rust/src/libproc_macro/lib.rs

1473 lines
53 KiB
Rust
Raw Normal View History

2016-08-04 14:20:01 -05:00
// Copyright 2016 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
// http://rust-lang.org/COPYRIGHT.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! A support library for macro authors when defining new macros.
2016-08-04 14:20:01 -05:00
//!
//! This library, provided by the standard distribution, provides the types
//! consumed in the interfaces of procedurally defined macro definitions.
//! Currently the primary use of this crate is to provide the ability to define
//! new custom derive modes through `#[proc_macro_derive]`.
2016-08-04 14:20:01 -05:00
//!
//! Note that this crate is intentionally very bare-bones currently. The main
//! type, `TokenStream`, only supports `fmt::Display` and `FromStr`
//! implementations, indicating that it can only go to and come from a string.
//! This functionality is intended to be expanded over time as more surface
//! area for macro authors is stabilized.
//!
//! See [the book](../book/first-edition/procedural-macros.html) for more.
2016-08-04 14:20:01 -05:00
#![stable(feature = "proc_macro_lib", since = "1.15.0")]
#![deny(missing_docs)]
#![doc(html_logo_url = "https://www.rust-lang.org/logos/rust-logo-128x128-blk-v2.png",
html_favicon_url = "https://doc.rust-lang.org/favicon.ico",
html_root_url = "https://doc.rust-lang.org/nightly/",
html_playground_url = "https://play.rust-lang.org/",
issue_tracker_base_url = "https://github.com/rust-lang/rust/issues/",
test(no_crate_inject, attr(deny(warnings))),
test(attr(allow(dead_code, deprecated, unused_variables, unused_mut))))]
2016-08-04 14:20:01 -05:00
#![feature(rustc_private)]
#![feature(staged_api)]
#![feature(lang_items)]
2018-03-01 19:42:22 -06:00
#![feature(optin_builtin_traits)]
2016-08-04 14:20:01 -05:00
extern crate syntax;
extern crate syntax_pos;
extern crate rustc_errors;
2018-02-27 10:11:14 -06:00
extern crate rustc_data_structures;
mod diagnostic;
#[unstable(feature = "proc_macro", issue = "38356")]
pub use diagnostic::{Diagnostic, Level};
2016-08-04 14:20:01 -05:00
2017-06-04 20:41:33 -05:00
use std::{ascii, fmt, iter};
2018-02-27 10:11:14 -06:00
use rustc_data_structures::sync::Lrc;
use std::str::FromStr;
use syntax::ast;
use syntax::errors::DiagnosticBuilder;
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 10:28:34 -05:00
use syntax::parse::{self, token};
2017-06-04 20:41:33 -05:00
use syntax::symbol::Symbol;
use syntax::tokenstream;
use syntax::parse::lexer::comments;
use syntax_pos::{FileMap, Pos, SyntaxContext, FileName};
use syntax_pos::hygiene::Mark;
/// The main type provided by this crate, representing an abstract stream of
/// tokens, or, more specifically, a sequence of token trees.
/// The type provide interfaces for iterating over those token trees and, conversely,
/// collecting a number of token trees into one stream.
///
/// This is both the input and output of `#[proc_macro]`, `#[proc_macro_attribute]`
/// and `#[proc_macro_derive]` definitions.
///
/// The API of this type is intentionally bare-bones, but it'll be expanded over
/// time!
#[stable(feature = "proc_macro_lib", since = "1.15.0")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[derive(Clone)]
pub struct TokenStream(tokenstream::TokenStream);
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for TokenStream {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for TokenStream {}
/// Error returned from `TokenStream::from_str`.
#[stable(feature = "proc_macro_lib", since = "1.15.0")]
2017-06-04 20:41:33 -05:00
#[derive(Debug)]
pub struct LexError {
_inner: (),
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for LexError {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for LexError {}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl TokenStream {
/// Returns an empty `TokenStream` containing no token trees.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn empty() -> TokenStream {
TokenStream(tokenstream::TokenStream::empty())
}
/// Checks if this `TokenStream` is empty.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn is_empty(&self) -> bool {
self.0.is_empty()
}
}
/// Attempts to break the string into tokens and parse those tokens into a token stream.
/// May fail for a number of reasons, for example, if the string contains unbalanced delimiters
/// or characters not existing in the language.
///
/// REVIEW The function actually panics on any error and never returns `LexError`.
/// REVIEW Should the panics be documented?
#[stable(feature = "proc_macro_lib", since = "1.15.0")]
impl FromStr for TokenStream {
type Err = LexError;
fn from_str(src: &str) -> Result<TokenStream, LexError> {
__internal::with_sess(|(sess, mark)| {
let src = src.to_string();
let name = FileName::ProcMacroSourceCode;
let expn_info = mark.expn_info().unwrap();
let call_site = expn_info.call_site;
// notify the expansion info that it is unhygienic
let mark = Mark::fresh(mark);
mark.set_expn_info(expn_info);
let span = call_site.with_ctxt(SyntaxContext::empty().apply_mark(mark));
let stream = parse::parse_stream_from_source_str(name, src, sess, Some(span));
Ok(__internal::token_stream_wrap(stream))
})
}
}
/// Prints the token stream as a string that should be losslessly convertible back
/// into the same token stream (modulo spans), except for possibly `TokenTree::Group`s
/// with `Delimiter::None` delimiters.
#[stable(feature = "proc_macro_lib", since = "1.15.0")]
impl fmt::Display for TokenStream {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.0.fmt(f)
}
}
/// Prints token in a form convenient for debugging.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[stable(feature = "proc_macro_lib", since = "1.15.0")]
impl fmt::Debug for TokenStream {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.write_str("TokenStream ")?;
f.debug_list().entries(self.clone()).finish()
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
}
2017-06-04 20:41:33 -05:00
/// Creates a token stream containing a single token tree.
///
/// REVIEW We don't generally have impls `From<T> for Collection<T>`, but I see why this exists
/// REVIEW from practical point of view.
#[unstable(feature = "proc_macro", issue = "38356")]
impl From<TokenTree> for TokenStream {
fn from(tree: TokenTree) -> TokenStream {
2017-06-04 20:41:33 -05:00
TokenStream(tree.to_internal())
}
}
/// Collects a number of token trees into a single stream.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl iter::FromIterator<TokenTree> for TokenStream {
fn from_iter<I: IntoIterator<Item = TokenTree>>(trees: I) -> Self {
trees.into_iter().map(TokenStream::from).collect()
}
}
/// A "flattening" operation on token streams, collects token trees
/// from multiple token streams into a single stream.
#[unstable(feature = "proc_macro", issue = "38356")]
impl iter::FromIterator<TokenStream> for TokenStream {
fn from_iter<I: IntoIterator<Item = TokenStream>>(streams: I) -> Self {
2017-06-04 20:41:33 -05:00
let mut builder = tokenstream::TokenStreamBuilder::new();
for stream in streams {
builder.push(stream.0);
}
TokenStream(builder.build())
}
}
/// Public implementation details for the `TokenStream` type, such as iterators.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub mod token_stream {
use syntax::tokenstream;
use syntax_pos::DUMMY_SP;
use {TokenTree, TokenStream, Delimiter};
/// An iterator over `TokenStream`'s `TokenTree`s.
/// The iteration is "shallow", e.g. the iterator doesn't recurse into delimited groups,
/// and returns whole groups as token trees.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[derive(Clone)]
#[unstable(feature = "proc_macro", issue = "38356")]
pub struct IntoIter {
cursor: tokenstream::Cursor,
stack: Vec<TokenTree>,
}
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl Iterator for IntoIter {
type Item = TokenTree;
fn next(&mut self) -> Option<TokenTree> {
loop {
let tree = self.stack.pop().or_else(|| {
let next = self.cursor.next_as_stream()?;
Some(TokenTree::from_internal(next, &mut self.stack))
})?;
// HACK: The condition "dummy span + group with empty delimiter" represents an AST
// fragment approximately converted into a token stream. This may happen, for
// example, with inputs to proc macro attributes, including derives. Such "groups"
// need to flattened during iteration over stream's token trees.
// Eventually this needs to be removed in favor of keeping original token trees
// and not doing the roundtrip through AST.
//
// REVIEW This may actually be observable if we can create a dummy span via
// proc macro API, but it looks like we can't do it with 1.2 yet.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
if tree.span().0 == DUMMY_SP {
if let TokenTree::Group(ref group) = tree {
if group.delimiter() == Delimiter::None {
self.cursor.insert(group.stream.clone().0);
continue
}
}
}
return Some(tree);
}
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl IntoIterator for TokenStream {
type Item = TokenTree;
type IntoIter = IntoIter;
fn into_iter(self) -> IntoIter {
IntoIter { cursor: self.0.trees(), stack: Vec::new() }
}
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// `quote!(..)` accepts arbitrary tokens and expands into a `TokenStream` describing the input.
/// For example, `quote!(a + b)` will produce a expression, that, when evaluated, constructs
2018-05-05 13:09:41 -05:00
/// the `TokenStream` `[Word("a"), Punct('+', Alone), Word("b")]`.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
/// Unquoting is done with `$`, and works by taking the single next ident as the unquoted term.
/// To quote `$` itself, use `$$`.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[macro_export]
macro_rules! quote { () => {} }
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro_internals", issue = "27812")]
#[doc(hidden)]
mod quote;
2017-06-04 20:41:33 -05:00
/// Quote a `Span` into a `TokenStream`.
/// This is needed to implement a custom quoter.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn quote_span(span: Span) -> TokenStream {
quote::Quote::quote(span)
2017-06-04 20:41:33 -05:00
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// A region of source code, along with macro expansion information.
///
/// REVIEW ATTENTION: `Copy` impl on a struct with private fields.
/// REVIEW Do we want to guarantee `Span` to be `Copy`? Yes.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
#[derive(Copy, Clone)]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub struct Span(syntax_pos::Span);
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for Span {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for Span {}
macro_rules! diagnostic_method {
($name:ident, $level:expr) => (
/// Create a new `Diagnostic` with the given `message` at the span
/// `self`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn $name<T: Into<String>>(self, message: T) -> Diagnostic {
Diagnostic::spanned(self, $level, message)
}
)
}
impl Span {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// A span that resolves at the macro definition site.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn def_site() -> Span {
::__internal::with_sess(|(_, mark)| {
let call_site = mark.expn_info().unwrap().call_site;
Span(call_site.with_ctxt(SyntaxContext::empty().apply_mark(mark)))
})
}
/// The span of the invocation of the current procedural macro.
/// Identifiers created with this span will be resolved as if they were written
/// directly at the macro call location (call-site hygiene) and other code
/// at the macro call site will be able to refer to them as well.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn call_site() -> Span {
::__internal::with_sess(|(_, mark)| Span(mark.expn_info().unwrap().call_site))
}
2017-08-01 20:05:08 -05:00
/// The original source file into which this span points.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn source_file(&self) -> SourceFile {
SourceFile {
filemap: __internal::lookup_char_pos(self.0.lo()).file,
}
}
/// The `Span` for the tokens in the previous macro expansion from which
/// `self` was generated from, if any.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn parent(&self) -> Option<Span> {
2018-04-26 17:28:34 -05:00
self.0.parent().map(Span)
}
/// The span for the origin source code that `self` was generated from. If
/// this `Span` wasn't generated from other macro expansions then the return
/// value is the same as `*self`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn source(&self) -> Span {
Span(self.0.source_callsite())
}
2017-08-01 20:05:08 -05:00
/// Get the starting line/column in the source file for this span.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn start(&self) -> LineColumn {
let loc = __internal::lookup_char_pos(self.0.lo());
LineColumn {
line: loc.line,
column: loc.col.to_usize()
}
}
/// Get the ending line/column in the source file for this span.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn end(&self) -> LineColumn {
let loc = __internal::lookup_char_pos(self.0.hi());
LineColumn {
line: loc.line,
column: loc.col.to_usize()
}
}
/// Create a new span encompassing `self` and `other`.
///
/// Returns `None` if `self` and `other` are from different files.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn join(&self, other: Span) -> Option<Span> {
let self_loc = __internal::lookup_char_pos(self.0.lo());
let other_loc = __internal::lookup_char_pos(other.0.lo());
2017-08-01 20:05:08 -05:00
if self_loc.file.name != other_loc.file.name { return None }
Some(Span(self.0.to(other.0)))
}
/// Creates a new span with the same line/column information as `self` but
/// that resolves symbols as though it were at `other`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn resolved_at(&self, other: Span) -> Span {
Span(self.0.with_ctxt(other.0.ctxt()))
}
/// Creates a new span with the same name resolution behavior as `self` but
/// with the line/column information of `other`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn located_at(&self, other: Span) -> Span {
other.resolved_at(*self)
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Compares to spans to see if they're equal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn eq(&self, other: &Span) -> bool {
self.0 == other.0
}
diagnostic_method!(error, Level::Error);
diagnostic_method!(warning, Level::Warning);
diagnostic_method!(note, Level::Note);
diagnostic_method!(help, Level::Help);
}
/// Prints a span in a form convenient for debugging.
#[unstable(feature = "proc_macro", issue = "38356")]
impl fmt::Debug for Span {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "{:?} bytes({}..{})",
self.0.ctxt(),
self.0.lo().0,
self.0.hi().0)
}
}
2017-08-01 20:05:08 -05:00
/// A line-column pair representing the start or end of a `Span`.
#[unstable(feature = "proc_macro", issue = "38356")]
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
pub struct LineColumn {
/// The 1-indexed line in the source file on which the span starts or ends (inclusive).
#[unstable(feature = "proc_macro", issue = "38356")]
pub line: usize,
2017-08-01 20:05:08 -05:00
/// The 0-indexed column (in UTF-8 characters) in the source file on which
/// the span starts or ends (inclusive).
#[unstable(feature = "proc_macro", issue = "38356")]
pub column: usize
2017-08-01 20:05:08 -05:00
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for LineColumn {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for LineColumn {}
2017-08-01 20:05:08 -05:00
/// The source file of a given `Span`.
#[unstable(feature = "proc_macro", issue = "38356")]
#[derive(Clone)]
pub struct SourceFile {
2018-02-27 10:11:14 -06:00
filemap: Lrc<FileMap>,
2017-08-01 20:05:08 -05:00
}
2018-03-01 19:42:22 -06:00
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for SourceFile {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for SourceFile {}
2017-08-01 20:05:08 -05:00
impl SourceFile {
/// Get the path to this source file.
2017-08-01 20:05:08 -05:00
///
/// ### Note
/// If the code span associated with this `SourceFile` was generated by an external macro, this
/// may not be an actual path on the filesystem. Use [`is_real`] to check.
///
/// Also note that even if `is_real` returns `true`, if `--remap-path-prefix` was passed on
2017-08-01 20:05:08 -05:00
/// the command line, the path as given may not actually be valid.
///
/// [`is_real`]: #method.is_real
# [unstable(feature = "proc_macro", issue = "38356")]
pub fn path(&self) -> &FileName {
2017-08-01 20:05:08 -05:00
&self.filemap.name
}
/// Returns `true` if this source file is a real source file, and not generated by an external
/// macro's expansion.
#[unstable(feature = "proc_macro", issue = "38356")]
2017-08-01 20:05:08 -05:00
pub fn is_real(&self) -> bool {
// This is a hack until intercrate spans are implemented and we can have real source files
// for spans generated in external macros.
// https://github.com/rust-lang/rust/pull/43604#issuecomment-333334368
self.filemap.is_real_file()
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl AsRef<FileName> for SourceFile {
fn as_ref(&self) -> &FileName {
self.path()
2017-08-01 20:05:08 -05:00
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl fmt::Debug for SourceFile {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("SourceFile")
.field("path", self.path())
2017-08-01 20:05:08 -05:00
.field("is_real", &self.is_real())
.finish()
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl PartialEq for SourceFile {
fn eq(&self, other: &Self) -> bool {
2018-02-27 10:11:14 -06:00
Lrc::ptr_eq(&self.filemap, &other.filemap)
2017-08-01 20:05:08 -05:00
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl Eq for SourceFile {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl PartialEq<FileName> for SourceFile {
fn eq(&self, other: &FileName) -> bool {
2017-08-01 20:05:08 -05:00
self.as_ref() == other
}
}
/// A single token or a delimited sequence of token trees (e.g. `[1, (), ..]`).
#[unstable(feature = "proc_macro", issue = "38356")]
#[derive(Clone)]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub enum TokenTree {
/// A token stream surrounded by bracket delimiters.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Group(Group),
/// An identifier or lifetime identifier.
2018-05-05 13:09:41 -05:00
Ident(Ident),
/// A single punctuation character (`+`, `,`, `$`, etc.).
2018-05-05 13:09:41 -05:00
Punct(Punct),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// A literal character (`'a'`), string (`"hello"`), number (`2.3`), etc.
Literal(Literal),
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for TokenTree {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for TokenTree {}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl TokenTree {
/// Returns the span of this tree, delegating to the `span` method of
/// the contained token or a delimited stream.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn span(&self) -> Span {
match *self {
TokenTree::Group(ref t) => t.span(),
2018-05-05 13:09:41 -05:00
TokenTree::Ident(ref t) => t.span(),
TokenTree::Punct(ref t) => t.span(),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
TokenTree::Literal(ref t) => t.span(),
}
}
/// Configures the span for *only this token*.
///
/// Note that if this token is a `Group` then this method will not configure
/// the span of each of the internal tokens, this will simply delegate to
/// the `set_span` method of each variant.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn set_span(&mut self, span: Span) {
match *self {
TokenTree::Group(ref mut t) => t.set_span(span),
2018-05-05 13:09:41 -05:00
TokenTree::Ident(ref mut t) => t.set_span(span),
TokenTree::Punct(ref mut t) => t.set_span(span),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
TokenTree::Literal(ref mut t) => t.set_span(span),
}
}
}
/// Prints token treee in a form convenient for debugging.
#[unstable(feature = "proc_macro", issue = "38356")]
impl fmt::Debug for TokenTree {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
// Each of these has the name in the struct type in the derived debug,
// so don't bother with an extra layer of indirection
match *self {
TokenTree::Group(ref tt) => tt.fmt(f),
2018-05-05 13:09:41 -05:00
TokenTree::Ident(ref tt) => tt.fmt(f),
TokenTree::Punct(ref tt) => tt.fmt(f),
TokenTree::Literal(ref tt) => tt.fmt(f),
}
}
}
/// REVIEW the impls below are kind of `From<T> for Option<T>`, not strictly necessary,
/// REVIEW but convenient. No harm, I guess. I'd actually like to see impls
2018-05-05 13:09:41 -05:00
/// REVIEW `From<Group/Ident/Punct/Literal> for TokenStream` to avoid stuttering like
/// REVIEW `TokenTree::Literal(Literal::string("lalala")).into()`.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
impl From<Group> for TokenTree {
fn from(g: Group) -> TokenTree {
TokenTree::Group(g)
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl From<Ident> for TokenTree {
fn from(g: Ident) -> TokenTree {
TokenTree::Ident(g)
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl From<Punct> for TokenTree {
fn from(g: Punct) -> TokenTree {
TokenTree::Punct(g)
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
}
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl From<Literal> for TokenTree {
fn from(g: Literal) -> TokenTree {
TokenTree::Literal(g)
}
}
/// Prints the token tree as a string that should be losslessly convertible back
/// into the same token tree (modulo spans), except for possibly `TokenTree::Group`s
/// with `Delimiter::None` delimiters.
#[unstable(feature = "proc_macro", issue = "38356")]
impl fmt::Display for TokenTree {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
match *self {
TokenTree::Group(ref t) => t.fmt(f),
2018-05-05 13:09:41 -05:00
TokenTree::Ident(ref t) => t.fmt(f),
TokenTree::Punct(ref t) => t.fmt(f),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
TokenTree::Literal(ref t) => t.fmt(f),
}
}
}
/// A delimited token stream.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
/// A `Group` internally contains a `TokenStream` which is surrounded by `Delimiter`s.
2017-06-04 20:41:33 -05:00
#[derive(Clone, Debug)]
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub struct Group {
delimiter: Delimiter,
stream: TokenStream,
span: Span,
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for Group {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for Group {}
/// Describes how a sequence of token trees is delimited.
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
#[unstable(feature = "proc_macro", issue = "38356")]
pub enum Delimiter {
/// `( ... )`
Parenthesis,
/// `{ ... }`
Brace,
/// `[ ... ]`
Bracket,
/// `Ø ... Ø`
/// An implicit delimiter, that may, for example, appear around tokens coming from a
/// "macro variable" `$var`. It is important to preserve operator priorities in cases like
/// `$var * 3` where `$var` is `1 + 2`.
/// Implicit delimiters may not survive roundtrip of a token stream through a string.
None,
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl Group {
/// Creates a new `Group` with the given delimiter and token stream.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
/// This constructor will set the span for this group to
/// `Span::call_site()`. To change the span you can use the `set_span`
/// method below.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn new(delimiter: Delimiter, stream: TokenStream) -> Group {
Group {
delimiter: delimiter,
stream: stream,
span: Span::call_site(),
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Returns the delimiter of this `Group`
2017-06-04 20:41:33 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn delimiter(&self) -> Delimiter {
self.delimiter
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Returns the `TokenStream` of tokens that are delimited in this `Group`.
///
/// Note that the returned token stream does not include the delimiter
/// returned above.
2017-06-04 20:41:33 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn stream(&self) -> TokenStream {
self.stream.clone()
}
/// Returns the span for the delimiters of this token stream, spanning the
/// entire `Group`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn span(&self) -> Span {
self.span
}
/// Configures the span for this `Group`'s delimiters, but not its internal
/// tokens.
///
/// This method will **not** set the span of all the internal tokens spanned
/// by this group, but rather it will only set the span of the delimiter
/// tokens at the level of the `Group`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn set_span(&mut self, span: Span) {
self.span = span;
}
}
/// Prints the group as a string that should be losslessly convertible back
/// into the same group (modulo spans), except for possibly `TokenTree::Group`s
/// with `Delimiter::None` delimiters.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
impl fmt::Display for Group {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
TokenStream::from(TokenTree::from(self.clone())).fmt(f)
}
}
2018-05-05 13:09:41 -05:00
/// An `Punct` is an single punctuation character like `+`, `-` or `#`.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
2018-05-05 13:09:41 -05:00
/// Multicharacter operators like `+=` are represented as two instances of `Punct` with different
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// forms of `Spacing` returned.
///
2018-05-05 13:09:41 -05:00
/// REVIEW We should guarantee that `Punct` contains a valid punctuation character permitted by
/// REVIEW the language and not a random unicode code point. The check is already performed in
/// REVIEW `TokenTree::to_internal`, but we should do it on construction.
2018-05-05 13:09:41 -05:00
/// REVIEW `Punct` can also avoid using `char` internally and keep an u8-like enum.
///
/// REVIEW ATTENTION: `Copy` impl on a struct with private fields.
2018-05-05 13:09:41 -05:00
/// REVIEW Do we want to guarantee `Punct` to be `Copy`?
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
#[derive(Copy, Clone, Debug)]
2018-05-05 13:09:41 -05:00
pub struct Punct {
ch: char,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
spacing: Spacing,
span: Span,
}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl !Send for Punct {}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl !Sync for Punct {}
2018-05-05 13:09:41 -05:00
/// Whether an `Punct` is followed immediately by another `Punct` or
/// followed by another token or whitespace.
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
#[unstable(feature = "proc_macro", issue = "38356")]
2017-06-04 20:41:33 -05:00
pub enum Spacing {
/// e.g. `+` is `Alone` in `+ =`, `+ident` or `+()`.
Alone,
/// e.g. `+` is `Joint` in `+=` or `+#`.
Joint,
}
2018-05-05 13:09:41 -05:00
impl Punct {
/// Creates a new `Punct` from the given character and spacing.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
2018-05-05 13:09:41 -05:00
/// The returned `Punct` will have the default span of `Span::call_site()`
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// which can be further configured with the `set_span` method below.
///
/// REVIEW Why we even use `char` here? There's no reason to use unicode here.
/// REVIEW I guess because it's more convenient to write `new('+')` than `new(b'+')`, that's ok.
///
/// REVIEW TO_DO Do input validation on construction, the argument should be a valid punctuation
/// REVIEW character permitted by the language.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
pub fn new(ch: char, spacing: Spacing) -> Punct {
Punct {
ch: ch,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
spacing: spacing,
span: Span::call_site(),
}
}
/// Returns the value of this punctuation character as `char`.
///
/// REVIEW Again, there's no need for unicode here,
/// REVIEW except for maybe future compatibility in case Rust turns into APL,
/// REVIEW but if it's more convenient to use `char` then that's okay.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
pub fn as_char(&self) -> char {
self.ch
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
/// Returns the spacing of this punctuation character, indicating whether it's immediately
2018-05-05 13:09:41 -05:00
/// followed by another `Punct` in the token stream, so they can potentially be combined into
/// a multicharacter operator (`Joint`), or it's followed by some other token or whitespace
/// (`Alone`) so the operator has certainly ended.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn spacing(&self) -> Spacing {
self.spacing
}
/// Returns the span for this punctuation character.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn span(&self) -> Span {
self.span
}
/// Configure the span for this punctuation character.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn set_span(&mut self, span: Span) {
self.span = span;
}
}
/// Prints the punctuation character as a string that should be losslessly convertible
/// back into the same character.
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl fmt::Display for Punct {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
TokenStream::from(TokenTree::from(self.clone())).fmt(f)
}
}
/// An identifier (`ident`) or lifetime identifier (`'ident`).
///
2018-05-05 13:09:41 -05:00
/// REVIEW We should guarantee that `Ident` contains a valid identifier permitted by
/// REVIEW the language and not a random unicode string, at least for a start.
///
/// REVIEW We need to support raw identifiers here (`r#ident`) or at least be future compatible
/// REVIEW with them. Currently they are supported using "string typing" - if string "r#ident" is
2018-05-05 13:09:41 -05:00
/// REVIEW passed to `Ident::new` it will be interpreted as a raw identifier later on, we should add
/// REVIEW a field `is_raw` and a separate constructor for it (`Ident::new_raw` or something) and
/// REVIEW keep it unstable until raw identifiers are stabilized.
///
/// REVIEW ATTENTION: `Copy` impl on a struct with private fields.
2018-05-05 13:09:41 -05:00
/// REVIEW Do we want to guarantee `Ident` to be `Copy`?
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[derive(Copy, Clone, Debug)]
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
pub struct Ident {
// REVIEW(INTERNAL) Symbol + Span is actually `ast::Ident`! We can use it here.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
sym: Symbol,
span: Span,
}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl !Send for Ident {}
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl !Sync for Ident {}
2018-05-05 13:09:41 -05:00
impl Ident {
/// Creates a new `Ident` with the given `string` as well as the specified
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// `span`.
///
/// Note that `span`, currently in rustc, configures the hygiene information
/// for this identifier.
///
/// As of this time `Span::call_site()` explicitly opts-in to "call-site" hygiene
/// meaning that identifiers created with this span will be resolved as if they were written
/// directly at the location of the macro call, and other code at the macro call site will be
/// able to refer to them as well.
///
/// Later spans like `Span::def_site()` will allow to opt-in to "definition-site" hygiene
/// meaning that identifiers created with this span will be resolved at the location of the
/// macro definition and other code at the macro call site will not be able to refer to them.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
///
/// Due to the current importance of hygiene this constructor, unlike other
/// tokens, requires a `Span` to be specified at construction.
///
/// REVIEW TO_DO Do input validation, the argument should be a valid identifier or
/// REVIEW lifetime identifier.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
pub fn new(string: &str, span: Span) -> Ident {
Ident {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
sym: Symbol::intern(string),
span,
}
}
// FIXME: Remove this, do not stabilize
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Get a reference to the interned string.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn as_str(&self) -> &str {
unsafe { &*(&*self.sym.as_str() as *const str) }
}
2018-05-05 13:09:41 -05:00
/// Returns the span of this `Ident`, encompassing the entire string returned
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// by `as_str`.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn span(&self) -> Span {
self.span
}
2018-05-05 13:09:41 -05:00
/// Configures the span of this `Ident`, possibly changing its hygiene context.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn set_span(&mut self, span: Span) {
self.span = span;
}
}
/// Prints the identifier as a string that should be losslessly convertible
/// back into the same identifier.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
2018-05-05 13:09:41 -05:00
impl fmt::Display for Ident {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.sym.as_str().fmt(f)
}
}
/// A literal string (`"hello"`), byte string (`b"hello"`),
/// character (`'a'`), byte character (`b'a'`), an integer or floating point number
/// with or without a suffix (`1`, `1u8`, `2.3`, `2.3f32`).
2018-05-05 13:09:41 -05:00
/// Boolean literals like `true` and `false` do not belong here, they are `Ident`s.
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
#[derive(Clone, Debug)]
#[unstable(feature = "proc_macro", issue = "38356")]
pub struct Literal {
lit: token::Lit,
suffix: Option<ast::Name>,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span,
}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Send for Literal {}
#[unstable(feature = "proc_macro", issue = "38356")]
impl !Sync for Literal {}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
macro_rules! suffixed_int_literals {
($($name:ident => $kind:ident,)*) => ($(
/// Creates a new suffixed integer literal with the specified value.
///
/// This function will create an integer like `1u32` where the integer
/// value specified is the first part of the token and the integral is
/// also suffixed at the end.
///
/// Literals created through this method have the `Span::call_site()`
/// span by default, which can be configured with the `set_span` method
/// below.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn $name(n: $kind) -> Literal {
Literal {
lit: token::Lit::Integer(Symbol::intern(&n.to_string())),
suffix: Some(Symbol::intern(stringify!($kind))),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
)*)
}
macro_rules! unsuffixed_int_literals {
($($name:ident => $kind:ident,)*) => ($(
/// Creates a new unsuffixed integer literal with the specified value.
///
/// This function will create an integer like `1` where the integer
/// value specified is the first part of the token. No suffix is
/// specified on this token, meaning that invocations like
/// `Literal::i8_unsuffixed(1)` are equivalent to
/// `Literal::u32_unsuffixed(1)`.
///
/// Literals created through this method have the `Span::call_site()`
/// span by default, which can be configured with the `set_span` method
/// below.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn $name(n: $kind) -> Literal {
Literal {
lit: token::Lit::Integer(Symbol::intern(&n.to_string())),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
)*)
}
impl Literal {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
suffixed_int_literals! {
u8_suffixed => u8,
u16_suffixed => u16,
u32_suffixed => u32,
u64_suffixed => u64,
u128_suffixed => u128,
usize_suffixed => usize,
i8_suffixed => i8,
i16_suffixed => i16,
i32_suffixed => i32,
i64_suffixed => i64,
i128_suffixed => i128,
isize_suffixed => isize,
2017-06-04 20:41:33 -05:00
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
unsuffixed_int_literals! {
u8_unsuffixed => u8,
u16_unsuffixed => u16,
u32_unsuffixed => u32,
u64_unsuffixed => u64,
u128_unsuffixed => u128,
usize_unsuffixed => usize,
i8_unsuffixed => i8,
i16_unsuffixed => i16,
i32_unsuffixed => i32,
i64_unsuffixed => i64,
i128_unsuffixed => i128,
isize_unsuffixed => isize,
2017-06-04 20:41:33 -05:00
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Creates a new unsuffixed floating-point literal.
///
/// This constructor is similar to those like `Literal::i8_unsuffixed` where
/// the float's value is emitted directly into the token but no suffix is
/// used, so it may be inferred to be a `f64` later in the compiler.
///
/// # Panics
///
/// This function requires that the specified float is finite, for
/// example if it is infinity or NaN this function will panic.
2017-06-04 20:41:33 -05:00
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn f32_unsuffixed(n: f32) -> Literal {
2017-06-04 20:41:33 -05:00
if !n.is_finite() {
panic!("Invalid float literal {}", n);
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Literal {
lit: token::Lit::Float(Symbol::intern(&n.to_string())),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Creates a new suffixed floating-point literal.
///
/// This consturctor will create a literal like `1.0f32` where the value
/// specified is the preceding part of the token and `f32` is the suffix of
/// the token. This token will always be inferred to be an `f32` in the
/// compiler.
///
/// # Panics
///
/// This function requires that the specified float is finite, for
/// example if it is infinity or NaN this function will panic.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn f32_suffixed(n: f32) -> Literal {
2017-06-04 20:41:33 -05:00
if !n.is_finite() {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
panic!("Invalid float literal {}", n);
}
Literal {
lit: token::Lit::Float(Symbol::intern(&n.to_string())),
suffix: Some(Symbol::intern("f32")),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
2017-06-04 20:41:33 -05:00
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Creates a new unsuffixed floating-point literal.
///
/// This constructor is similar to those like `Literal::i8_unsuffixed` where
/// the float's value is emitted directly into the token but no suffix is
/// used, so it may be inferred to be a `f64` later in the compiler.
///
/// # Panics
///
/// This function requires that the specified float is finite, for
/// example if it is infinity or NaN this function will panic.
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
pub fn f64_unsuffixed(n: f64) -> Literal {
2017-06-04 20:41:33 -05:00
if !n.is_finite() {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
panic!("Invalid float literal {}", n);
}
Literal {
lit: token::Lit::Float(Symbol::intern(&n.to_string())),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
/// Creates a new suffixed floating-point literal.
///
/// This consturctor will create a literal like `1.0f64` where the value
/// specified is the preceding part of the token and `f64` is the suffix of
/// the token. This token will always be inferred to be an `f64` in the
/// compiler.
///
/// # Panics
///
/// This function requires that the specified float is finite, for
/// example if it is infinity or NaN this function will panic.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn f64_suffixed(n: f64) -> Literal {
if !n.is_finite() {
panic!("Invalid float literal {}", n);
}
Literal {
lit: token::Lit::Float(Symbol::intern(&n.to_string())),
suffix: Some(Symbol::intern("f64")),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
2017-06-04 20:41:33 -05:00
}
}
/// String literal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn string(string: &str) -> Literal {
let mut escaped = String::new();
for ch in string.chars() {
escaped.extend(ch.escape_debug());
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Literal {
lit: token::Lit::Str_(Symbol::intern(&escaped)),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
/// Character literal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn character(ch: char) -> Literal {
let mut escaped = String::new();
escaped.extend(ch.escape_unicode());
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Literal {
lit: token::Lit::Char(Symbol::intern(&escaped)),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
2017-06-04 20:41:33 -05:00
}
/// Byte string literal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn byte_string(bytes: &[u8]) -> Literal {
let string = bytes.iter().cloned().flat_map(ascii::escape_default)
.map(Into::<char>::into).collect::<String>();
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Literal {
lit: token::Lit::ByteStr(Symbol::intern(&string)),
suffix: None,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span: Span::call_site(),
}
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
/// Returns the span encompassing this literal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn span(&self) -> Span {
self.span
}
/// Configures the span associated for this literal.
#[unstable(feature = "proc_macro", issue = "38356")]
pub fn set_span(&mut self, span: Span) {
self.span = span;
}
}
/// Prints the literal as a string that should be losslessly convertible
/// back into the same literal (except for possible rounding for floating point literals).
#[unstable(feature = "proc_macro", issue = "38356")]
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
impl fmt::Display for Literal {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
TokenStream::from(TokenTree::from(self.clone())).fmt(f)
}
}
impl Delimiter {
2017-06-04 20:41:33 -05:00
fn from_internal(delim: token::DelimToken) -> Delimiter {
match delim {
token::Paren => Delimiter::Parenthesis,
token::Brace => Delimiter::Brace,
token::Bracket => Delimiter::Bracket,
token::NoDelim => Delimiter::None,
}
}
2017-06-04 20:41:33 -05:00
fn to_internal(self) -> token::DelimToken {
match self {
Delimiter::Parenthesis => token::Paren,
Delimiter::Brace => token::Brace,
Delimiter::Bracket => token::Bracket,
Delimiter::None => token::NoDelim,
}
}
}
impl TokenTree {
fn from_internal(stream: tokenstream::TokenStream, stack: &mut Vec<TokenTree>)
-> TokenTree {
use syntax::parse::token::*;
let (tree, is_joint) = stream.as_tree();
let (span, token) = match tree {
tokenstream::TokenTree::Token(span, token) => (span, token),
tokenstream::TokenTree::Delimited(span, delimed) => {
2017-06-04 20:41:33 -05:00
let delimiter = Delimiter::from_internal(delimed.delim);
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
let mut g = Group::new(delimiter, TokenStream(delimed.tts.into()));
g.set_span(Span(span));
return g.into()
}
};
2017-06-04 20:41:33 -05:00
let op_kind = if is_joint { Spacing::Joint } else { Spacing::Alone };
macro_rules! tt {
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
($e:expr) => ({
let mut x = TokenTree::from($e);
x.set_span(Span(span));
x
})
}
macro_rules! op {
2018-05-05 13:09:41 -05:00
($a:expr) => (tt!(Punct::new($a, op_kind)));
($a:expr, $b:expr) => ({
2018-05-05 13:09:41 -05:00
stack.push(tt!(Punct::new($b, op_kind)));
tt!(Punct::new($a, Spacing::Joint))
});
($a:expr, $b:expr, $c:expr) => ({
2018-05-05 13:09:41 -05:00
stack.push(tt!(Punct::new($c, op_kind)));
stack.push(tt!(Punct::new($b, Spacing::Joint)));
tt!(Punct::new($a, Spacing::Joint))
})
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
match token {
Eq => op!('='),
Lt => op!('<'),
Le => op!('<', '='),
EqEq => op!('=', '='),
Ne => op!('!', '='),
Ge => op!('>', '='),
Gt => op!('>'),
AndAnd => op!('&', '&'),
OrOr => op!('|', '|'),
Not => op!('!'),
Tilde => op!('~'),
BinOp(Plus) => op!('+'),
BinOp(Minus) => op!('-'),
BinOp(Star) => op!('*'),
BinOp(Slash) => op!('/'),
BinOp(Percent) => op!('%'),
BinOp(Caret) => op!('^'),
BinOp(And) => op!('&'),
BinOp(Or) => op!('|'),
BinOp(Shl) => op!('<', '<'),
BinOp(Shr) => op!('>', '>'),
BinOpEq(Plus) => op!('+', '='),
BinOpEq(Minus) => op!('-', '='),
BinOpEq(Star) => op!('*', '='),
BinOpEq(Slash) => op!('/', '='),
BinOpEq(Percent) => op!('%', '='),
BinOpEq(Caret) => op!('^', '='),
BinOpEq(And) => op!('&', '='),
BinOpEq(Or) => op!('|', '='),
BinOpEq(Shl) => op!('<', '<', '='),
BinOpEq(Shr) => op!('>', '>', '='),
At => op!('@'),
Dot => op!('.'),
DotDot => op!('.', '.'),
DotDotDot => op!('.', '.', '.'),
DotDotEq => op!('.', '.', '='),
Comma => op!(','),
Semi => op!(';'),
Colon => op!(':'),
ModSep => op!(':', ':'),
RArrow => op!('-', '>'),
LArrow => op!('<', '-'),
FatArrow => op!('=', '>'),
Pound => op!('#'),
Dollar => op!('$'),
Question => op!('?'),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
Ident(ident, false) | Lifetime(ident) => {
2018-05-05 13:09:41 -05:00
tt!(self::Ident::new(&ident.name.as_str(), Span(span)))
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
Ident(ident, true) => {
2018-05-05 13:09:41 -05:00
tt!(self::Ident::new(&format!("r#{}", ident), Span(span)))
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
Literal(lit, suffix) => tt!(self::Literal { lit, suffix, span: Span(span) }),
DocComment(c) => {
let style = comments::doc_comment_style(&c.as_str());
let stripped = comments::strip_doc_comment_decoration(&c.as_str());
let stream = vec![
2018-05-05 13:09:41 -05:00
tt!(self::Ident::new("doc", Span(span))),
tt!(Punct::new('=', Spacing::Alone)),
tt!(self::Literal::string(&stripped)),
].into_iter().collect();
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
stack.push(tt!(Group::new(Delimiter::Bracket, stream)));
if style == ast::AttrStyle::Inner {
2018-05-05 13:09:41 -05:00
stack.push(tt!(Punct::new('!', Spacing::Alone)));
}
2018-05-05 13:09:41 -05:00
tt!(Punct::new('#', Spacing::Alone))
}
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 10:28:34 -05:00
Interpolated(_) => {
__internal::with_sess(|(sess, _)| {
let tts = token.interpolated_to_tokenstream(sess, span);
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
tt!(Group::new(Delimiter::None, TokenStream(tts)))
})
}
DotEq => op!('.', '='),
OpenDelim(..) | CloseDelim(..) => unreachable!(),
Whitespace | Comment | Shebang(..) | Eof => unreachable!(),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
}
}
2017-06-04 20:41:33 -05:00
fn to_internal(self) -> tokenstream::TokenStream {
use syntax::parse::token::*;
use syntax::tokenstream::{TokenTree, Delimited};
2018-05-05 13:09:41 -05:00
let (ch, kind, span) = match self {
self::TokenTree::Punct(tt) => (tt.as_char(), tt.spacing(), tt.span()),
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
self::TokenTree::Group(tt) => {
return TokenTree::Delimited(tt.span.0, Delimited {
delim: tt.delimiter.to_internal(),
tts: tt.stream.0.into(),
}).into();
},
2018-05-05 13:09:41 -05:00
self::TokenTree::Ident(tt) => {
let ident = ast::Ident::new(tt.sym, tt.span.0);
let sym_str = tt.sym.to_string();
2018-03-18 19:54:56 -05:00
let token = if sym_str.starts_with("'") {
Lifetime(ident)
} else if sym_str.starts_with("r#") {
let name = Symbol::intern(&sym_str[2..]);
let ident = ast::Ident::new(name, ident.span);
Ident(ident, true)
} else {
Ident(ident, false)
};
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
return TokenTree::Token(tt.span.0, token).into();
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
self::TokenTree::Literal(self::Literal {
lit: Lit::Integer(ref a),
suffix,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span,
})
if a.as_str().starts_with("-") =>
{
let minus = BinOp(BinOpToken::Minus);
let integer = Symbol::intern(&a.as_str()[1..]);
let integer = Literal(Lit::Integer(integer), suffix);
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
let a = TokenTree::Token(span.0, minus);
let b = TokenTree::Token(span.0, integer);
return vec![a, b].into_iter().collect()
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
self::TokenTree::Literal(self::Literal {
lit: Lit::Float(ref a),
suffix,
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
span,
})
if a.as_str().starts_with("-") =>
{
let minus = BinOp(BinOpToken::Minus);
let float = Symbol::intern(&a.as_str()[1..]);
let float = Literal(Lit::Float(float), suffix);
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
let a = TokenTree::Token(span.0, minus);
let b = TokenTree::Token(span.0, float);
return vec![a, b].into_iter().collect()
}
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
self::TokenTree::Literal(tt) => {
let token = Literal(tt.lit, tt.suffix);
return TokenTree::Token(tt.span.0, token).into()
}
};
2018-05-05 13:09:41 -05:00
let token = match ch {
'=' => Eq,
'<' => Lt,
'>' => Gt,
'!' => Not,
'~' => Tilde,
'+' => BinOp(Plus),
'-' => BinOp(Minus),
'*' => BinOp(Star),
'/' => BinOp(Slash),
'%' => BinOp(Percent),
'^' => BinOp(Caret),
'&' => BinOp(And),
'|' => BinOp(Or),
'@' => At,
'.' => Dot,
',' => Comma,
';' => Semi,
':' => Colon,
'#' => Pound,
'$' => Dollar,
'?' => Question,
2018-05-05 13:09:41 -05:00
_ => panic!("unsupported character {}", ch),
};
proc_macro: Reorganize public API This commit is a reorganization of the `proc_macro` crate's public user-facing API. This is the result of a number of discussions at the recent Rust All-Hands where we're hoping to get the `proc_macro` crate into ship shape for stabilization of a subset of its functionality in the Rust 2018 release. The reorganization here is motivated by experiences from the `proc-macro2`, `quote`, and `syn` crates on crates.io (and other crates which depend on them). The main focus is future flexibility along with making a few more operations consistent and/or fixing bugs. A summary of the changes made from today's `proc_macro` API is: * The `TokenNode` enum has been removed and the public fields of `TokenTree` have also been removed. Instead the `TokenTree` type is now a public enum (what `TokenNode` was) and each variant is an opaque struct which internally contains `Span` information. This makes the various tokens a bit more consistent, require fewer wrappers, and otherwise provides good future-compatibility as opaque structs are easy to modify later on. * `Literal` integer constructors have been expanded to be unambiguous as to what they're doing and also allow for more future flexibility. Previously constructors like `Literal::float` and `Literal::integer` were used to create unsuffixed literals and the concrete methods like `Literal::i32` would create a suffixed token. This wasn't immediately clear to all users (the suffixed/unsuffixed aspect) and having *one* constructor for unsuffixed literals required us to pick a largest type which may not always be true. To fix these issues all constructors are now of the form `Literal::i32_unsuffixed` or `Literal::i32_suffixed` (for all integral types). This should allow future compatibility as well as being immediately clear what's suffixed and what isn't. * Each variant of `TokenTree` internally contains a `Span` which can also be configured via `set_span`. For example `Literal` and `Term` now both internally contain a `Span` rather than having it stored in an auxiliary location. * Constructors of all tokens are called `new` now (aka `Term::intern` is gone) and most do not take spans. Manufactured tokens typically don't have a fresh span to go with them and the span is purely used for error-reporting **except** the span for `Term`, which currently affects hygiene. The default spans for all these constructed tokens is `Span::call_site()` for now. The `Term` type's constructor explicitly requires passing in a `Span` to provide future-proofing against possible hygiene changes. It's intended that a first pass of stabilization will likely only stabilize `Span::call_site()` which is an explicit opt-in for "I would like no hygiene here please". The intention here is to make this explicit in procedural macros to be forwards-compatible with a hygiene-specifying solution. * Some of the conversions for `TokenStream` have been simplified a little. * The `TokenTreeIter` iterator was renamed to `token_stream::IntoIter`. Overall the hope is that this is the "final pass" at the API of `TokenStream` and most of `TokenTree` before stabilization. Explicitly left out here is any changes to `Span`'s API which will likely need to be re-evaluated before stabilization. All changes in this PR have already been reflected to the [`proc-macro2`], `quote`, and `syn` crates. New versions of all these crates have also been published to crates.io. Once this lands in nightly I plan on making an internals post again summarizing the changes made here and also calling on all macro authors to give the APIs a spin and see how they work. Hopefully pending no major issues we can then have an FCP to stabilize later this cycle! [`proc-macro2`]: https://docs.rs/proc-macro2/0.3.1/proc_macro2/
2018-04-02 10:19:32 -05:00
let tree = TokenTree::Token(span.0, token);
match kind {
2017-06-04 20:41:33 -05:00
Spacing::Alone => tree.into(),
Spacing::Joint => tree.joint(),
}
}
}
/// Permanently unstable internal implementation details of this crate. This
/// should not be used.
///
/// These methods are used by the rest of the compiler to generate instances of
/// `TokenStream` to hand to macro definitions, as well as consume the output.
///
/// Note that this module is also intentionally separate from the rest of the
/// crate. This allows the `#[unstable]` directive below to naturally apply to
/// all of the contents.
#[unstable(feature = "proc_macro_internals", issue = "27812")]
#[doc(hidden)]
pub mod __internal {
pub use quote::{LiteralKind, Quoter, unquote};
use std::cell::Cell;
use syntax::ast;
use syntax::ext::base::ExtCtxt;
use syntax::ext::hygiene::Mark;
use syntax::ptr::P;
2017-03-28 20:55:01 -05:00
use syntax::parse::{self, ParseSess};
use syntax::parse::token::{self, Token};
use syntax::tokenstream;
2017-08-01 20:05:08 -05:00
use syntax_pos::{BytePos, Loc, DUMMY_SP};
use super::{TokenStream, LexError};
2017-08-01 20:05:08 -05:00
pub fn lookup_char_pos(pos: BytePos) -> Loc {
with_sess(|(sess, _)| sess.codemap().lookup_char_pos(pos))
}
pub fn new_token_stream(item: P<ast::Item>) -> TokenStream {
2017-03-28 20:55:01 -05:00
let token = Token::interpolated(token::NtItem(item));
TokenStream(tokenstream::TokenTree::Token(DUMMY_SP, token).into())
}
pub fn token_stream_wrap(inner: tokenstream::TokenStream) -> TokenStream {
TokenStream(inner)
}
pub fn token_stream_parse_items(stream: TokenStream) -> Result<Vec<P<ast::Item>>, LexError> {
with_sess(move |(sess, _)| {
let mut parser = parse::stream_to_parser(sess, stream.0);
let mut items = Vec::new();
while let Some(item) = try!(parser.parse_item().map_err(super::parse_to_lex_err)) {
items.push(item)
}
Ok(items)
})
}
2016-08-04 14:20:01 -05:00
pub fn token_stream_inner(stream: TokenStream) -> tokenstream::TokenStream {
stream.0
}
2016-08-04 14:20:01 -05:00
pub trait Registry {
fn register_custom_derive(&mut self,
trait_name: &str,
expand: fn(TokenStream) -> TokenStream,
attributes: &[&'static str]);
fn register_attr_proc_macro(&mut self,
name: &str,
expand: fn(TokenStream, TokenStream) -> TokenStream);
fn register_bang_proc_macro(&mut self,
name: &str,
expand: fn(TokenStream) -> TokenStream);
}
// Emulate scoped_thread_local!() here essentially
thread_local! {
static CURRENT_SESS: Cell<(*const ParseSess, Mark)> =
Cell::new((0 as *const _, Mark::root()));
}
pub fn set_sess<F, R>(cx: &ExtCtxt, f: F) -> R
where F: FnOnce() -> R
{
struct Reset { prev: (*const ParseSess, Mark) }
impl Drop for Reset {
fn drop(&mut self) {
CURRENT_SESS.with(|p| p.set(self.prev));
}
}
CURRENT_SESS.with(|p| {
let _reset = Reset { prev: p.get() };
p.set((cx.parse_sess, cx.current_expansion.mark));
f()
})
}
pub fn in_sess() -> bool
{
let p = CURRENT_SESS.with(|p| p.get());
!p.0.is_null()
}
pub fn with_sess<F, R>(f: F) -> R
where F: FnOnce((&ParseSess, Mark)) -> R
{
let p = CURRENT_SESS.with(|p| p.get());
assert!(!p.0.is_null(), "proc_macro::__internal::with_sess() called \
before set_parse_sess()!");
f(unsafe { (&*p.0, p.1) })
}
}
fn parse_to_lex_err(mut err: DiagnosticBuilder) -> LexError {
err.cancel();
LexError { _inner: () }
}