rust/src/librustdoc/html/render.rs

1641 lines
59 KiB
Rust
Raw Normal View History

// Copyright 2013 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.
//! Rustdoc's HTML Rendering module
//!
//! This modules contains the bulk of the logic necessary for rendering a
//! rustdoc `clean::Crate` instance to a set of static HTML pages. This
//! rendering process is largely driven by the `format!` syntax extension to
//! perform all I/O into files and streams.
//!
//! The rendering process is largely driven by the `Context` and `Cache`
//! structures. The cache is pre-populated by crawling the crate in question,
//! and then it is shared among the various rendering tasks. The cache is meant
//! to be a fairly large structure not implementing `Clone` (because it's shared
//! among tasks). The context, however, should be a lightweight structure. This
//! is cloned per-task and contains information about what is currently being
//! rendered.
//!
//! In order to speed up rendering (mostly because of markdown rendering), the
//! rendering process has been parallelized. This parallelization is only
//! exposed through the `crate` method on the context, and then also from the
//! fact that the shared cache is stored in TLS (and must be accessed as such).
//!
//! In addition to rendering the crate itself, this module is also responsible
//! for creating the corresponding search index and source file renderings.
//! These tasks are not parallelized (they haven't been a bottleneck yet), and
//! both occur before the crate is rendered.
use std::cell::Cell;
use std::comm::{SharedPort, SharedChan};
use std::comm;
use std::fmt;
use std::hashmap::{HashMap, HashSet};
use std::local_data;
2013-11-10 22:46:32 -08:00
use std::io::buffered::BufferedWriter;
use std::io;
use std::io::fs;
use std::io::File;
use std::os;
use std::str;
use std::task;
use std::unstable::finally::Finally;
use std::vec;
use extra::arc::RWArc;
use extra::json::ToJson;
use extra::sort;
use syntax::ast;
use syntax::attr;
use clean;
use doctree;
use fold::DocFolder;
use html::escape::Escape;
use html::format::{VisSpace, Method, PuritySpace};
use html::layout;
use html::markdown::Markdown;
/// Major driving force in all rustdoc rendering. This contains information
/// about where in the tree-like hierarchy rendering is occurring and controls
/// how the current page is being rendered.
///
/// It is intended that this context is a lightweight object which can be fairly
/// easily cloned because it is cloned per work-job (about once per item in the
/// rustdoc tree).
#[deriving(Clone)]
pub struct Context {
/// Current hierarchy of components leading down to what's currently being
/// rendered
current: ~[~str],
/// String representation of how to get back to the root path of the 'doc/'
/// folder in terms of a relative URL.
root_path: ~str,
/// The current destination folder of where HTML artifacts should be placed.
/// This changes as the context descends into the module hierarchy.
dst: Path,
/// This describes the layout of each page, and is not modified after
/// creation of the context (contains info like the favicon)
layout: layout::Layout,
/// This map is a list of what should be displayed on the sidebar of the
/// current page. The key is the section header (traits, modules,
/// functions), and the value is the list of containers belonging to this
/// header. This map will change depending on the surrounding context of the
/// page.
sidebar: HashMap<~str, ~[~str]>,
/// This flag indicates whether [src] links should be generated or not. If
/// the source files are present in the html rendering, then this will be
/// `true`.
include_sources: bool,
}
/// Indicates where an external crate can be found.
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
pub enum ExternalLocation {
/// Remote URL root of the external crate
Remote(~str),
/// This external crate can be found in the local doc/ folder
Local,
/// The external crate could not be found.
Unknown,
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
}
/// Different ways an implementor of a trait can be rendered.
enum Implementor {
/// Paths are displayed specially by omitting the `impl XX for` cruft
PathType(clean::Type),
/// This is the generic representation of an trait implementor, used for
/// primitive types and otherwise non-path types.
OtherType(clean::Generics, /* trait */ clean::Type, /* for */ clean::Type),
}
/// This cache is used to store information about the `clean::Crate` being
/// rendered in order to provide more useful documentation. This contains
/// information like all implementors of a trait, all traits a type implements,
/// documentation for all known traits, etc.
///
/// This structure purposefully does not implement `Clone` because it's intended
/// to be a fairly large and expensive structure to clone. Instead this adheres
/// to both `Send` and `Freeze` so it may be stored in a `RWArc` instance and
/// shared among the various rendering tasks.
pub struct Cache {
/// Mapping of typaram ids to the name of the type parameter. This is used
/// when pretty-printing a type (so pretty printing doesn't have to
/// painfully maintain a context like this)
typarams: HashMap<ast::NodeId, ~str>,
/// Maps a type id to all known implementations for that type. This is only
/// recognized for intra-crate `ResolvedPath` types, and is used to print
/// out extra documentation on the page of an enum/struct.
///
/// The values of the map are a list of implementations and documentation
/// found on that implementation.
impls: HashMap<ast::NodeId, ~[(clean::Impl, Option<~str>)]>,
/// Maintains a mapping of local crate node ids to the fully qualified name
/// and "short type description" of that node. This is used when generating
/// URLs when a type is being linked to. External paths are not located in
/// this map because the `External` type itself has all the information
/// necessary.
paths: HashMap<ast::NodeId, (~[~str], &'static str)>,
/// This map contains information about all known traits of this crate.
/// Implementations of a crate should inherit the documentation of the
/// parent trait if no extra documentation is specified, and default methods
/// should show up in documentation about trait implementations.
traits: HashMap<ast::NodeId, clean::Trait>,
/// When rendering traits, it's often useful to be able to list all
/// implementors of the trait, and this mapping is exactly, that: a mapping
/// of trait ids to the list of known implementors of the trait
implementors: HashMap<ast::NodeId, ~[Implementor]>,
/// Cache of where external crate documentation can be found.
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
extern_locations: HashMap<ast::CrateNum, ExternalLocation>,
// Private fields only used when initially crawling a crate to build a cache
priv stack: ~[~str],
priv parent_stack: ~[ast::NodeId],
priv search_index: ~[IndexItem],
}
/// Helper struct to render all source code to HTML pages
struct SourceCollector<'self> {
cx: &'self mut Context,
/// Processed source-file paths
seen: HashSet<~str>,
/// Root destination to place all HTML output into
dst: Path,
}
/// Wrapper struct to render the source code of a file. This will do things like
/// adding line numbers to the left-hand side.
struct Source<'self>(&'self str);
// Helper structs for rendering items/sidebars and carrying along contextual
// information
struct Item<'self> { cx: &'self Context, item: &'self clean::Item, }
struct Sidebar<'self> { cx: &'self Context, item: &'self clean::Item, }
/// Struct representing one entry in the JS search index. These are all emitted
/// by hand to a large JS file at the end of cache-creation.
struct IndexItem {
ty: &'static str,
name: ~str,
path: ~str,
desc: ~str,
parent: Option<ast::NodeId>,
}
// TLS keys used to carry information around during rendering.
local_data_key!(pub cache_key: RWArc<Cache>)
local_data_key!(pub current_location_key: ~[~str])
/// Generates the documentation for `crate` into the directory `dst`
pub fn run(mut crate: clean::Crate, dst: Path) {
let mut cx = Context {
dst: dst,
current: ~[],
root_path: ~"",
sidebar: HashMap::new(),
layout: layout::Layout {
logo: ~"",
favicon: ~"",
crate: crate.name.clone(),
},
include_sources: true,
};
mkdir(&cx.dst);
match crate.module.as_ref().map(|m| m.doc_list().unwrap_or(&[])) {
Some(attrs) => {
for attr in attrs.iter() {
match *attr {
clean::NameValue(~"html_favicon_url", ref s) => {
cx.layout.favicon = s.to_owned();
}
clean::NameValue(~"html_logo_url", ref s) => {
cx.layout.logo = s.to_owned();
}
clean::Word(~"html_no_source") => {
cx.include_sources = false;
}
_ => {}
}
}
}
None => {}
}
// Crawl the crate to build various caches used for the output
let mut cache = Cache {
impls: HashMap::new(),
typarams: HashMap::new(),
paths: HashMap::new(),
traits: HashMap::new(),
implementors: HashMap::new(),
stack: ~[],
parent_stack: ~[],
search_index: ~[],
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
extern_locations: HashMap::new(),
};
cache.stack.push(crate.name.clone());
crate = cache.fold_crate(crate);
// Add all the static files
let mut dst = cx.dst.join(crate.name.as_slice());
mkdir(&dst);
write(dst.join("jquery.js"), include_str!("static/jquery-2.0.3.min.js"));
write(dst.join("main.js"), include_str!("static/main.js"));
write(dst.join("main.css"), include_str!("static/main.css"));
write(dst.join("normalize.css"), include_str!("static/normalize.css"));
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
// Publish the search index
{
dst.push("search-index.js");
let mut w = BufferedWriter::new(File::create(&dst).unwrap());
let w = &mut w as &mut Writer;
write!(w, "var searchIndex = [");
for (i, item) in cache.search_index.iter().enumerate() {
if i > 0 { write!(w, ","); }
write!(w, "\\{ty:\"{}\",name:\"{}\",path:\"{}\",desc:{}",
item.ty, item.name, item.path,
item.desc.to_json().to_str())
match item.parent {
Some(id) => { write!(w, ",parent:'{}'", id); }
None => {}
}
write!(w, "\\}");
}
write!(w, "];");
write!(w, "var allPaths = \\{");
for (i, (&id, &(ref fqp, short))) in cache.paths.iter().enumerate() {
if i > 0 { write!(w, ","); }
write!(w, "'{}':\\{type:'{}',name:'{}'\\}", id, short, *fqp.last());
}
write!(w, "\\};");
w.flush();
}
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
// Render all source files (this may turn into a giant no-op)
{
info!("emitting source files");
let dst = cx.dst.join("src");
mkdir(&dst);
let dst = dst.join(crate.name.as_slice());
mkdir(&dst);
let mut folder = SourceCollector {
dst: dst,
seen: HashSet::new(),
cx: &mut cx,
};
crate = folder.fold_crate(crate);
}
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
for (&n, e) in crate.externs.iter() {
cache.extern_locations.insert(n, extern_location(e, &cx.dst));
}
// And finally render the whole crate's documentation
cx.crate(crate, cache);
}
/// Writes the entire contents of a string to a destination, not attempting to
/// catch any errors.
fn write(dst: Path, contents: &str) {
File::create(&dst).write(contents.as_bytes());
}
/// Makes a directory on the filesystem, failing the task if an error occurs and
/// skipping if the directory already exists.
fn mkdir(path: &Path) {
do io::io_error::cond.trap(|err| {
error!("Couldn't create directory `{}`: {}",
path.display(), err.desc);
fail!()
}).inside {
if !path.is_dir() {
fs::mkdir(path, io::UserRWX);
}
}
}
/// Takes a path to a source file and cleans the path to it. This canonicalizes
/// things like ".." to components which preserve the "top down" hierarchy of a
/// static HTML tree.
// FIXME (#9639): The closure should deal with &[u8] instead of &str
fn clean_srcpath(src: &[u8], f: &fn(&str)) {
let p = Path::new(src);
if p.as_vec() != bytes!(".") {
for c in p.str_component_iter().map(|x|x.unwrap()) {
if ".." == c {
f("up");
} else {
f(c.as_slice())
}
}
}
}
/// Attempts to find where an external crate is located, given that we're
/// rendering in to the specified source destination.
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
fn extern_location(e: &clean::ExternalCrate, dst: &Path) -> ExternalLocation {
// See if there's documentation generated into the local directory
let local_location = dst.join(e.name.as_slice());
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
if local_location.is_dir() {
return Local;
}
// Failing that, see if there's an attribute specifying where to find this
// external crate
for attr in e.attrs.iter() {
match *attr {
clean::List(~"doc", ref list) => {
for attr in list.iter() {
match *attr {
clean::NameValue(~"html_root_url", ref s) => {
if s.ends_with("/") {
return Remote(s.to_owned());
}
return Remote(*s + "/");
}
_ => {}
}
}
}
_ => {}
}
}
// Well, at least we tried.
return Unknown;
}
impl<'self> DocFolder for SourceCollector<'self> {
fn fold_item(&mut self, item: clean::Item) -> Option<clean::Item> {
// If we're including source files, and we haven't seen this file yet,
// then we need to render it out to the filesystem
if self.cx.include_sources && !self.seen.contains(&item.source.filename) {
// If it turns out that we couldn't read this file, then we probably
// can't read any of the files (generating html output from json or
// something like that), so just don't include sources for the
// entire crate. The other option is maintaining this mapping on a
// per-file basis, but that's probably not worth it...
self.cx.include_sources = self.emit_source(item.source.filename);
self.seen.insert(item.source.filename.clone());
if !self.cx.include_sources {
println!("warning: source code was requested to be rendered, \
but `{}` is a missing source file.",
item.source.filename);
println!(" skipping rendering of source code");
}
}
self.fold_item_recur(item)
}
}
impl<'self> SourceCollector<'self> {
/// Renders the given filename into its corresponding HTML source file.
fn emit_source(&mut self, filename: &str) -> bool {
let p = Path::new(filename);
// Read the contents of the file
let mut contents = ~[];
{
let mut buf = [0, ..1024];
// If we couldn't open this file, then just returns because it
// probably means that it's some standard library macro thing and we
// can't have the source to it anyway.
let mut r = match io::result(|| File::open(&p)) {
Ok(r) => r,
// eew macro hacks
Err(*) => return filename == "<std-macros>"
};
// read everything
loop {
match r.read(buf) {
Some(n) => contents.push_all(buf.slice_to(n)),
None => break
}
}
}
let contents = str::from_utf8_owned(contents);
// Create the intermediate directories
let mut cur = self.dst.clone();
let mut root_path = ~"../../";
do clean_srcpath(p.dirname()) |component| {
cur.push(component);
mkdir(&cur);
root_path.push_str("../");
}
cur.push(p.filename().expect("source has no filename") + bytes!(".html"));
let mut w = BufferedWriter::new(File::create(&cur).unwrap());
let title = cur.filename_display().with_str(|s| format!("{} -- source", s));
let page = layout::Page {
title: title,
ty: "source",
root_path: root_path,
};
layout::render(&mut w as &mut Writer, &self.cx.layout,
&page, &(""), &Source(contents.as_slice()));
w.flush();
return true;
}
}
impl DocFolder for Cache {
fn fold_item(&mut self, item: clean::Item) -> Option<clean::Item> {
// Register any generics to their corresponding string. This is used
// when pretty-printing types
match item.inner {
clean::StructItem(ref s) => self.generics(&s.generics),
clean::EnumItem(ref e) => self.generics(&e.generics),
clean::FunctionItem(ref f) => self.generics(&f.generics),
clean::TypedefItem(ref t) => self.generics(&t.generics),
clean::TraitItem(ref t) => self.generics(&t.generics),
clean::ImplItem(ref i) => self.generics(&i.generics),
clean::TyMethodItem(ref i) => self.generics(&i.generics),
clean::MethodItem(ref i) => self.generics(&i.generics),
clean::ForeignFunctionItem(ref f) => self.generics(&f.generics),
_ => {}
}
// Propagate a trait methods' documentation to all implementors of the
// trait
match item.inner {
clean::TraitItem(ref t) => {
self.traits.insert(item.id, t.clone());
}
_ => {}
}
// Collect all the implementors of traits.
match item.inner {
clean::ImplItem(ref i) => {
match i.trait_ {
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
Some(clean::ResolvedPath{ id, _ }) => {
let v = do self.implementors.find_or_insert_with(id) |_|{
~[]
};
match i.for_ {
clean::ResolvedPath{_} => {
v.unshift(PathType(i.for_.clone()));
}
_ => {
v.push(OtherType(i.generics.clone(),
i.trait_.get_ref().clone(),
i.for_.clone()));
}
}
}
Some(*) | None => {}
}
}
_ => {}
}
// Index this method for searching later on
match item.name {
Some(ref s) => {
let parent = match item.inner {
clean::TyMethodItem(*) |
clean::StructFieldItem(*) |
clean::VariantItem(*) => {
Some((Some(*self.parent_stack.last()),
self.stack.slice_to(self.stack.len() - 1)))
}
clean::MethodItem(*) => {
if self.parent_stack.len() == 0 {
None
} else {
let last = self.parent_stack.last();
let amt = match self.paths.find(last) {
Some(&(_, "trait")) => self.stack.len() - 1,
Some(*) | None => self.stack.len(),
};
Some((Some(*last), self.stack.slice_to(amt)))
}
}
_ => Some((None, self.stack.as_slice()))
};
match parent {
Some((parent, path)) => {
self.search_index.push(IndexItem {
ty: shortty(&item),
name: s.to_owned(),
path: path.connect("::"),
desc: shorter(item.doc_value()).to_owned(),
parent: parent,
});
}
None => {}
}
}
None => {}
}
// Keep track of the fully qualified path for this item.
let pushed = if item.name.is_some() {
let n = item.name.get_ref();
if n.len() > 0 {
self.stack.push(n.to_owned());
true
} else { false }
} else { false };
match item.inner {
clean::StructItem(*) | clean::EnumItem(*) |
clean::TypedefItem(*) | clean::TraitItem(*) |
clean::FunctionItem(*) | clean::ModuleItem(*) |
clean::ForeignFunctionItem(*) | clean::VariantItem(*) => {
self.paths.insert(item.id, (self.stack.clone(), shortty(&item)));
}
_ => {}
}
// Maintain the parent stack
let parent_pushed = match item.inner {
clean::TraitItem(*) | clean::EnumItem(*) | clean::StructItem(*) => {
self.parent_stack.push(item.id); true
}
clean::ImplItem(ref i) => {
match i.for_ {
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
clean::ResolvedPath{ id, _ } => {
self.parent_stack.push(id); true
}
_ => false
}
}
_ => false
};
// Once we've recursively found all the generics, then hoard off all the
// implementations elsewhere
let ret = match self.fold_item_recur(item) {
Some(item) => {
match item {
clean::Item{ attrs, inner: clean::ImplItem(i), _ } => {
match i.for_ {
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
clean::ResolvedPath { id, _ } => {
let v = do self.impls.find_or_insert_with(id) |_| {
~[]
};
// extract relevant documentation for this impl
match attrs.move_iter().find(|a| {
match *a {
clean::NameValue(~"doc", _) => true,
_ => false
}
}) {
Some(clean::NameValue(_, dox)) => {
v.push((i, Some(dox)));
}
Some(*) | None => {
v.push((i, None));
}
}
}
_ => {}
}
None
}
// Private modules may survive the strip-private pass if
// they contain impls for public types, but those will get
// stripped here
clean::Item { inner: clean::ModuleItem(ref m), _ }
if m.items.len() == 0 => None,
i => Some(i),
}
}
i => i,
};
if pushed { self.stack.pop(); }
if parent_pushed { self.parent_stack.pop(); }
return ret;
}
}
impl<'self> Cache {
fn generics(&mut self, generics: &clean::Generics) {
for typ in generics.type_params.iter() {
self.typarams.insert(typ.id, typ.name.clone());
}
}
}
impl Context {
/// Recurse in the directory structure and change the "root path" to make
/// sure it always points to the top (relatively)
fn recurse<T>(&mut self, s: ~str, f: &fn(&mut Context) -> T) -> T {
if s.len() == 0 {
fail!("what {:?}", self);
}
let prev = self.dst.clone();
self.dst.push(s.as_slice());
self.root_path.push_str("../");
self.current.push(s);
mkdir(&self.dst);
let ret = f(self);
// Go back to where we were at
self.dst = prev;
let len = self.root_path.len();
self.root_path.truncate(len - 3);
self.current.pop();
return ret;
}
/// Main method for rendering a crate. This parallelizes the task of
/// rendering a crate, and requires ownership of the crate in order to break
/// it up into its separate components.
fn crate(self, mut crate: clean::Crate, cache: Cache) {
enum Work {
Die,
Process(Context, clean::Item),
}
enum Progress { JobNew, JobDone }
let workers = match os::getenv("RUSTDOC_WORKERS") {
Some(s) => {
match from_str::<uint>(s) {
Some(n) => n, None => fail!("{} not a number", s)
}
}
None => 10,
};
let mut item = match crate.module.take() {
Some(i) => i,
None => return
};
item.name = Some(crate.name);
let (port, chan) = comm::stream::<Work>();
let port = SharedPort::new(port);
let chan = SharedChan::new(chan);
let (prog_port, prog_chan) = comm::stream();
let prog_chan = SharedChan::new(prog_chan);
let cache = RWArc::new(cache);
// Each worker thread receives work from a shared port and publishes
// new work onto the corresponding shared port. All of the workers are
// using the same channel/port. Through this, the crate is recursed on
// in a hierarchical fashion, and parallelization is only achieved if
// one node in the hierarchy has more than one child (very common).
for i in range(0, workers) {
let port = port.clone();
let chan = chan.clone();
let prog_chan = prog_chan.clone();
let mut task = task::task();
task.name(format!("worker{}", i));
let cache = cache.clone();
do task.spawn {
worker(cache, &port, &chan, &prog_chan);
}
fn worker(cache: RWArc<Cache>,
port: &SharedPort<Work>,
chan: &SharedChan<Work>,
prog_chan: &SharedChan<Progress>) {
local_data::set(cache_key, cache);
loop {
match port.recv() {
Process(cx, item) => {
let mut cx = cx;
let item = Cell::new(item);
do (|| {
do cx.item(item.take()) |cx, item| {
prog_chan.send(JobNew);
chan.send(Process(cx.clone(), item));
}
}).finally {
// If we fail, everything else should still get
// completed
prog_chan.send(JobDone);
}
}
Die => break,
}
}
}
}
// Send off the initial job
chan.send(Process(self, item));
let mut jobs = 1;
// Keep track of the number of jobs active in the system and kill
// everything once there are no more jobs remaining.
loop {
match prog_port.recv() {
JobNew => jobs += 1,
JobDone => jobs -= 1,
}
if jobs == 0 { break }
}
for _ in range(0, workers) {
chan.send(Die);
}
}
/// Non-parellelized version of rendering an item. This will take the input
/// item, render its contents, and then invoke the specified closure with
/// all sub-items which need to be rendered.
///
/// The rendering driver uses this closure to queue up more work.
fn item(&mut self, item: clean::Item, f: &fn(&mut Context, clean::Item)) {
fn render(w: io::File, cx: &mut Context, it: &clean::Item,
pushname: bool) {
// A little unfortunate that this is done like this, but it sure
// does make formatting *a lot* nicer.
local_data::set(current_location_key, cx.current.clone());
let mut title = cx.current.connect("::");
if pushname {
if title.len() > 0 { title.push_str("::"); }
title.push_str(*it.name.get_ref());
}
title.push_str(" - Rust");
let page = layout::Page {
ty: shortty(it),
root_path: cx.root_path,
title: title,
};
// We have a huge number of calls to write, so try to alleviate some
// of the pain by using a buffered writer instead of invoking the
// write sycall all the time.
let mut writer = BufferedWriter::new(w);
layout::render(&mut writer as &mut Writer, &cx.layout, &page,
&Sidebar{ cx: cx, item: it },
&Item{ cx: cx, item: it });
writer.flush();
}
match item.inner {
// modules are special because they add a namespace. We also need to
// recurse into the items of the module as well.
clean::ModuleItem(*) => {
let name = item.name.get_ref().to_owned();
let item = Cell::new(item);
do self.recurse(name) |this| {
let item = item.take();
let dst = this.dst.join("index.html");
render(File::create(&dst).unwrap(), this, &item, false);
let m = match item.inner {
clean::ModuleItem(m) => m,
_ => unreachable!()
};
this.sidebar = build_sidebar(&m);
for item in m.items.move_iter() {
f(this, item);
}
}
}
// Things which don't have names (like impls) don't get special
// pages dedicated to them.
_ if item.name.is_some() => {
let dst = self.dst.join(item_path(&item));
render(File::create(&dst).unwrap(), self, &item, true);
}
_ => {}
}
}
}
fn shortty(item: &clean::Item) -> &'static str {
match item.inner {
clean::ModuleItem(*) => "mod",
clean::StructItem(*) => "struct",
clean::EnumItem(*) => "enum",
clean::FunctionItem(*) => "fn",
clean::TypedefItem(*) => "typedef",
clean::StaticItem(*) => "static",
clean::TraitItem(*) => "trait",
clean::ImplItem(*) => "impl",
clean::ViewItemItem(*) => "viewitem",
clean::TyMethodItem(*) => "tymethod",
clean::MethodItem(*) => "method",
clean::StructFieldItem(*) => "structfield",
clean::VariantItem(*) => "variant",
clean::ForeignFunctionItem(*) => "ffi",
clean::ForeignStaticItem(*) => "ffs",
}
}
impl<'self> Item<'self> {
fn ismodule(&self) -> bool {
match self.item.inner {
clean::ModuleItem(*) => true, _ => false
}
}
}
impl<'self> fmt::Default for Item<'self> {
fn fmt(it: &Item<'self>, fmt: &mut fmt::Formatter) {
match attr::find_stability(it.item.attrs.iter()) {
Some(stability) => {
write!(fmt.buf,
"<a class='stability {lvl}' title='{reason}'>{lvl}</a>",
lvl = stability.level.to_str(),
reason = match stability.text {
Some(s) => s, None => @"",
});
}
None => {}
}
if it.cx.include_sources {
let mut path = ~[];
do clean_srcpath(it.item.source.filename.as_bytes()) |component| {
path.push(component.to_owned());
}
let href = if it.item.source.loline == it.item.source.hiline {
format!("{}", it.item.source.loline)
} else {
format!("{}-{}", it.item.source.loline, it.item.source.hiline)
};
write!(fmt.buf,
"<a class='source'
href='{root}src/{crate}/{path}.html\\#{href}'>[src]</a>",
root = it.cx.root_path,
crate = it.cx.layout.crate,
path = path.connect("/"),
href = href);
}
// Write the breadcrumb trail header for the top
write!(fmt.buf, "<h1 class='fqn'>");
match it.item.inner {
clean::ModuleItem(*) => write!(fmt.buf, "Module "),
clean::FunctionItem(*) => write!(fmt.buf, "Function "),
clean::TraitItem(*) => write!(fmt.buf, "Trait "),
clean::StructItem(*) => write!(fmt.buf, "Struct "),
clean::EnumItem(*) => write!(fmt.buf, "Enum "),
_ => {}
}
let cur = it.cx.current.as_slice();
let amt = if it.ismodule() { cur.len() - 1 } else { cur.len() };
for (i, component) in cur.iter().enumerate().take(amt) {
let mut trail = ~"";
for _ in range(0, cur.len() - i - 1) {
trail.push_str("../");
}
write!(fmt.buf, "<a href='{}index.html'>{}</a>::",
trail, component.as_slice());
}
write!(fmt.buf, "<a class='{}' href=''>{}</a></h1>",
shortty(it.item), it.item.name.get_ref().as_slice());
match it.item.inner {
clean::ModuleItem(ref m) => item_module(fmt.buf, it.cx,
it.item, m.items),
clean::FunctionItem(ref f) | clean::ForeignFunctionItem(ref f) =>
item_function(fmt.buf, it.item, f),
clean::TraitItem(ref t) => item_trait(fmt.buf, it.item, t),
clean::StructItem(ref s) => item_struct(fmt.buf, it.item, s),
clean::EnumItem(ref e) => item_enum(fmt.buf, it.item, e),
clean::TypedefItem(ref t) => item_typedef(fmt.buf, it.item, t),
_ => {}
}
}
}
fn item_path(item: &clean::Item) -> ~str {
match item.inner {
clean::ModuleItem(*) => *item.name.get_ref() + "/index.html",
_ => shortty(item) + "." + *item.name.get_ref() + ".html"
}
}
fn full_path(cx: &Context, item: &clean::Item) -> ~str {
let mut s = cx.current.connect("::");
s.push_str("::");
s.push_str(item.name.get_ref().as_slice());
return s;
}
fn blank<'a>(s: Option<&'a str>) -> &'a str {
match s {
Some(s) => s,
None => ""
}
}
fn shorter<'a>(s: Option<&'a str>) -> &'a str {
match s {
Some(s) => match s.find_str("\n\n") {
Some(pos) => s.slice_to(pos),
None => s,
},
None => ""
}
}
fn document(w: &mut Writer, item: &clean::Item) {
match item.doc_value() {
Some(s) => {
write!(w, "<div class='docblock'>{}</div>", Markdown(s));
}
None => {}
}
}
fn item_module(w: &mut Writer, cx: &Context,
item: &clean::Item, items: &[clean::Item]) {
document(w, item);
debug!("{:?}", items);
let mut indices = vec::from_fn(items.len(), |i| i);
fn lt(i1: &clean::Item, i2: &clean::Item, idx1: uint, idx2: uint) -> bool {
if shortty(i1) == shortty(i2) {
return i1.name < i2.name;
}
match (&i1.inner, &i2.inner) {
(&clean::ViewItemItem(ref a), &clean::ViewItemItem(ref b)) => {
match (&a.inner, &b.inner) {
(&clean::ExternMod(*), _) => true,
(_, &clean::ExternMod(*)) => false,
_ => idx1 < idx2,
}
}
(&clean::ViewItemItem(*), _) => true,
(_, &clean::ViewItemItem(*)) => false,
(&clean::ModuleItem(*), _) => true,
(_, &clean::ModuleItem(*)) => false,
(&clean::StructItem(*), _) => true,
(_, &clean::StructItem(*)) => false,
(&clean::EnumItem(*), _) => true,
(_, &clean::EnumItem(*)) => false,
(&clean::StaticItem(*), _) => true,
(_, &clean::StaticItem(*)) => false,
(&clean::ForeignFunctionItem(*), _) => true,
(_, &clean::ForeignFunctionItem(*)) => false,
(&clean::ForeignStaticItem(*), _) => true,
(_, &clean::ForeignStaticItem(*)) => false,
(&clean::TraitItem(*), _) => true,
(_, &clean::TraitItem(*)) => false,
(&clean::FunctionItem(*), _) => true,
(_, &clean::FunctionItem(*)) => false,
(&clean::TypedefItem(*), _) => true,
(_, &clean::TypedefItem(*)) => false,
_ => idx1 < idx2,
}
}
debug!("{:?}", indices);
do sort::quick_sort(indices) |&i1, &i2| {
lt(&items[i1], &items[i2], i1, i2)
}
debug!("{:?}", indices);
let mut curty = "";
for &idx in indices.iter() {
let myitem = &items[idx];
let myty = shortty(myitem);
if myty != curty {
if curty != "" {
write!(w, "</table>");
}
curty = myty;
write!(w, "<h2>{}</h2>\n<table>", match myitem.inner {
clean::ModuleItem(*) => "Modules",
clean::StructItem(*) => "Structs",
clean::EnumItem(*) => "Enums",
clean::FunctionItem(*) => "Functions",
clean::TypedefItem(*) => "Type Definitions",
clean::StaticItem(*) => "Statics",
clean::TraitItem(*) => "Traits",
clean::ImplItem(*) => "Implementations",
clean::ViewItemItem(*) => "Reexports",
clean::TyMethodItem(*) => "Type Methods",
clean::MethodItem(*) => "Methods",
clean::StructFieldItem(*) => "Struct Fields",
clean::VariantItem(*) => "Variants",
clean::ForeignFunctionItem(*) => "Foreign Functions",
clean::ForeignStaticItem(*) => "Foreign Statics",
});
}
match myitem.inner {
clean::StaticItem(ref s) | clean::ForeignStaticItem(ref s) => {
struct Initializer<'self>(&'self str);
impl<'self> fmt::Default for Initializer<'self> {
fn fmt(s: &Initializer<'self>, f: &mut fmt::Formatter) {
if s.len() == 0 { return; }
write!(f.buf, "<code> = </code>");
let tag = if s.contains("\n") { "pre" } else { "code" };
write!(f.buf, "<{tag}>{}</{tag}>",
s.as_slice(), tag=tag);
}
}
write!(w, "
<tr>
<td><code>{}static {}: {}</code>{}</td>
<td class='docblock'>{}&nbsp;</td>
</tr>
",
VisSpace(myitem.visibility),
*myitem.name.get_ref(),
s.type_,
Initializer(s.expr),
Markdown(blank(myitem.doc_value())));
}
clean::ViewItemItem(ref item) => {
match item.inner {
clean::ExternMod(ref name, ref src, _, _) => {
write!(w, "<tr><td><code>extern mod {}",
name.as_slice());
match *src {
Some(ref src) => write!(w, " = \"{}\"",
src.as_slice()),
None => {}
}
write!(w, ";</code></td></tr>");
}
clean::Import(ref imports) => {
for import in imports.iter() {
write!(w, "<tr><td><code>{}{}</code></td></tr>",
VisSpace(myitem.visibility),
*import);
}
}
}
}
_ => {
if myitem.name.is_none() { continue }
write!(w, "
<tr>
<td><a class='{class}' href='{href}'
title='{title}'>{}</a></td>
<td class='docblock short'>{}</td>
</tr>
",
*myitem.name.get_ref(),
Markdown(shorter(myitem.doc_value())),
class = shortty(myitem),
href = item_path(myitem),
title = full_path(cx, myitem));
}
}
}
write!(w, "</table>");
}
fn item_function(w: &mut Writer, it: &clean::Item, f: &clean::Function) {
write!(w, "<pre class='fn'>{vis}{purity}fn {name}{generics}{decl}</pre>",
vis = VisSpace(it.visibility),
purity = PuritySpace(f.purity),
name = it.name.get_ref().as_slice(),
generics = f.generics,
decl = f.decl);
document(w, it);
}
fn item_trait(w: &mut Writer, it: &clean::Item, t: &clean::Trait) {
let mut parents = ~"";
if t.parents.len() > 0 {
parents.push_str(": ");
for (i, p) in t.parents.iter().enumerate() {
if i > 0 { parents.push_str(" + "); }
parents.push_str(format!("{}", *p));
}
}
// Output the trait definition
write!(w, "<pre class='trait'>{}trait {}{}{} ",
VisSpace(it.visibility),
it.name.get_ref().as_slice(),
t.generics,
parents);
let required = t.methods.iter().filter(|m| m.is_req()).to_owned_vec();
let provided = t.methods.iter().filter(|m| !m.is_req()).to_owned_vec();
if t.methods.len() == 0 {
write!(w, "\\{ \\}");
} else {
write!(w, "\\{\n");
for m in required.iter() {
write!(w, " ");
render_method(w, m.item(), true);
write!(w, ";\n");
}
if required.len() > 0 && provided.len() > 0 {
w.write("\n".as_bytes());
}
for m in provided.iter() {
write!(w, " ");
render_method(w, m.item(), true);
write!(w, " \\{ ... \\}\n");
}
write!(w, "\\}");
}
write!(w, "</pre>");
// Trait documentation
document(w, it);
fn meth(w: &mut Writer, m: &clean::TraitMethod) {
write!(w, "<h3 id='{}.{}' class='method'><code>",
shortty(m.item()),
*m.item().name.get_ref());
render_method(w, m.item(), false);
write!(w, "</code></h3>");
document(w, m.item());
}
// Output the documentation for each function individually
if required.len() > 0 {
write!(w, "
<h2 id='required-methods'>Required Methods</h2>
<div class='methods'>
");
for m in required.iter() {
meth(w, *m);
}
write!(w, "</div>");
}
if provided.len() > 0 {
write!(w, "
<h2 id='provided-methods'>Provided Methods</h2>
<div class='methods'>
");
for m in provided.iter() {
meth(w, *m);
}
write!(w, "</div>");
}
do local_data::get(cache_key) |cache| {
do cache.unwrap().read |cache| {
match cache.implementors.find(&it.id) {
Some(implementors) => {
write!(w, "
<h2 id='implementors'>Implementors</h2>
<ul class='item-list'>
");
for i in implementors.iter() {
match *i {
PathType(ref ty) => {
write!(w, "<li><code>{}</code></li>", *ty);
}
OtherType(ref generics, ref trait_, ref for_) => {
write!(w, "<li><code>impl{} {} for {}</code></li>",
*generics, *trait_, *for_);
}
}
}
write!(w, "</ul>");
}
None => {}
}
}
}
}
fn render_method(w: &mut Writer, meth: &clean::Item, withlink: bool) {
fn fun(w: &mut Writer, it: &clean::Item, purity: ast::purity,
g: &clean::Generics, selfty: &clean::SelfTy, d: &clean::FnDecl,
withlink: bool) {
write!(w, "{}fn {withlink, select,
true{<a href='\\#{ty}.{name}'
class='fnname'>{name}</a>}
other{<span class='fnname'>{name}</span>}
}{generics}{decl}",
match purity {
ast::unsafe_fn => "unsafe ",
_ => "",
},
ty = shortty(it),
name = it.name.get_ref().as_slice(),
generics = *g,
decl = Method(selfty, d),
withlink = if withlink {"true"} else {"false"});
}
match meth.inner {
clean::TyMethodItem(ref m) => {
fun(w, meth, m.purity, &m.generics, &m.self_, &m.decl, withlink);
}
clean::MethodItem(ref m) => {
fun(w, meth, m.purity, &m.generics, &m.self_, &m.decl, withlink);
}
_ => unreachable!()
}
}
fn item_struct(w: &mut Writer, it: &clean::Item, s: &clean::Struct) {
write!(w, "<pre class='struct'>");
render_struct(w, it, Some(&s.generics), s.struct_type, s.fields,
s.fields_stripped, "", true);
write!(w, "</pre>");
document(w, it);
match s.struct_type {
doctree::Plain => {
write!(w, "<h2 class='fields'>Fields</h2>\n<table>");
for field in s.fields.iter() {
write!(w, "<tr><td id='structfield.{name}'>\
<code>{name}</code></td><td>",
name = field.name.get_ref().as_slice());
document(w, field);
write!(w, "</td></tr>");
}
write!(w, "</table>");
}
_ => {}
}
render_methods(w, it);
}
fn item_enum(w: &mut Writer, it: &clean::Item, e: &clean::Enum) {
write!(w, "<pre class='enum'>{}enum {}{}",
VisSpace(it.visibility),
it.name.get_ref().as_slice(),
e.generics);
if e.variants.len() == 0 {
write!(w, " \\{\\}");
} else {
write!(w, " \\{\n");
for v in e.variants.iter() {
write!(w, " ");
let name = v.name.get_ref().as_slice();
match v.inner {
clean::VariantItem(ref var) => {
match var.kind {
clean::CLikeVariant => write!(w, "{}", name),
clean::TupleVariant(ref tys) => {
write!(w, "{}(", name);
for (i, ty) in tys.iter().enumerate() {
if i > 0 { write!(w, ", ") }
write!(w, "{}", *ty);
}
write!(w, ")");
}
clean::StructVariant(ref s) => {
render_struct(w, v, None, s.struct_type, s.fields,
s.fields_stripped, " ", false);
}
}
}
_ => unreachable!()
}
write!(w, ",\n");
}
if e.variants_stripped {
write!(w, " // some variants omitted\n");
}
write!(w, "\\}");
}
write!(w, "</pre>");
document(w, it);
if e.variants.len() > 0 {
write!(w, "<h2 class='variants'>Variants</h2>\n<table>");
for variant in e.variants.iter() {
write!(w, "<tr><td id='variant.{name}'><code>{name}</code></td><td>",
name = variant.name.get_ref().as_slice());
document(w, variant);
match variant.inner {
clean::VariantItem(ref var) => {
match var.kind {
clean::StructVariant(ref s) => {
write!(w, "<h3 class='fields'>Fields</h3>\n<table>");
for field in s.fields.iter() {
write!(w, "<tr><td id='variant.{v}.field.{f}'>\
<code>{f}</code></td><td>",
v = variant.name.get_ref().as_slice(),
f = field.name.get_ref().as_slice());
document(w, field);
write!(w, "</td></tr>");
}
write!(w, "</table>");
}
_ => ()
}
}
_ => ()
}
write!(w, "</td></tr>");
}
write!(w, "</table>");
}
render_methods(w, it);
}
fn render_struct(w: &mut Writer, it: &clean::Item,
g: Option<&clean::Generics>,
ty: doctree::StructType,
fields: &[clean::Item],
fields_stripped: bool,
tab: &str,
structhead: bool) {
write!(w, "{}{}{}",
VisSpace(it.visibility),
if structhead {"struct "} else {""},
it.name.get_ref().as_slice());
match g {
Some(g) => write!(w, "{}", *g),
None => {}
}
match ty {
doctree::Plain => {
write!(w, " \\{\n{}", tab);
for field in fields.iter() {
match field.inner {
clean::StructFieldItem(ref ty) => {
write!(w, " {}{}: {},\n{}",
VisSpace(field.visibility),
field.name.get_ref().as_slice(),
ty.type_,
tab);
}
_ => unreachable!()
}
}
if fields_stripped {
write!(w, " // some fields omitted\n{}", tab);
}
write!(w, "\\}");
}
doctree::Tuple | doctree::Newtype => {
write!(w, "(");
for (i, field) in fields.iter().enumerate() {
if i > 0 { write!(w, ", ") }
match field.inner {
clean::StructFieldItem(ref field) => {
write!(w, "{}", field.type_);
}
_ => unreachable!()
}
}
write!(w, ");");
}
doctree::Unit => { write!(w, ";"); }
}
}
fn render_methods(w: &mut Writer, it: &clean::Item) {
do local_data::get(cache_key) |cache| {
let cache = cache.unwrap();
do cache.read |c| {
match c.impls.find(&it.id) {
Some(v) => {
let mut non_trait = v.iter().filter(|p| {
p.n0_ref().trait_.is_none()
});
let non_trait = non_trait.to_owned_vec();
let mut traits = v.iter().filter(|p| {
p.n0_ref().trait_.is_some()
});
let traits = traits.to_owned_vec();
if non_trait.len() > 0 {
write!(w, "<h2 id='methods'>Methods</h2>");
for &(ref i, ref dox) in non_trait.move_iter() {
render_impl(w, i, dox);
}
}
if traits.len() > 0 {
write!(w, "<h2 id='implementations'>Trait \
Implementations</h2>");
for &(ref i, ref dox) in traits.move_iter() {
render_impl(w, i, dox);
}
}
}
None => {}
}
}
}
}
fn render_impl(w: &mut Writer, i: &clean::Impl, dox: &Option<~str>) {
write!(w, "<h3 class='impl'><code>impl{} ", i.generics);
let trait_id = match i.trait_ {
Some(ref ty) => {
write!(w, "{} for ", *ty);
match *ty {
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
clean::ResolvedPath { id, _ } => Some(id),
_ => None,
}
}
None => None
};
write!(w, "{}</code></h3>", i.for_);
match *dox {
Some(ref dox) => {
write!(w, "<div class='docblock'>{}</div>",
Markdown(dox.as_slice()));
}
None => {}
}
fn docmeth(w: &mut Writer, item: &clean::Item) -> bool {
write!(w, "<h4 id='method.{}' class='method'><code>",
*item.name.get_ref());
render_method(w, item, false);
write!(w, "</code></h4>\n");
match item.doc_value() {
Some(s) => {
write!(w, "<div class='docblock'>{}</div>", Markdown(s));
true
}
None => false
}
}
write!(w, "<div class='methods'>");
for meth in i.methods.iter() {
if docmeth(w, meth) {
continue
}
// No documentation? Attempt to slurp in the trait's documentation
let trait_id = match trait_id {
None => continue,
rustdoc: Generate hyperlinks between crates The general idea of hyperlinking between crates is that it should require as little configuration as possible, if any at all. In this vein, there are two separate ways to generate hyperlinks between crates: 1. When you're generating documentation for a crate 'foo' into folder 'doc', then if foo's external crate dependencies already have documented in the folder 'doc', then hyperlinks will be generated. This will work because all documentation is in the same folder, allowing links to work seamlessly both on the web and on the local filesystem browser. The rationale for this use case is a package with multiple libraries/crates that all want to link to one another, and you don't want to have to deal with going to the web. In theory this could be extended to have a RUST_PATH-style searching situtation, but I'm not sure that it would work seamlessly on the web as it does on the local filesystem, so I'm not attempting to explore this case in this pull request. I believe to fully realize this potential rustdoc would have to be acting as a server instead of a static site generator. 2. One of foo's external dependencies has a #[doc(html_root_url = "...")] attribute. This means that all hyperlinks to the dependency will be rooted at this url. This use case encompasses all packages using libstd/libextra. These two crates now have this attribute encoded (currently at the /doc/master url) and will be read by anything which has a dependency on libstd/libextra. This should also work for arbitrary crates in the wild that have online documentation. I don't like how the version is hard-wired into the url, but I think that this may be a case-by-case thing which doesn't end up being too bad in the long run. Closes #9539
2013-10-02 15:39:32 -07:00
Some(id) => id,
};
do local_data::get(cache_key) |cache| {
do cache.unwrap().read |cache| {
match cache.traits.find(&trait_id) {
Some(t) => {
let name = meth.name.clone();
match t.methods.iter().find(|t| t.item().name == name) {
Some(method) => {
match method.item().doc_value() {
Some(s) => {
write!(w,
"<div class='docblock'>{}</div>",
Markdown(s));
}
None => {}
}
}
None => {}
}
}
None => {}
}
}
}
}
// If we've implemented a trait, then also emit documentation for all
// default methods which weren't overridden in the implementation block.
match trait_id {
None => {}
Some(id) => {
do local_data::get(cache_key) |cache| {
do cache.unwrap().read |cache| {
match cache.traits.find(&id) {
Some(t) => {
for method in t.methods.iter() {
let n = method.item().name.clone();
match i.methods.iter().find(|m| m.name == n) {
Some(*) => continue,
None => {}
}
docmeth(w, method.item());
}
}
None => {}
}
}
}
}
}
write!(w, "</div>");
}
fn item_typedef(w: &mut Writer, it: &clean::Item, t: &clean::Typedef) {
write!(w, "<pre class='typedef'>type {}{} = {};</pre>",
it.name.get_ref().as_slice(),
t.generics,
t.type_);
document(w, it);
}
impl<'self> fmt::Default for Sidebar<'self> {
fn fmt(s: &Sidebar<'self>, fmt: &mut fmt::Formatter) {
let cx = s.cx;
let it = s.item;
write!(fmt.buf, "<p class='location'>");
let len = cx.current.len() - if it.is_mod() {1} else {0};
for (i, name) in cx.current.iter().take(len).enumerate() {
if i > 0 { write!(fmt.buf, "&\\#8203;::") }
write!(fmt.buf, "<a href='{}index.html'>{}</a>",
cx.root_path.slice_to((cx.current.len() - i - 1) * 3), *name);
}
write!(fmt.buf, "</p>");
fn block(w: &mut Writer, short: &str, longty: &str,
cur: &clean::Item, cx: &Context) {
let items = match cx.sidebar.find_equiv(&short) {
Some(items) => items.as_slice(),
None => return
};
write!(w, "<div class='block {}'><h2>{}</h2>", short, longty);
for item in items.iter() {
let class = if cur.name.get_ref() == item &&
short == shortty(cur) { "current" } else { "" };
write!(w, "<a class='{ty} {class}' href='{curty, select,
mod{../}
other{}
}{tysel, select,
mod{{name}/index.html}
other{#.{name}.html}
}'>{name}</a><br/>",
ty = short,
tysel = short,
class = class,
curty = shortty(cur),
name = item.as_slice());
}
write!(w, "</div>");
}
block(fmt.buf, "mod", "Modules", it, cx);
block(fmt.buf, "struct", "Structs", it, cx);
block(fmt.buf, "enum", "Enums", it, cx);
block(fmt.buf, "trait", "Traits", it, cx);
block(fmt.buf, "fn", "Functions", it, cx);
}
}
fn build_sidebar(m: &clean::Module) -> HashMap<~str, ~[~str]> {
let mut map = HashMap::new();
for item in m.items.iter() {
let short = shortty(item);
let myname = match item.name {
None => continue,
Some(ref s) => s.to_owned(),
};
let v = map.find_or_insert_with(short.to_owned(), |_| ~[]);
v.push(myname);
}
for (_, items) in map.mut_iter() {
sort::quick_sort(*items, |i1, i2| i1 < i2);
}
return map;
}
impl<'self> fmt::Default for Source<'self> {
fn fmt(s: &Source<'self>, fmt: &mut fmt::Formatter) {
let lines = s.line_iter().len();
let mut cols = 0;
let mut tmp = lines;
while tmp > 0 {
cols += 1;
tmp /= 10;
}
write!(fmt.buf, "<pre class='line-numbers'>");
for i in range(1, lines + 1) {
write!(fmt.buf, "<span id='{0:u}'>{0:1$u}</span>\n", i, cols);
}
write!(fmt.buf, "</pre>");
write!(fmt.buf, "<pre class='rust'>");
write!(fmt.buf, "{}", Escape(s.as_slice()));
write!(fmt.buf, "</pre>");
}
}