Only have one source of truth for keywords.

`rustc_symbol` is the source of truth for keywords.

rustdoc has its own implicit definition of keywords, via the
`is_doc_keyword`. It (presumably) intends to include all keywords, but
it omits `yeet`.

rustfmt has its own explicit list of Rust keywords. It also (presumably)
intends to include all keywords, but it omits `await`, `builtin`, `gen`,
`macro_rules`, `raw`, `reuse`, `safe`, and `yeet`. Also, it does linear
searches through this list, which is inefficient.

This commit fixes all of the above problems by introducing a new
predicate `is_any_keyword` in rustc and using it in rustdoc and rustfmt.
It documents that it's not the right predicate in most cases.

Author: nnethercote

compiler/rustc_ast/src/token.rs

@@ -903,7 +903,8 @@ impl Token {
         self.is_non_raw_ident_where(|id| id.name == kw)
     }
 
-    /// Returns `true` if the token is a given keyword, `kw` or if `case` is `Insensitive` and this token is an identifier equal to `kw` ignoring the case.
+    /// Returns `true` if the token is a given keyword, `kw` or if `case` is `Insensitive` and this
+    /// token is an identifier equal to `kw` ignoring the case.
     pub fn is_keyword_case(&self, kw: Symbol, case: Case) -> bool {
         self.is_keyword(kw)
             || (case == Case::Insensitive
@@ -916,6 +917,11 @@ impl Token {
         self.is_non_raw_ident_where(Ident::is_path_segment_keyword)
     }
 
+    /// Don't use this unless you're doing something very loose and heuristic-y.
+    pub fn is_any_keyword(&self) -> bool {
+        self.is_non_raw_ident_where(Ident::is_any_keyword)
+    }
+
     /// Returns true for reserved identifiers used internally for elided lifetimes,
     /// unnamed method parameters, crate root module, error recovery etc.
     pub fn is_special_ident(&self) -> bool {

compiler/rustc_parse/src/parser/diagnostics.rs

@@ -815,8 +815,8 @@ impl<'a> Parser<'a> {
 
             // Otherwise, check the previous token with all the keywords as possible candidates.
             // This handles code like `Struct Human;` and `While a < b {}`.
-            // We check the previous token only when the current token is an identifier to avoid false
-            // positives like suggesting keyword `for` for `extern crate foo {}`.
+            // We check the previous token only when the current token is an identifier to avoid
+            // false positives like suggesting keyword `for` for `extern crate foo {}`.
             if let Some(misspelled_kw) = find_similar_kw(prev_ident, &all_keywords) {
                 err.subdiagnostic(misspelled_kw);
                 // We don't want other suggestions to be added as they are most likely meaningless

compiler/rustc_span/src/symbol.rs

@@ -2589,6 +2589,11 @@ pub mod sym {
 }
 
 impl Symbol {
+    /// Don't use this unless you're doing something very loose and heuristic-y.
+    pub fn is_any_keyword(self) -> bool {
+        self >= kw::As && self <= kw::Yeet
+    }
+
     fn is_special(self) -> bool {
         self <= kw::Underscore
     }
@@ -2645,6 +2650,11 @@ impl Symbol {
 }
 
 impl Ident {
+    /// Don't use this unless you're doing something very loose and heuristic-y.
+    pub fn is_any_keyword(self) -> bool {
+        self.name.is_any_keyword()
+    }
+
     /// Returns `true` for reserved identifiers used internally for elided lifetimes,
     /// unnamed method parameters, crate root module, error recovery etc.
     pub fn is_special(self) -> bool {

src/tools/rustfmt/src/parse/macros/mod.rs

@@ -4,8 +4,7 @@ use rustc_ast::{ast, ptr};
 use rustc_parse::MACRO_ARGUMENTS;
 use rustc_parse::parser::{ForceCollect, Parser, Recovery};
 use rustc_session::parse::ParseSess;
-use rustc_span::Symbol;
-use rustc_span::symbol::{self, kw};
+use rustc_span::symbol;
 
 use crate::macros::MacroArg;
 use crate::rewrite::RewriteContext;
@@ -82,18 +81,18 @@ pub(crate) struct ParsedMacroArgs {
 }
 
 fn check_keyword<'a, 'b: 'a>(parser: &'a mut Parser<'b>) -> Option<MacroArg> {
-    for &keyword in RUST_KW.iter() {
-        if parser.token.is_keyword(keyword)
-            && parser.look_ahead(1, |t| *t == TokenKind::Eof || *t == TokenKind::Comma)
-        {
-            parser.bump();
-            return Some(MacroArg::Keyword(
-                symbol::Ident::with_dummy_span(keyword),
-                parser.prev_token.span,
-            ));
-        }
+    if parser.token.is_any_keyword()
+        && parser.look_ahead(1, |t| *t == TokenKind::Eof || *t == TokenKind::Comma)
+    {
+        let keyword = parser.token.ident().unwrap().0.name;
+        parser.bump();
+        Some(MacroArg::Keyword(
+            symbol::Ident::with_dummy_span(keyword),
+            parser.prev_token.span,
+        ))
+    } else {
+        None
     }
-    None
 }
 
 pub(crate) fn parse_macro_args(
@@ -169,65 +168,3 @@ pub(crate) fn parse_expr(
     let mut parser = build_parser(context, tokens);
     parser.parse_expr().ok()
 }
-
-const RUST_KW: [Symbol; 59] = [
-    kw::PathRoot,
-    kw::DollarCrate,
-    kw::Underscore,
-    kw::As,
-    kw::Box,
-    kw::Break,
-    kw::Const,
-    kw::Continue,
-    kw::Crate,
-    kw::Else,
-    kw::Enum,
-    kw::Extern,
-    kw::False,
-    kw::Fn,
-    kw::For,
-    kw::If,
-    kw::Impl,
-    kw::In,
-    kw::Let,
-    kw::Loop,
-    kw::Match,
-    kw::Mod,
-    kw::Move,
-    kw::Mut,
-    kw::Pub,
-    kw::Ref,
-    kw::Return,
-    kw::SelfLower,
-    kw::SelfUpper,
-    kw::Static,
-    kw::Struct,
-    kw::Super,
-    kw::Trait,
-    kw::True,
-    kw::Type,
-    kw::Unsafe,
-    kw::Use,
-    kw::Where,
-    kw::While,
-    kw::Abstract,
-    kw::Become,
-    kw::Do,
-    kw::Final,
-    kw::Macro,
-    kw::Override,
-    kw::Priv,
-    kw::Typeof,
-    kw::Unsized,
-    kw::Virtual,
-    kw::Yield,
-    kw::Dyn,
-    kw::Async,
-    kw::Try,
-    kw::UnderscoreLifetime,
-    kw::StaticLifetime,
-    kw::Auto,
-    kw::Catch,
-    kw::Default,
-    kw::Union,
-];