rust/src/libsyntax/lib.rs

190 lines
4.4 KiB
Rust
Raw Normal View History

2014-06-09 20:12:30 +00:00
//! The Rust parser and macro expander.
//!
//! # Note
//!
//! This API is completely unstable and subject to change.
2013-03-29 19:51:10 +00:00
#![doc(html_root_url = "https://doc.rust-lang.org/nightly/",
2015-11-03 16:34:11 +00:00
test(attr(deny(warnings))))]
#![feature(bind_by_move_pattern_guards)]
#![feature(box_syntax)]
#![feature(const_fn)]
#![feature(const_transmute)]
#![feature(crate_visibility_modifier)]
#![feature(label_break_value)]
#![feature(mem_take)]
2019-02-10 07:13:30 +00:00
#![feature(nll)]
2019-07-18 18:02:34 +00:00
#![feature(proc_macro_diagnostic)]
#![feature(proc_macro_internals)]
#![feature(proc_macro_span)]
#![feature(rustc_diagnostic_macros)]
2018-08-20 00:51:02 +00:00
#![feature(try_trait)]
#![feature(unicode_internals)]
2018-03-03 05:24:49 +00:00
#![recursion_limit="256"]
2019-07-18 18:02:34 +00:00
extern crate proc_macro;
pub use errors;
use rustc_data_structures::sync::Lock;
Merge indexed_set.rs into bitvec.rs, and rename it bit_set.rs. Currently we have two files implementing bitsets (and 2D bit matrices). This commit combines them into one, taking the best features from each. This involves renaming a lot of things. The high level changes are as follows. - bitvec.rs --> bit_set.rs - indexed_set.rs --> (removed) - BitArray + IdxSet --> BitSet (merged, see below) - BitVector --> GrowableBitSet - {,Sparse,Hybrid}IdxSet --> {,Sparse,Hybrid}BitSet - BitMatrix --> BitMatrix - SparseBitMatrix --> SparseBitMatrix The changes within the bitset types themselves are as follows. ``` OLD OLD NEW BitArray<C> IdxSet<T> BitSet<T> -------- ------ ------ grow - grow new - (remove) new_empty new_empty new_empty new_filled new_filled new_filled - to_hybrid to_hybrid clear clear clear set_up_to set_up_to set_up_to clear_above - clear_above count - count contains(T) contains(&T) contains(T) contains_all - superset is_empty - is_empty insert(T) add(&T) insert(T) insert_all - insert_all() remove(T) remove(&T) remove(T) words words words words_mut words_mut words_mut - overwrite overwrite merge union union - subtract subtract - intersect intersect iter iter iter ``` In general, when choosing names I went with: - names that are more obvious (e.g. `BitSet` over `IdxSet`). - names that are more like the Rust libraries (e.g. `T` over `C`, `insert` over `add`); - names that are more set-like (e.g. `union` over `merge`, `superset` over `contains_all`, `domain_size` over `num_bits`). Also, using `T` for index arguments seems more sensible than `&T` -- even though the latter is standard in Rust collection types -- because indices are always copyable. It also results in fewer `&` and `*` sigils in practice.
2018-09-14 05:07:25 +00:00
use rustc_data_structures::bit_set::GrowableBitSet;
pub use rustc_data_structures::thin_vec::ThinVec;
use ast::AttrId;
use syntax_pos::edition::Edition;
2019-05-22 05:17:53 +00:00
const MACRO_ARGUMENTS: Option<&'static str> = Some("macro arguments");
2015-12-20 21:00:43 +00:00
// A variant of 'try!' that panics on an Err. This is used as a crutch on the
// way towards a non-panic!-prone parser. It should be used for fatal parsing
// errors; eventually we plan to convert all code using panictry to just use
// normal try.
#[macro_export]
macro_rules! panictry {
($e:expr) => ({
use std::result::Result::{Ok, Err};
use errors::FatalError;
match $e {
Ok(e) => e,
2015-12-20 21:00:43 +00:00
Err(mut e) => {
e.emit();
FatalError.raise()
2015-12-20 21:00:43 +00:00
}
}
})
}
// A variant of 'panictry!' that works on a Vec<Diagnostic> instead of a single DiagnosticBuilder.
macro_rules! panictry_buffer {
($handler:expr, $e:expr) => ({
use std::result::Result::{Ok, Err};
use errors::{FatalError, DiagnosticBuilder};
match $e {
Ok(e) => e,
Err(errs) => {
for e in errs {
DiagnosticBuilder::new_diagnostic($handler, e).emit();
}
FatalError.raise()
}
}
})
}
#[macro_export]
macro_rules! unwrap_or {
($opt:expr, $default:expr) => {
match $opt {
Some(x) => x,
None => $default,
}
}
}
2018-04-25 22:49:52 +00:00
pub struct Globals {
Merge indexed_set.rs into bitvec.rs, and rename it bit_set.rs. Currently we have two files implementing bitsets (and 2D bit matrices). This commit combines them into one, taking the best features from each. This involves renaming a lot of things. The high level changes are as follows. - bitvec.rs --> bit_set.rs - indexed_set.rs --> (removed) - BitArray + IdxSet --> BitSet (merged, see below) - BitVector --> GrowableBitSet - {,Sparse,Hybrid}IdxSet --> {,Sparse,Hybrid}BitSet - BitMatrix --> BitMatrix - SparseBitMatrix --> SparseBitMatrix The changes within the bitset types themselves are as follows. ``` OLD OLD NEW BitArray<C> IdxSet<T> BitSet<T> -------- ------ ------ grow - grow new - (remove) new_empty new_empty new_empty new_filled new_filled new_filled - to_hybrid to_hybrid clear clear clear set_up_to set_up_to set_up_to clear_above - clear_above count - count contains(T) contains(&T) contains(T) contains_all - superset is_empty - is_empty insert(T) add(&T) insert(T) insert_all - insert_all() remove(T) remove(&T) remove(T) words words words words_mut words_mut words_mut - overwrite overwrite merge union union - subtract subtract - intersect intersect iter iter iter ``` In general, when choosing names I went with: - names that are more obvious (e.g. `BitSet` over `IdxSet`). - names that are more like the Rust libraries (e.g. `T` over `C`, `insert` over `add`); - names that are more set-like (e.g. `union` over `merge`, `superset` over `contains_all`, `domain_size` over `num_bits`). Also, using `T` for index arguments seems more sensible than `&T` -- even though the latter is standard in Rust collection types -- because indices are always copyable. It also results in fewer `&` and `*` sigils in practice.
2018-09-14 05:07:25 +00:00
used_attrs: Lock<GrowableBitSet<AttrId>>,
known_attrs: Lock<GrowableBitSet<AttrId>>,
syntax_pos_globals: syntax_pos::Globals,
}
impl Globals {
fn new(edition: Edition) -> Globals {
Globals {
// We have no idea how many attributes their will be, so just
// initiate the vectors with 0 bits. We'll grow them as necessary.
Merge indexed_set.rs into bitvec.rs, and rename it bit_set.rs. Currently we have two files implementing bitsets (and 2D bit matrices). This commit combines them into one, taking the best features from each. This involves renaming a lot of things. The high level changes are as follows. - bitvec.rs --> bit_set.rs - indexed_set.rs --> (removed) - BitArray + IdxSet --> BitSet (merged, see below) - BitVector --> GrowableBitSet - {,Sparse,Hybrid}IdxSet --> {,Sparse,Hybrid}BitSet - BitMatrix --> BitMatrix - SparseBitMatrix --> SparseBitMatrix The changes within the bitset types themselves are as follows. ``` OLD OLD NEW BitArray<C> IdxSet<T> BitSet<T> -------- ------ ------ grow - grow new - (remove) new_empty new_empty new_empty new_filled new_filled new_filled - to_hybrid to_hybrid clear clear clear set_up_to set_up_to set_up_to clear_above - clear_above count - count contains(T) contains(&T) contains(T) contains_all - superset is_empty - is_empty insert(T) add(&T) insert(T) insert_all - insert_all() remove(T) remove(&T) remove(T) words words words words_mut words_mut words_mut - overwrite overwrite merge union union - subtract subtract - intersect intersect iter iter iter ``` In general, when choosing names I went with: - names that are more obvious (e.g. `BitSet` over `IdxSet`). - names that are more like the Rust libraries (e.g. `T` over `C`, `insert` over `add`); - names that are more set-like (e.g. `union` over `merge`, `superset` over `contains_all`, `domain_size` over `num_bits`). Also, using `T` for index arguments seems more sensible than `&T` -- even though the latter is standard in Rust collection types -- because indices are always copyable. It also results in fewer `&` and `*` sigils in practice.
2018-09-14 05:07:25 +00:00
used_attrs: Lock::new(GrowableBitSet::new_empty()),
known_attrs: Lock::new(GrowableBitSet::new_empty()),
syntax_pos_globals: syntax_pos::Globals::new(edition),
}
}
}
pub fn with_globals<F, R>(edition: Edition, f: F) -> R
where F: FnOnce() -> R
{
let globals = Globals::new(edition);
GLOBALS.set(&globals, || {
syntax_pos::GLOBALS.set(&globals.syntax_pos_globals, f)
})
}
pub fn with_default_globals<F, R>(f: F) -> R
where F: FnOnce() -> R
{
with_globals(edition::DEFAULT_EDITION, f)
}
2019-02-06 17:33:01 +00:00
scoped_tls::scoped_thread_local!(pub static GLOBALS: Globals);
2016-06-28 17:40:40 +00:00
#[macro_use]
pub mod diagnostics {
#[macro_use]
pub mod macros;
pub mod plugin;
pub mod metadata;
}
// N.B., this module needs to be declared first so diagnostics are
2016-06-28 17:40:40 +00:00
// registered before they are used.
2019-04-17 21:27:30 +00:00
pub mod error_codes;
2016-06-28 17:40:40 +00:00
2013-04-03 16:41:40 +00:00
pub mod util {
pub mod lev_distance;
pub mod node_count;
pub mod parser;
2013-05-17 00:41:47 +00:00
#[cfg(test)]
pub mod parser_testing;
Overhaul `syntax::fold::Folder`. This commit changes `syntax::fold::Folder` from a functional style (where most methods take a `T` and produce a new `T`) to a more imperative style (where most methods take and modify a `&mut T`), and renames it `syntax::mut_visit::MutVisitor`. The first benefit is speed. The functional style does not require any reallocations, due to the use of `P::map` and `MoveMap::move_{,flat_}map`. However, every field in the AST must be overwritten; even those fields that are unchanged are overwritten with the same value. This causes a lot of unnecessary memory writes. The imperative style reduces instruction counts by 1--3% across a wide range of workloads, particularly incremental workloads. The second benefit is conciseness; the imperative style is usually more concise. E.g. compare the old functional style: ``` fn fold_abc(&mut self, abc: ABC) { ABC { a: fold_a(abc.a), b: fold_b(abc.b), c: abc.c, } } ``` with the imperative style: ``` fn visit_abc(&mut self, ABC { a, b, c: _ }: &mut ABC) { visit_a(a); visit_b(b); } ``` (The reductions get larger in more complex examples.) Overall, the patch removes over 200 lines of code -- even though the new code has more comments -- and a lot of the remaining lines have fewer characters. Some notes: - The old style used methods called `fold_*`. The new style mostly uses methods called `visit_*`, but there are a few methods that map a `T` to something other than a `T`, which are called `flat_map_*` (`T` maps to multiple `T`s) or `filter_map_*` (`T` maps to 0 or 1 `T`s). - `move_map.rs`/`MoveMap`/`move_map`/`move_flat_map` are renamed `map_in_place.rs`/`MapInPlace`/`map_in_place`/`flat_map_in_place` to reflect their slightly changed signatures. - Although this commit renames the `fold` module as `mut_visit`, it keeps it in the `fold.rs` file, so as not to confuse git. The next commit will rename the file.
2019-02-05 04:20:55 +00:00
pub mod map_in_place;
2013-04-03 16:41:40 +00:00
}
pub mod json;
pub mod ast;
pub mod attr;
2018-08-18 10:14:03 +00:00
pub mod source_map;
#[macro_use]
2014-07-25 02:44:24 +00:00
pub mod config;
pub mod entry;
2014-09-11 00:55:42 +00:00
pub mod feature_gate;
2019-02-05 22:10:05 +00:00
pub mod mut_visit;
pub mod parse;
pub mod ptr;
2014-07-25 02:44:24 +00:00
pub mod show_span;
pub use syntax_pos::edition;
2017-03-17 04:04:41 +00:00
pub use syntax_pos::symbol;
pub mod tokenstream;
pub mod visit;
pub mod print {
pub mod pp;
pub mod pprust;
mod helpers;
2012-03-29 20:48:05 +00:00
}
pub mod ext {
2019-07-18 18:02:34 +00:00
mod placeholders;
mod proc_macro_server;
pub use syntax_pos::hygiene;
pub mod allocator;
pub mod base;
pub mod build;
pub mod expand;
pub mod proc_macro;
pub mod tt {
pub mod transcribe;
pub mod macro_check;
pub mod macro_parser;
pub mod macro_rules;
pub mod quoted;
}
2012-03-29 20:48:05 +00:00
}
2016-06-28 17:40:40 +00:00
pub mod early_buffered_lints;
#[cfg(test)]
mod test_snippet;
__build_diagnostic_array! { libsyntax, DIAGNOSTICS }