Auto merge of #34860 - jseyfried:encapsulate_hygiene, r=nrc

Clean up and encapsulate `syntax::ext::mtwt`, rename `mtwt` to `hygiene`

r? @nrc

Author: bors

src/librustc/session/config.rs

@@ -735,8 +735,8 @@ options! {DebuggingOptions, DebuggingSetter, basic_debugging_options,
           "for every macro invocation, print its name and arguments"),
     enable_nonzeroing_move_hints: bool = (false, parse_bool,
           "force nonzeroing move optimization on"),
-    keep_mtwt_tables: bool = (false, parse_bool,
-          "don't clear the resolution tables after analysis"),
+    keep_hygiene_data: bool = (false, parse_bool,
+          "don't clear the hygiene data after analysis"),
     keep_ast: bool = (false, parse_bool,
           "keep the AST after lowering it to HIR"),
     show_span: Option<String> = (None, parse_opt_string,

src/librustc_driver/driver.rs

@@ -236,8 +236,8 @@ pub fn compile_input(sess: &Session,
     Ok(())
 }
 
-fn keep_mtwt_tables(sess: &Session) -> bool {
-    sess.opts.debugging_opts.keep_mtwt_tables
+fn keep_hygiene_data(sess: &Session) -> bool {
+    sess.opts.debugging_opts.keep_hygiene_data
 }
 
 fn keep_ast(sess: &Session) -> bool {
@@ -479,9 +479,8 @@ pub fn phase_1_parse_input<'a>(sess: &'a Session,
                                input: &Input)
                                -> PResult<'a, ast::Crate> {
     // These may be left in an incoherent state after a previous compile.
-    // `clear_tables` and `clear_ident_interner` can be used to free
-    // memory, but they do not restore the initial state.
-    syntax::ext::mtwt::reset_tables();
+    syntax::ext::hygiene::reset_hygiene_data();
+    // `clear_ident_interner` can be used to free memory, but it does not restore the initial state.
     token::reset_ident_interner();
     let continue_after_error = sess.opts.continue_parse_after_error;
     sess.diagnostic().set_continue_after_error(continue_after_error);
@@ -761,9 +760,9 @@ pub fn phase_2_configure_and_expand<'a, F>(sess: &Session,
         hir_map::Forest::new(lower_crate(sess, &krate, &mut resolver), &sess.dep_graph)
     });
 
-    // Discard MTWT tables that aren't required past lowering to HIR.
-    if !keep_mtwt_tables(sess) {
-        syntax::ext::mtwt::clear_tables();
+    // Discard hygiene data, which isn't required past lowering to HIR.
+    if !keep_hygiene_data(sess) {
+        syntax::ext::hygiene::reset_hygiene_data();
     }
 
     Ok(ExpansionResult {

src/librustc_driver/pretty.rs

@@ -456,7 +456,7 @@ impl<'ast> pprust::PpAnn for HygieneAnnotation<'ast> {
                 pp::space(&mut s.s)?;
                 // FIXME #16420: this doesn't display the connections
                 // between syntax contexts
-                s.synth_comment(format!("{}#{}", nm, ctxt.0))
+                s.synth_comment(format!("{}{:?}", nm, ctxt))
             }
             pprust::NodeName(&ast::Name(nm)) => {
                 pp::space(&mut s.s)?;

src/librustc_resolve/assign_ids.rs

@@ -11,7 +11,7 @@
 use Resolver;
 use rustc::session::Session;
 use syntax::ast;
-use syntax::ext::mtwt;
+use syntax::ext::hygiene::Mark;
 use syntax::fold::{self, Folder};
 use syntax::ptr::P;
 use syntax::util::move_map::MoveMap;
@@ -31,7 +31,7 @@ impl<'a> Resolver<'a> {
 
 struct NodeIdAssigner<'a> {
     sess: &'a Session,
-    macros_at_scope: &'a mut HashMap<ast::NodeId, Vec<ast::Mrk>>,
+    macros_at_scope: &'a mut HashMap<ast::NodeId, Vec<Mark>>,
 }
 
 impl<'a> Folder for NodeIdAssigner<'a> {
@@ -49,7 +49,7 @@ impl<'a> Folder for NodeIdAssigner<'a> {
             block.stmts = block.stmts.move_flat_map(|stmt| {
                 if let ast::StmtKind::Item(ref item) = stmt.node {
                     if let ast::ItemKind::Mac(..) = item.node {
-                        macros.push(mtwt::outer_mark(item.ident.ctxt));
+                        macros.push(item.ident.ctxt.data().outer_mark);
                         return None;
                     }
                 }

src/librustc_resolve/lib.rs

@@ -53,7 +53,7 @@ use rustc::ty::subst::{ParamSpace, FnSpace, TypeSpace};
 use rustc::hir::{Freevar, FreevarMap, TraitCandidate, TraitMap, GlobMap};
 use rustc::util::nodemap::{NodeMap, NodeSet, FnvHashMap, FnvHashSet};
 
-use syntax::ext::mtwt;
+use syntax::ext::hygiene::Mark;
 use syntax::ast::{self, FloatTy};
 use syntax::ast::{CRATE_NODE_ID, Name, NodeId, CrateNum, IntTy, UintTy};
 use syntax::parse::token::{self, keywords};
@@ -654,7 +654,7 @@ enum RibKind<'a> {
     ModuleRibKind(Module<'a>),
 
     // We passed through a `macro_rules!` statement with the given expansion
-    MacroDefinition(ast::Mrk),
+    MacroDefinition(Mark),
 }
 
 #[derive(Copy, Clone)]
@@ -933,7 +933,7 @@ pub struct Resolver<'a> {
 
     // Maps the node id of a statement to the expansions of the `macro_rules!`s
     // immediately above the statement (if appropriate).
-    macros_at_scope: HashMap<NodeId, Vec<ast::Mrk>>,
+    macros_at_scope: HashMap<NodeId, Vec<Mark>>,
 
     graph_root: Module<'a>,
 
@@ -1434,10 +1434,9 @@ impl<'a> Resolver<'a> {
             if let MacroDefinition(mac) = self.get_ribs(ns)[i].kind {
                 // If an invocation of this macro created `ident`, give up on `ident`
                 // and switch to `ident`'s source from the macro definition.
-                if let Some((source_ident, source_macro)) = mtwt::source(ident) {
-                    if mac == source_macro {
-                        ident = source_ident;
-                    }
+                let (source_ctxt, source_macro) = ident.ctxt.source();
+                if source_macro == mac {
+                    ident.ctxt = source_ctxt;
                 }
             }
         }
@@ -1585,10 +1584,9 @@ impl<'a> Resolver<'a> {
                 MacroDefinition(mac) => {
                     // If an invocation of this macro created `ident`, give up on `ident`
                     // and switch to `ident`'s source from the macro definition.
-                    if let Some((source_ident, source_macro)) = mtwt::source(ident) {
-                        if mac == source_macro {
-                            ident = source_ident;
-                        }
+                    let (source_ctxt, source_macro) = ident.ctxt.source();
+                    if source_macro == mac {
+                        ident.ctxt = source_ctxt;
                     }
                 }
                 _ => {

src/libsyntax/ast.rs

@@ -19,6 +19,7 @@ pub use util::ThinVec;
 use syntax_pos::{mk_sp, Span, DUMMY_SP, ExpnId};
 use codemap::{respan, Spanned};
 use abi::Abi;
+use ext::hygiene::SyntaxContext;
 use parse::token::{self, keywords, InternedString};
 use print::pprust;
 use ptr::P;
@@ -33,15 +34,6 @@ use serialize::{Encodable, Decodable, Encoder, Decoder};
 #[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
 pub struct Name(pub u32);
 
-/// A SyntaxContext represents a chain of macro-expandings
-/// and renamings. Each macro expansion corresponds to
-/// a fresh u32. This u32 is a reference to a table stored
-/// in thread-local storage.
-/// The special value EMPTY_CTXT is used to indicate an empty
-/// syntax context.
-#[derive(Clone, Copy, PartialEq, Eq, Hash, Debug, RustcEncodable, RustcDecodable)]
-pub struct SyntaxContext(pub u32);
-
 /// An identifier contains a Name (index into the interner
 /// table) and a SyntaxContext to track renaming and
 /// macro expansion per Flatt et al., "Macros That Work Together"
@@ -81,20 +73,15 @@ impl Decodable for Name {
     }
 }
 
-pub const EMPTY_CTXT : SyntaxContext = SyntaxContext(0);
-
 impl Ident {
-    pub fn new(name: Name, ctxt: SyntaxContext) -> Ident {
-        Ident {name: name, ctxt: ctxt}
-    }
     pub const fn with_empty_ctxt(name: Name) -> Ident {
-        Ident {name: name, ctxt: EMPTY_CTXT}
+        Ident { name: name, ctxt: SyntaxContext::empty() }
     }
 }
 
 impl fmt::Debug for Ident {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(f, "{}#{}", self.name, self.ctxt.0)
+        write!(f, "{}{:?}", self.name, self.ctxt)
     }
 }
 
@@ -116,9 +103,6 @@ impl Decodable for Ident {
     }
 }
 
-/// A mark represents a unique id associated with a macro expansion
-pub type Mrk = u32;
-
 #[derive(Clone, PartialEq, Eq, RustcEncodable, RustcDecodable, Hash, Copy)]
 pub struct Lifetime {
     pub id: NodeId,

src/libsyntax/ext/expand.rs

@@ -9,20 +9,19 @@
 // except according to those terms.
 
 use ast::{Block, Crate, Ident, Mac_, Name, PatKind};
-use ast::{MacStmtStyle, Mrk, Stmt, StmtKind, ItemKind};
+use ast::{MacStmtStyle, Stmt, StmtKind, ItemKind};
 use ast;
-use attr::HasAttrs;
-use ext::mtwt;
-use attr;
+use ext::hygiene::Mark;
+use attr::{self, HasAttrs};
 use attr::AttrMetaMethods;
-use codemap::{dummy_spanned, Spanned, ExpnInfo, NameAndSpan, MacroBang, MacroAttribute};
+use codemap::{dummy_spanned, ExpnInfo, NameAndSpan, MacroBang, MacroAttribute};
 use syntax_pos::{self, Span, ExpnId};
 use config::StripUnconfigured;
 use ext::base::*;
 use feature_gate::{self, Features};
 use fold;
 use fold::*;
-use parse::token::{fresh_mark, intern, keywords};
+use parse::token::{intern, keywords};
 use ptr::P;
 use tokenstream::TokenTree;
 use util::small_vector::SmallVector;
@@ -130,9 +129,9 @@ fn expand_mac_invoc<T>(mac: ast::Mac, ident: Option<Ident>, attrs: Vec<ast::Attr
     // It would almost certainly be cleaner to pass the whole macro invocation in,
     // rather than pulling it apart and marking the tts and the ctxt separately.
     let Mac_ { path, tts, .. } = mac.node;
-    let mark = fresh_mark();
+    let mark = Mark::fresh();
 
-    fn mac_result<'a>(path: &ast::Path, ident: Option<Ident>, tts: Vec<TokenTree>, mark: Mrk,
+    fn mac_result<'a>(path: &ast::Path, ident: Option<Ident>, tts: Vec<TokenTree>, mark: Mark,
                       attrs: Vec<ast::Attribute>, call_site: Span, fld: &'a mut MacroExpander)
                       -> Option<Box<MacResult + 'a>> {
         // Detect use of feature-gated or invalid attributes on macro invoations
@@ -708,30 +707,17 @@ pub fn expand_crate(mut cx: ExtCtxt,
     return (ret, cx.syntax_env.names);
 }
 
-// HYGIENIC CONTEXT EXTENSION:
-// all of these functions are for walking over
-// ASTs and making some change to the context of every
-// element that has one. a CtxtFn is a trait-ified
-// version of a closure in (SyntaxContext -> SyntaxContext).
-// the ones defined here include:
-// Marker - add a mark to a context
-
 // A Marker adds the given mark to the syntax context and
 // sets spans' `expn_id` to the given expn_id (unless it is `None`).
-struct Marker { mark: Mrk, expn_id: Option<ExpnId> }
+struct Marker { mark: Mark, expn_id: Option<ExpnId> }
 
 impl Folder for Marker {
-    fn fold_ident(&mut self, id: Ident) -> Ident {
-        ast::Ident::new(id.name, mtwt::apply_mark(self.mark, id.ctxt))
-    }
-    fn fold_mac(&mut self, Spanned {node, span}: ast::Mac) -> ast::Mac {
-        Spanned {
-            node: Mac_ {
-                path: self.fold_path(node.path),
-                tts: self.fold_tts(&node.tts),
-            },
-            span: self.new_span(span),
-        }
+    fn fold_ident(&mut self, mut ident: Ident) -> Ident {
+        ident.ctxt = ident.ctxt.apply_mark(self.mark);
+        ident
+    }
+    fn fold_mac(&mut self, mac: ast::Mac) -> ast::Mac {
+        noop_fold_mac(mac, self)
     }
 
     fn new_span(&mut self, mut span: Span) -> Span {
@@ -743,7 +729,7 @@ impl Folder for Marker {
 }
 
 // apply a given mark to the given token trees. Used prior to expansion of a macro.
-fn mark_tts(tts: &[TokenTree], m: Mrk) -> Vec<TokenTree> {
+fn mark_tts(tts: &[TokenTree], m: Mark) -> Vec<TokenTree> {
     noop_fold_tts(tts, &mut Marker{mark:m, expn_id: None})
 }
 

src/libsyntax/ext/hygiene.rs

@@ -0,0 +1,116 @@
+// Copyright 2012-2014 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.
+
+//! Machinery for hygienic macros, inspired by the MTWT[1] paper.
+//!
+//! [1] Matthew Flatt, Ryan Culpepper, David Darais, and Robert Bruce Findler.
+//! 2012. *Macros that work together: Compile-time bindings, partial expansion,
+//! and definition contexts*. J. Funct. Program. 22, 2 (March 2012), 181-216.
+//! DOI=10.1017/S0956796812000093 http://dx.doi.org/10.1017/S0956796812000093
+
+use std::cell::RefCell;
+use std::collections::HashMap;
+use std::fmt;
+
+/// A SyntaxContext represents a chain of macro expansions (represented by marks).
+#[derive(Clone, Copy, PartialEq, Eq, Hash, RustcEncodable, RustcDecodable, Default)]
+pub struct SyntaxContext(u32);
+
+#[derive(Copy, Clone)]
+pub struct SyntaxContextData {
+    pub outer_mark: Mark,
+    pub prev_ctxt: SyntaxContext,
+}
+
+/// A mark represents a unique id associated with a macro expansion.
+#[derive(Copy, Clone, PartialEq, Eq, Hash, Debug, Default)]
+pub struct Mark(u32);
+
+impl Mark {
+    pub fn fresh() -> Self {
+        HygieneData::with(|data| {
+            let next_mark = Mark(data.next_mark.0 + 1);
+            ::std::mem::replace(&mut data.next_mark, next_mark)
+        })
+    }
+}
+
+struct HygieneData {
+    syntax_contexts: Vec<SyntaxContextData>,
+    markings: HashMap<(SyntaxContext, Mark), SyntaxContext>,
+    next_mark: Mark,
+}
+
+impl HygieneData {
+    fn new() -> Self {
+        HygieneData {
+            syntax_contexts: vec![SyntaxContextData {
+                outer_mark: Mark(0), // the null mark
+                prev_ctxt: SyntaxContext(0), // the empty context
+            }],
+            markings: HashMap::new(),
+            next_mark: Mark(1),
+        }
+    }
+
+    fn with<T, F: FnOnce(&mut HygieneData) -> T>(f: F) -> T {
+        thread_local! {
+            static HYGIENE_DATA: RefCell<HygieneData> = RefCell::new(HygieneData::new());
+        }
+        HYGIENE_DATA.with(|data| f(&mut *data.borrow_mut()))
+    }
+}
+
+pub fn reset_hygiene_data() {
+    HygieneData::with(|data| *data = HygieneData::new())
+}
+
+impl SyntaxContext {
+    pub const fn empty() -> Self {
+        SyntaxContext(0)
+    }
+
+    pub fn data(self) -> SyntaxContextData {
+        HygieneData::with(|data| data.syntax_contexts[self.0 as usize])
+    }
+
+    /// Extend a syntax context with a given mark
+    pub fn apply_mark(self, mark: Mark) -> SyntaxContext {
+        // Applying the same mark twice is a no-op
+        let ctxt_data = self.data();
+        if mark == ctxt_data.outer_mark {
+            return ctxt_data.prev_ctxt;
+        }
+
+        HygieneData::with(|data| {
+            let syntax_contexts = &mut data.syntax_contexts;
+            *data.markings.entry((self, mark)).or_insert_with(|| {
+                syntax_contexts.push(SyntaxContextData {
+                    outer_mark: mark,
+                    prev_ctxt: self,
+                });
+                SyntaxContext(syntax_contexts.len() as u32 - 1)
+            })
+        })
+    }
+
+   /// If `ident` is macro expanded, return the source ident from the macro definition
+   /// and the mark of the expansion that created the macro definition.
+   pub fn source(self) -> (Self /* source context */, Mark /* source macro */) {
+        let macro_def_ctxt = self.data().prev_ctxt.data();
+        (macro_def_ctxt.prev_ctxt, macro_def_ctxt.outer_mark)
+   }
+}
+
+impl fmt::Debug for SyntaxContext {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "#{}", self.0)
+    }
+}