Auto merge of #60679 - petrochenkov:lit2, r=matklad

Keep original literal tokens in AST

The original literal tokens (`token::Lit`) are kept in AST until lowering to HIR.

The tokens are kept together with their lowered "semantic" representation (`ast::LitKind`), so the size of `ast::Lit` is increased (this also increases the size of meta-item structs used for processing built-in attributes).
However, the size of `ast::Expr` stays the same.

The intent is to remove the "semantic" representation from AST eventually and keep literals as tokens until lowering to HIR (at least), and I'm going to work on that, but it would be good to land this sooner to unblock progress on the [lexer refactoring](https://github.com/rust-lang/rust/pull/59706).

Fixes a part of https://github.com/rust-lang/rust/issues/43081 (literal tokens that are passed to proc macros are always precise, including hexadecimal numbers, strings with their original escaping, etc)
Fixes a part of https://github.com/rust-lang/rust/issues/60495 (everything except for proc macro API doesn't need escaping anymore)
This also allows to eliminate a certain hack from the lexer (https://rust-lang.zulipchat.com/#narrow/stream/131828-t-compiler/topic/pretty-printing.20comments/near/165005357).

cc @matklad

1 files changed, 6 insertions, 132 deletions

diff --git a/src/libsyntax/parse/parser.rs b/src/libsyntax/parse/parser.rs
index d97d1e2f0f4..f95981680b9 100644
--- a/src/libsyntax/parse/parser.rs
+++ b/src/libsyntax/parse/parser.rs
@@ -15,7 +15,7 @@ use crate::ast::{ForeignItem, ForeignItemKind, FunctionRetTy};
 use crate::ast::{GenericParam, GenericParamKind};
 use crate::ast::GenericArg;
 use crate::ast::{Ident, ImplItem, IsAsync, IsAuto, Item, ItemKind};
-use crate::ast::{Label, Lifetime, Lit, LitKind};
+use crate::ast::{Label, Lifetime};
 use crate::ast::{Local, LocalSource};
 use crate::ast::MacStmtStyle;
 use crate::ast::{Mac, Mac_, MacDelimiter};
@@ -35,7 +35,7 @@ use crate::ast::{RangeEnd, RangeSyntax};
 use crate::{ast, attr};
 use crate::ext::base::DummyResult;
 use crate::source_map::{self, SourceMap, Spanned, respan};
-use crate::parse::{self, SeqSep, classify, token};
+use crate::parse::{SeqSep, classify, literal, token};
 use crate::parse::lexer::{TokenAndSpan, UnmatchedBrace};
 use crate::parse::lexer::comments::{doc_comment_style, strip_doc_comment_decoration};
 use crate::parse::token::DelimToken;
@@ -46,7 +46,7 @@ use crate::ptr::P;
 use crate::parse::PResult;
 use crate::ThinVec;
 use crate::tokenstream::{self, DelimSpan, TokenTree, TokenStream, TreeAndJoint};
-use crate::symbol::{Symbol, keywords};
+use crate::symbol::{keywords, Symbol};
 
 use errors::{Applicability, DiagnosticBuilder, DiagnosticId, FatalError};
 use rustc_target::spec::abi::{self, Abi};
@@ -613,7 +613,7 @@ impl<'a> Parser<'a> {
         })
     }
 
-    fn this_token_descr(&self) -> String {
+    crate fn this_token_descr(&self) -> String {
         if let Some(prefix) = self.token_descr() {
             format!("{} `{}`", prefix, self.this_token_to_string())
         } else {
@@ -621,11 +621,6 @@ impl<'a> Parser<'a> {
         }
     }
 
-    fn unexpected_last<T>(&self, t: &token::Token) -> PResult<'a, T> {
-        let token_str = pprust::token_to_string(t);
-        Err(self.span_fatal(self.prev_span, &format!("unexpected token: `{}`", token_str)))
-    }
-
     crate fn unexpected<T>(&mut self) -> PResult<'a, T> {
         match self.expect_one_of(&[], &[]) {
             Err(e) => Err(e),
@@ -1109,43 +1104,7 @@ impl<'a> Parser<'a> {
     }
 
     fn expect_no_suffix(&self, sp: Span, kind: &str, suffix: Option<ast::Name>) {
-        match suffix {
-            None => {/* everything ok */}
-            Some(suf) => {
-                let text = suf.as_str();
-                if text.is_empty() {
-                    self.span_bug(sp, "found empty literal suffix in Some")
-                }
-                let mut err = if kind == "a tuple index" &&
-                    ["i32", "u32", "isize", "usize"].contains(&text.to_string().as_str())
-                {
-                    // #59553: warn instead of reject out of hand to allow the fix to percolate
-                    // through the ecosystem when people fix their macros
-                    let mut err = self.struct_span_warn(
-                        sp,
-                        &format!("suffixes on {} are invalid", kind),
-                    );
-                    err.note(&format!(
-                        "`{}` is *temporarily* accepted on tuple index fields as it was \
-                         incorrectly accepted on stable for a few releases",
-                        text,
-                    ));
-                    err.help(
-                        "on proc macros, you'll want to use `syn::Index::from` or \
-                         `proc_macro::Literal::*_unsuffixed` for code that will desugar \
-                         to tuple field access",
-                    );
-                    err.note(
-                        "for more context, see https://github.com/rust-lang/rust/issues/60210",
-                    );
-                    err
-                } else {
-                    self.struct_span_err(sp, &format!("suffixes on {} are invalid", kind))
-                };
-                err.span_label(sp, format!("invalid suffix `{}`", text));
-                err.emit();
-            }
-        }
+        literal::expect_no_suffix(sp, &self.sess.span_diagnostic, kind, suffix)
     }
 
     /// Attempts to consume a `<`. If `<<` is seen, replaces it with a single
@@ -1423,7 +1382,7 @@ impl<'a> Parser<'a> {
         })
     }
 
-    fn look_ahead_span(&self, dist: usize) -> Span {
+    crate fn look_ahead_span(&self, dist: usize) -> Span {
         if dist == 0 {
             return self.span
         }
@@ -1452,9 +1411,6 @@ impl<'a> Parser<'a> {
     crate fn struct_span_err<S: Into<MultiSpan>>(&self, sp: S, m: &str) -> DiagnosticBuilder<'a> {
         self.sess.span_diagnostic.struct_span_err(sp, m)
     }
-    fn struct_span_warn<S: Into<MultiSpan>>(&self, sp: S, m: &str) -> DiagnosticBuilder<'a> {
-        self.sess.span_diagnostic.struct_span_warn(sp, m)
-    }
     crate fn span_bug<S: Into<MultiSpan>>(&self, sp: S, m: &str) -> ! {
         self.sess.span_diagnostic.span_bug(sp, m)
     }
@@ -2069,88 +2025,6 @@ impl<'a> Parser<'a> {
         }
     }
 
-    /// Matches `token_lit = LIT_INTEGER | ...`.
-    fn parse_lit_token(&mut self) -> PResult<'a, LitKind> {
-        let out = match self.token {
-            token::Interpolated(ref nt) => match **nt {
-                token::NtExpr(ref v) | token::NtLiteral(ref v) => match v.node {
-                    ExprKind::Lit(ref lit) => { lit.node.clone() }
-                    _ => { return self.unexpected_last(&self.token); }
-                },
-                _ => { return self.unexpected_last(&self.token); }
-            },
-            token::Literal(lit, suf) => {
-                let diag = Some((self.span, &self.sess.span_diagnostic));
-                let (suffix_illegal, result) = parse::lit_token(lit, suf, diag);
-
-                if suffix_illegal {
-                    let sp = self.span;
-                    self.expect_no_suffix(sp, &format!("a {}", lit.literal_name()), suf)
-                }
-
-                result.unwrap()
-            }
-            token::Dot if self.look_ahead(1, |t| match t {
-                token::Literal(parse::token::Lit::Integer(_) , _) => true,
-                _ => false,
-            }) => { // recover from `let x = .4;`
-                let lo = self.span;
-                self.bump();
-                if let token::Literal(
-                    parse::token::Lit::Integer(val),
-                    suffix,
-                ) = self.token {
-                    let suffix = suffix.and_then(|s| {
-                        let s = s.as_str();
-                        if s == "f32" {
-                            Some("f32")
-                        } else if s == "f64" {
-                            Some("f64")
-                        } else {
-                            None
-                        }
-                    }).unwrap_or("");
-                    self.bump();
-                    let sp = lo.to(self.prev_span);
-                    let mut err = self.diagnostic()
-                        .struct_span_err(sp, "float literals must have an integer part");
-                    err.span_suggestion(
-                        sp,
-                        "must have an integer part",
-                        format!("0.{}{}", val, suffix),
-                        Applicability::MachineApplicable,
-                    );
-                    err.emit();
-                    return Ok(match suffix {
-                        "f32" => ast::LitKind::Float(val, ast::FloatTy::F32),
-                        "f64" => ast::LitKind::Float(val, ast::FloatTy::F64),
-                        _ => ast::LitKind::FloatUnsuffixed(val),
-                    });
-                } else {
-                    unreachable!();
-                };
-            }
-            _ => { return self.unexpected_last(&self.token); }
-        };
-
-        self.bump();
-        Ok(out)
-    }
-
-    /// Matches `lit = true | false | token_lit`.
-    crate fn parse_lit(&mut self) -> PResult<'a, Lit> {
-        let lo = self.span;
-        let lit = if self.eat_keyword(keywords::True) {
-            LitKind::Bool(true)
-        } else if self.eat_keyword(keywords::False) {
-            LitKind::Bool(false)
-        } else {
-            let lit = self.parse_lit_token()?;
-            lit
-        };
-        Ok(source_map::Spanned { node: lit, span: lo.to(self.prev_span) })
-    }
-
     /// Matches `'-' lit | lit` (cf. `ast_validation::AstValidator::check_expr_within_pat`).
     crate fn parse_literal_maybe_minus(&mut self) -> PResult<'a, P<Expr>> {
         maybe_whole_expr!(self);