rust/src/libsyntax_ext/proc_macro_registrar.rs

336 lines
12 KiB
Rust
Raw Normal View History

rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
// 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.
use std::mem;
use errors;
use syntax::ast::{self, Ident, NodeId};
use syntax::codemap::{ExpnInfo, NameAndSpan, MacroAttribute};
use syntax::ext::base::ExtCtxt;
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
use syntax::ext::build::AstBuilder;
use syntax::ext::expand::ExpansionConfig;
use syntax::parse::ParseSess;
use syntax::feature_gate::Features;
use syntax::fold::Folder;
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
use syntax::ptr::P;
use syntax::symbol::Symbol;
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
use syntax_pos::{Span, DUMMY_SP};
use syntax::visit::{self, Visitor};
use deriving;
struct CustomDerive {
trait_name: ast::Name,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
function_name: Ident,
span: Span,
attrs: Vec<ast::Name>,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}
struct CollectCustomDerives<'a> {
derives: Vec<CustomDerive>,
in_root: bool,
handler: &'a errors::Handler,
is_proc_macro_crate: bool,
is_test_crate: bool,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}
pub fn modify(sess: &ParseSess,
resolver: &mut ::syntax::ext::base::Resolver,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
mut krate: ast::Crate,
is_proc_macro_crate: bool,
is_test_crate: bool,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
num_crate_types: usize,
handler: &errors::Handler,
features: &Features) -> ast::Crate {
let ecfg = ExpansionConfig::default("proc_macro".to_string());
let mut cx = ExtCtxt::new(sess, ecfg, resolver);
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let mut collect = CollectCustomDerives {
derives: Vec::new(),
in_root: true,
handler: handler,
is_proc_macro_crate: is_proc_macro_crate,
is_test_crate: is_test_crate,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
};
visit::walk_crate(&mut collect, &krate);
if !is_proc_macro_crate {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
return krate
} else if !features.proc_macro {
let mut err = handler.struct_err("the `proc-macro` crate type is \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
experimental");
err.help("add #![feature(proc_macro)] to the crate attributes to \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
enable");
err.emit();
}
if num_crate_types > 1 {
handler.err("cannot mix `proc-macro` crate type with others");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}
if is_test_crate {
return krate;
}
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
krate.module.items.push(mk_registrar(&mut cx, &collect.derives));
if krate.exported_macros.len() > 0 {
handler.err("cannot export macro_rules! macros from a `proc-macro` \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
crate type currently");
}
return krate
}
impl<'a> CollectCustomDerives<'a> {
fn check_not_pub_in_root(&self, vis: &ast::Visibility, sp: Span) {
if self.is_proc_macro_crate &&
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
self.in_root &&
*vis == ast::Visibility::Public {
self.handler.span_err(sp,
"`proc-macro` crate types cannot \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
export any items other than functions \
tagged with `#[proc_macro_derive]` \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
currently");
}
}
}
impl<'a> Visitor for CollectCustomDerives<'a> {
fn visit_item(&mut self, item: &ast::Item) {
// First up, make sure we're checking a bare function. If we're not then
// we're just not interested in this item.
//
// If we find one, try to locate a `#[proc_macro_derive]` attribute on
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
// it.
match item.node {
ast::ItemKind::Fn(..) => {}
_ => {
// Check for invalid use of proc_macro_derive
let attr = item.attrs.iter()
.filter(|a| a.check_name("proc_macro_derive"))
.next();
if let Some(attr) = attr {
self.handler.span_err(attr.span(),
"the `#[proc_macro_derive]` \
attribute may only be used \
on bare functions");
return;
}
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
self.check_not_pub_in_root(&item.vis, item.span);
return visit::walk_item(self, item)
}
}
let mut attrs = item.attrs.iter()
.filter(|a| a.check_name("proc_macro_derive"));
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let attr = match attrs.next() {
Some(attr) => attr,
None => {
self.check_not_pub_in_root(&item.vis, item.span);
return visit::walk_item(self, item)
}
};
if let Some(a) = attrs.next() {
self.handler.span_err(a.span(), "multiple `#[proc_macro_derive]` \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
attributes found");
}
if self.is_test_crate {
return;
}
if !self.is_proc_macro_crate {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
self.handler.span_err(attr.span(),
"the `#[proc_macro_derive]` attribute is \
only usable with crates of the `proc-macro` \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
crate type");
}
// Once we've located the `#[proc_macro_derive]` attribute, verify
// that it's of the form `#[proc_macro_derive(Foo)]` or
// `#[proc_macro_derive(Foo, attributes(A, ..))]`
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let list = match attr.meta_item_list() {
Some(list) => list,
None => {
self.handler.span_err(attr.span(),
"attribute must be of form: \
#[proc_macro_derive(TraitName)]");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
return
}
};
if list.len() != 1 && list.len() != 2 {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
self.handler.span_err(attr.span(),
"attribute must have either one or two arguments");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
return
}
let trait_attr = &list[0];
let attributes_attr = list.get(1);
let trait_name = match trait_attr.name() {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
Some(name) => name,
_ => {
self.handler.span_err(trait_attr.span(), "not a meta item");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
return
}
};
if !trait_attr.is_word() {
self.handler.span_err(trait_attr.span(), "must only be one word");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}
if deriving::is_builtin_trait(trait_name) {
self.handler.span_err(trait_attr.span(),
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
"cannot override a built-in #[derive] mode");
}
if self.derives.iter().any(|d| d.trait_name == trait_name) {
self.handler.span_err(trait_attr.span(),
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
"derive mode defined twice in this crate");
}
let proc_attrs: Vec<_> = if let Some(attr) = attributes_attr {
if !attr.check_name("attributes") {
self.handler.span_err(attr.span(), "second argument must be `attributes`")
}
attr.meta_item_list().unwrap_or_else(|| {
self.handler.span_err(attr.span(),
"attribute must be of form: \
`attributes(foo, bar)`");
&[]
}).into_iter().filter_map(|attr| {
let name = match attr.name() {
Some(name) => name,
_ => {
self.handler.span_err(attr.span(), "not a meta item");
return None;
},
};
if !attr.is_word() {
self.handler.span_err(attr.span(), "must only be one word");
return None;
}
Some(name)
}).collect()
} else {
Vec::new()
};
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
if self.in_root {
self.derives.push(CustomDerive {
span: item.span,
trait_name: trait_name,
function_name: item.ident,
attrs: proc_attrs,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
});
} else {
let msg = "functions tagged with `#[proc_macro_derive]` must \
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
currently reside in the root of the crate";
self.handler.span_err(item.span, msg);
}
visit::walk_item(self, item);
}
fn visit_mod(&mut self, m: &ast::Mod, _s: Span, id: NodeId) {
let mut prev_in_root = self.in_root;
if id != ast::CRATE_NODE_ID {
prev_in_root = mem::replace(&mut self.in_root, false);
}
visit::walk_mod(self, m);
self.in_root = prev_in_root;
}
fn visit_mac(&mut self, mac: &ast::Mac) {
visit::walk_mac(self, mac)
}
}
// Creates a new module which looks like:
//
// mod $gensym {
// extern crate proc_macro;
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
//
// use proc_macro::__internal::Registry;
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
//
// #[plugin_registrar]
// fn registrar(registrar: &mut Registry) {
// registrar.register_custom_derive($name_trait1, ::$name1, &[]);
// registrar.register_custom_derive($name_trait2, ::$name2, &["attribute_name"]);
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
// // ...
// }
// }
fn mk_registrar(cx: &mut ExtCtxt,
custom_derives: &[CustomDerive]) -> P<ast::Item> {
let eid = cx.codemap().record_expansion(ExpnInfo {
call_site: DUMMY_SP,
callee: NameAndSpan {
format: MacroAttribute(Symbol::intern("proc_macro")),
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
span: None,
allow_internal_unstable: true,
}
});
let span = Span { expn_id: eid, ..DUMMY_SP };
let proc_macro = Ident::from_str("proc_macro");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let krate = cx.item(span,
proc_macro,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
Vec::new(),
ast::ItemKind::ExternCrate(None));
let __internal = Ident::from_str("__internal");
let registry = Ident::from_str("Registry");
let registrar = Ident::from_str("registrar");
let register_custom_derive = Ident::from_str("register_custom_derive");
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let stmts = custom_derives.iter().map(|cd| {
let path = cx.path_global(cd.span, vec![cd.function_name]);
let trait_name = cx.expr_str(cd.span, cd.trait_name);
let attrs = cx.expr_vec_slice(
span,
cd.attrs.iter().map(|&s| cx.expr_str(cd.span, s)).collect::<Vec<_>>()
);
(path, trait_name, attrs)
}).map(|(path, trait_name, attrs)| {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let registrar = cx.expr_ident(span, registrar);
let ufcs_path = cx.path(span, vec![proc_macro, __internal, registry,
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
register_custom_derive]);
cx.expr_call(span,
cx.expr_path(ufcs_path),
vec![registrar, trait_name, cx.expr_path(path), attrs])
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}).map(|expr| {
cx.stmt_expr(expr)
}).collect::<Vec<_>>();
let path = cx.path(span, vec![proc_macro, __internal, registry]);
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let registrar_path = cx.ty_path(path);
let arg_ty = cx.ty_rptr(span, registrar_path, None, ast::Mutability::Mutable);
let func = cx.item_fn(span,
registrar,
vec![cx.arg(span, registrar, arg_ty)],
cx.ty(span, ast::TyKind::Tup(Vec::new())),
cx.block(span, stmts));
let derive_registrar = cx.meta_word(span, Symbol::intern("rustc_derive_registrar"));
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
let derive_registrar = cx.attribute(span, derive_registrar);
let func = func.map(|mut i| {
i.attrs.push(derive_registrar);
i.vis = ast::Visibility::Public;
i
});
let ident = ast::Ident::with_empty_ctxt(Symbol::gensym("registrar"));
let module = cx.item_mod(span, span, ident, Vec::new(), vec![krate, func]).map(|mut i| {
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
i.vis = ast::Visibility::Public;
i
});
cx.monotonic_expander().fold_item(module).pop().unwrap()
rustc: Implement custom derive (macros 1.1) This commit is an implementation of [RFC 1681] which adds support to the compiler for first-class user-define custom `#[derive]` modes with a far more stable API than plugins have today. [RFC 1681]: https://github.com/rust-lang/rfcs/blob/master/text/1681-macros-1.1.md The main features added by this commit are: * A new `rustc-macro` crate-type. This crate type represents one which will provide custom `derive` implementations and perhaps eventually flower into the implementation of macros 2.0 as well. * A new `rustc_macro` crate in the standard distribution. This crate will provide the runtime interface between macro crates and the compiler. The API here is particularly conservative right now but has quite a bit of room to expand into any manner of APIs required by macro authors. * The ability to load new derive modes through the `#[macro_use]` annotations on other crates. All support added here is gated behind the `rustc_macro` feature gate, both for the library support (the `rustc_macro` crate) as well as the language features. There are a few minor differences from the implementation outlined in the RFC, such as the `rustc_macro` crate being available as a dylib and all symbols are `dlsym`'d directly instead of having a shim compiled. These should only affect the implementation, however, not the public interface. This commit also ended up touching a lot of code related to `#[derive]`, making a few notable changes: * Recognized derive attributes are no longer desugared to `derive_Foo`. Wasn't sure how to keep this behavior and *not* expose it to custom derive. * Derive attributes no longer have access to unstable features by default, they have to opt in on a granular level. * The `derive(Copy,Clone)` optimization is now done through another "obscure attribute" which is just intended to ferry along in the compiler that such an optimization is possible. The `derive(PartialEq,Eq)` optimization was also updated to do something similar. --- One part of this PR which needs to be improved before stabilizing are the errors and exact interfaces here. The error messages are relatively poor quality and there are surprising spects of this such as `#[derive(PartialEq, Eq, MyTrait)]` not working by default. The custom attributes added by the compiler end up becoming unstable again when going through a custom impl. Hopefully though this is enough to start allowing experimentation on crates.io! syntax-[breaking-change]
2016-08-22 17:07:11 -07:00
}