2014-07-21 15:04:35 -05:00
|
|
|
// Copyright 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.
|
|
|
|
|
2014-07-14 03:52:18 -05:00
|
|
|
#![feature(globs, phase, macro_rules)]
|
|
|
|
|
|
|
|
extern crate syntax;
|
|
|
|
extern crate rustc;
|
|
|
|
|
|
|
|
#[phase(link)]
|
|
|
|
extern crate regex;
|
|
|
|
|
|
|
|
#[phase(link, plugin)]
|
|
|
|
extern crate log;
|
|
|
|
|
|
|
|
#[phase(plugin)] extern crate regex_macros;
|
|
|
|
|
|
|
|
use std::collections::HashMap;
|
|
|
|
use std::io::File;
|
|
|
|
|
|
|
|
use syntax::parse;
|
|
|
|
use syntax::parse::lexer;
|
2015-01-03 21:42:21 -06:00
|
|
|
use rustc::session::{self, config};
|
2014-07-14 03:52:18 -05:00
|
|
|
|
|
|
|
use syntax::ast;
|
|
|
|
use syntax::ast::Name;
|
2014-10-27 03:22:52 -05:00
|
|
|
use syntax::parse::token;
|
2014-07-14 03:52:18 -05:00
|
|
|
use syntax::parse::lexer::TokenAndSpan;
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
fn parse_token_list(file: &str) -> HashMap<String, token::Token> {
|
|
|
|
fn id() -> token::Token {
|
2014-10-27 10:01:44 -05:00
|
|
|
token::Ident(ast::Ident { name: Name(0), ctxt: 0, }, token::Plain)
|
2014-07-14 03:52:18 -05:00
|
|
|
}
|
|
|
|
|
|
|
|
let mut res = HashMap::new();
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
res.insert("-1".to_string(), token::Eof);
|
2014-07-14 03:52:18 -05:00
|
|
|
|
|
|
|
for line in file.split('\n') {
|
|
|
|
let eq = match line.trim().rfind('=') {
|
|
|
|
Some(val) => val,
|
|
|
|
None => continue
|
|
|
|
};
|
|
|
|
|
|
|
|
let val = line.slice_to(eq);
|
|
|
|
let num = line.slice_from(eq + 1);
|
|
|
|
|
|
|
|
let tok = match val {
|
2014-10-27 03:22:52 -05:00
|
|
|
"SHR" => token::BinOp(token::Shr),
|
|
|
|
"DOLLAR" => token::Dollar,
|
|
|
|
"LT" => token::Lt,
|
|
|
|
"STAR" => token::BinOp(token::Star),
|
|
|
|
"FLOAT_SUFFIX" => id(),
|
|
|
|
"INT_SUFFIX" => id(),
|
|
|
|
"SHL" => token::BinOp(token::Shl),
|
2014-10-29 05:37:54 -05:00
|
|
|
"LBRACE" => token::OpenDelim(token::Brace),
|
2014-11-18 19:52:44 -06:00
|
|
|
"RARROW" => token::RArrow,
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_STR" => token::Literal(token::Str_(Name(0)), None),
|
2014-10-27 03:22:52 -05:00
|
|
|
"DOTDOT" => token::DotDot,
|
|
|
|
"MOD_SEP" => token::ModSep,
|
|
|
|
"DOTDOTDOT" => token::DotDotDot,
|
|
|
|
"NOT" => token::Not,
|
|
|
|
"AND" => token::BinOp(token::And),
|
2014-10-29 05:37:54 -05:00
|
|
|
"LPAREN" => token::OpenDelim(token::Paren),
|
2014-10-27 03:22:52 -05:00
|
|
|
"ANDAND" => token::AndAnd,
|
|
|
|
"AT" => token::At,
|
2014-10-29 05:37:54 -05:00
|
|
|
"LBRACKET" => token::OpenDelim(token::Bracket),
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_STR_RAW" => token::Literal(token::StrRaw(Name(0), 0), None),
|
2014-10-29 05:37:54 -05:00
|
|
|
"RPAREN" => token::CloseDelim(token::Paren),
|
2014-10-27 03:22:52 -05:00
|
|
|
"SLASH" => token::BinOp(token::Slash),
|
|
|
|
"COMMA" => token::Comma,
|
|
|
|
"LIFETIME" => token::Lifetime(ast::Ident { name: Name(0), ctxt: 0 }),
|
|
|
|
"CARET" => token::BinOp(token::Caret),
|
|
|
|
"TILDE" => token::Tilde,
|
2014-11-18 19:52:44 -06:00
|
|
|
"IDENT" => id(),
|
2014-10-27 03:22:52 -05:00
|
|
|
"PLUS" => token::BinOp(token::Plus),
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_CHAR" => token::Literal(token::Char(Name(0)), None),
|
|
|
|
"LIT_BYTE" => token::Literal(token::Byte(Name(0)), None),
|
2014-10-27 03:22:52 -05:00
|
|
|
"EQ" => token::Eq,
|
2014-10-29 05:37:54 -05:00
|
|
|
"RBRACKET" => token::CloseDelim(token::Bracket),
|
2014-10-27 03:22:52 -05:00
|
|
|
"COMMENT" => token::Comment,
|
|
|
|
"DOC_COMMENT" => token::DocComment(Name(0)),
|
|
|
|
"DOT" => token::Dot,
|
|
|
|
"EQEQ" => token::EqEq,
|
|
|
|
"NE" => token::Ne,
|
|
|
|
"GE" => token::Ge,
|
|
|
|
"PERCENT" => token::BinOp(token::Percent),
|
2014-10-29 05:37:54 -05:00
|
|
|
"RBRACE" => token::CloseDelim(token::Brace),
|
2014-10-27 03:22:52 -05:00
|
|
|
"BINOP" => token::BinOp(token::Plus),
|
|
|
|
"POUND" => token::Pound,
|
|
|
|
"OROR" => token::OrOr,
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_INTEGER" => token::Literal(token::Integer(Name(0)), None),
|
2014-10-27 03:22:52 -05:00
|
|
|
"BINOPEQ" => token::BinOpEq(token::Plus),
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_FLOAT" => token::Literal(token::Float(Name(0)), None),
|
2014-10-27 03:22:52 -05:00
|
|
|
"WHITESPACE" => token::Whitespace,
|
|
|
|
"UNDERSCORE" => token::Underscore,
|
|
|
|
"MINUS" => token::BinOp(token::Minus),
|
|
|
|
"SEMI" => token::Semi,
|
|
|
|
"COLON" => token::Colon,
|
|
|
|
"FAT_ARROW" => token::FatArrow,
|
|
|
|
"OR" => token::BinOp(token::Or),
|
|
|
|
"GT" => token::Gt,
|
|
|
|
"LE" => token::Le,
|
2014-12-26 13:49:36 -06:00
|
|
|
"LIT_BINARY" => token::Literal(token::Binary(Name(0)), None),
|
|
|
|
"LIT_BINARY_RAW" => token::Literal(token::BinaryRaw(Name(0), 0), None),
|
2014-10-27 03:22:52 -05:00
|
|
|
_ => continue,
|
2014-07-14 03:52:18 -05:00
|
|
|
};
|
|
|
|
|
|
|
|
res.insert(num.to_string(), tok);
|
|
|
|
}
|
|
|
|
|
|
|
|
debug!("Token map: {}", res);
|
|
|
|
res
|
|
|
|
}
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
fn str_to_binop(s: &str) -> token::BinOpToken {
|
2014-07-14 03:52:18 -05:00
|
|
|
match s {
|
2014-10-27 03:22:52 -05:00
|
|
|
"+" => token::Plus,
|
|
|
|
"/" => token::Slash,
|
|
|
|
"-" => token::Minus,
|
|
|
|
"*" => token::Star,
|
|
|
|
"%" => token::Percent,
|
|
|
|
"^" => token::Caret,
|
|
|
|
"&" => token::And,
|
|
|
|
"|" => token::Or,
|
|
|
|
"<<" => token::Shl,
|
|
|
|
">>" => token::Shr,
|
2014-10-09 14:17:22 -05:00
|
|
|
_ => panic!("Bad binop str `{}`", s),
|
2014-07-14 19:27:28 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-07-14 22:45:39 -05:00
|
|
|
/// Assuming a string/binary literal, strip out the leading/trailing
|
2014-07-14 19:27:28 -05:00
|
|
|
/// hashes and surrounding quotes/raw/binary prefix.
|
|
|
|
fn fix(mut lit: &str) -> ast::Name {
|
|
|
|
if lit.char_at(0) == 'r' {
|
|
|
|
if lit.char_at(1) == 'b' {
|
|
|
|
lit = lit.slice_from(2)
|
|
|
|
} else {
|
|
|
|
lit = lit.slice_from(1);
|
|
|
|
}
|
|
|
|
} else if lit.char_at(0) == 'b' {
|
|
|
|
lit = lit.slice_from(1);
|
2014-07-14 03:52:18 -05:00
|
|
|
}
|
2014-07-14 19:27:28 -05:00
|
|
|
|
|
|
|
let leading_hashes = count(lit);
|
|
|
|
|
|
|
|
// +1/-1 to adjust for single quotes
|
|
|
|
parse::token::intern(lit.slice(leading_hashes + 1, lit.len() - leading_hashes - 1))
|
|
|
|
}
|
|
|
|
|
2014-07-14 22:45:39 -05:00
|
|
|
/// Assuming a char/byte literal, strip the 'b' prefix and the single quotes.
|
|
|
|
fn fixchar(mut lit: &str) -> ast::Name {
|
|
|
|
if lit.char_at(0) == 'b' {
|
|
|
|
lit = lit.slice_from(1);
|
|
|
|
}
|
|
|
|
|
|
|
|
parse::token::intern(lit.slice(1, lit.len() - 1))
|
|
|
|
}
|
|
|
|
|
2014-07-14 19:27:28 -05:00
|
|
|
fn count(lit: &str) -> uint {
|
|
|
|
lit.chars().take_while(|c| *c == '#').count()
|
2014-07-14 03:52:18 -05:00
|
|
|
}
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
fn parse_antlr_token(s: &str, tokens: &HashMap<String, token::Token>) -> TokenAndSpan {
|
2014-07-21 15:04:35 -05:00
|
|
|
let re = regex!(
|
|
|
|
r"\[@(?P<seq>\d+),(?P<start>\d+):(?P<end>\d+)='(?P<content>.+?)',<(?P<toknum>-?\d+)>,\d+:\d+]"
|
|
|
|
);
|
2014-07-14 03:52:18 -05:00
|
|
|
|
|
|
|
let m = re.captures(s).expect(format!("The regex didn't match {}", s).as_slice());
|
2014-12-13 12:33:18 -06:00
|
|
|
let start = m.name("start").unwrap_or("");
|
|
|
|
let end = m.name("end").unwrap_or("");
|
|
|
|
let toknum = m.name("toknum").unwrap_or("");
|
|
|
|
let content = m.name("content").unwrap_or("");
|
2014-07-14 03:52:18 -05:00
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
let proto_tok = tokens.get(toknum).expect(format!("didn't find token {} in the map",
|
2014-07-21 15:04:35 -05:00
|
|
|
toknum).as_slice());
|
2014-07-14 19:27:28 -05:00
|
|
|
|
|
|
|
let nm = parse::token::intern(content);
|
|
|
|
|
|
|
|
debug!("What we got: content (`{}`), proto: {}", content, proto_tok);
|
|
|
|
|
2014-07-14 03:52:18 -05:00
|
|
|
let real_tok = match *proto_tok {
|
2014-10-27 03:22:52 -05:00
|
|
|
token::BinOp(..) => token::BinOp(str_to_binop(content)),
|
|
|
|
token::BinOpEq(..) => token::BinOpEq(str_to_binop(content.slice_to(
|
|
|
|
content.len() - 1))),
|
2014-12-26 13:49:36 -06:00
|
|
|
token::Literal(token::Str_(..), n) => token::Literal(token::Str_(fix(content)), n),
|
|
|
|
token::Literal(token::StrRaw(..), n) => token::Literal(token::StrRaw(fix(content),
|
|
|
|
count(content)), n),
|
|
|
|
token::Literal(token::Char(..), n) => token::Literal(token::Char(fixchar(content)), n),
|
|
|
|
token::Literal(token::Byte(..), n) => token::Literal(token::Byte(fixchar(content)), n),
|
2014-10-27 03:22:52 -05:00
|
|
|
token::DocComment(..) => token::DocComment(nm),
|
2014-12-26 13:49:36 -06:00
|
|
|
token::Literal(token::Integer(..), n) => token::Literal(token::Integer(nm), n),
|
|
|
|
token::Literal(token::Float(..), n) => token::Literal(token::Float(nm), n),
|
|
|
|
token::Literal(token::Binary(..), n) => token::Literal(token::Binary(nm), n),
|
|
|
|
token::Literal(token::BinaryRaw(..), n) => token::Literal(token::BinaryRaw(fix(content),
|
|
|
|
count(content)), n),
|
2014-10-27 10:01:44 -05:00
|
|
|
token::Ident(..) => token::Ident(ast::Ident { name: nm, ctxt: 0 },
|
|
|
|
token::ModName),
|
2014-10-27 03:22:52 -05:00
|
|
|
token::Lifetime(..) => token::Lifetime(ast::Ident { name: nm, ctxt: 0 }),
|
2014-07-14 03:52:18 -05:00
|
|
|
ref t => t.clone()
|
|
|
|
};
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
let offset = if real_tok == token::Eof
|
|
|
|
{
|
2014-07-14 03:52:18 -05:00
|
|
|
1
|
|
|
|
} else {
|
|
|
|
0
|
|
|
|
};
|
|
|
|
|
|
|
|
let sp = syntax::codemap::Span {
|
2014-12-26 13:49:36 -06:00
|
|
|
lo: syntax::codemap::BytePos(start.parse::<u32>().unwrap() - offset),
|
|
|
|
hi: syntax::codemap::BytePos(end.parse::<u32>().unwrap() + 1),
|
2014-09-18 06:36:01 -05:00
|
|
|
expn_id: syntax::codemap::NO_EXPANSION
|
2014-07-14 03:52:18 -05:00
|
|
|
};
|
|
|
|
|
|
|
|
TokenAndSpan {
|
|
|
|
tok: real_tok,
|
|
|
|
sp: sp
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
fn tok_cmp(a: &token::Token, b: &token::Token) -> bool {
|
2014-07-14 19:27:28 -05:00
|
|
|
match a {
|
2014-10-27 03:22:52 -05:00
|
|
|
&token::Ident(id, _) => match b {
|
|
|
|
&token::Ident(id2, _) => id == id2,
|
2014-07-14 19:27:28 -05:00
|
|
|
_ => false
|
|
|
|
},
|
|
|
|
_ => a == b
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-07-14 03:52:18 -05:00
|
|
|
fn main() {
|
|
|
|
fn next(r: &mut lexer::StringReader) -> TokenAndSpan {
|
|
|
|
use syntax::parse::lexer::Reader;
|
|
|
|
r.next_token()
|
|
|
|
}
|
|
|
|
|
2014-07-15 02:18:17 -05:00
|
|
|
let args = std::os::args();
|
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
let mut token_file = File::open(&Path::new(args[2].as_slice()));
|
2014-07-21 14:59:25 -05:00
|
|
|
let token_map = parse_token_list(token_file.read_to_string().unwrap().as_slice());
|
|
|
|
|
2014-07-14 03:52:18 -05:00
|
|
|
let mut stdin = std::io::stdin();
|
2014-12-26 13:49:36 -06:00
|
|
|
let mut lock = stdin.lock();
|
|
|
|
let lines = lock.lines();
|
|
|
|
let mut antlr_tokens = lines.map(|l| parse_antlr_token(l.unwrap().as_slice().trim(),
|
2014-07-21 15:04:35 -05:00
|
|
|
&token_map));
|
2014-07-14 03:52:18 -05:00
|
|
|
|
2014-11-18 19:52:44 -06:00
|
|
|
let code = File::open(&Path::new(args[1].as_slice())).unwrap().read_to_string().unwrap();
|
2014-07-14 03:52:18 -05:00
|
|
|
let options = config::basic_options();
|
2014-07-14 19:27:28 -05:00
|
|
|
let session = session::build_session(options, None,
|
2014-11-18 19:52:44 -06:00
|
|
|
syntax::diagnostics::registry::Registry::new(&[]));
|
2014-07-14 03:52:18 -05:00
|
|
|
let filemap = parse::string_to_filemap(&session.parse_sess,
|
|
|
|
code,
|
|
|
|
String::from_str("<n/a>"));
|
|
|
|
let mut lexer = lexer::StringReader::new(session.diagnostic(), filemap);
|
|
|
|
|
|
|
|
for antlr_tok in antlr_tokens {
|
|
|
|
let rustc_tok = next(&mut lexer);
|
2014-11-18 19:52:44 -06:00
|
|
|
if rustc_tok.tok == token::Eof && antlr_tok.tok == token::Eof {
|
2014-07-14 03:52:18 -05:00
|
|
|
continue
|
|
|
|
}
|
|
|
|
|
2014-07-21 15:04:35 -05:00
|
|
|
assert!(rustc_tok.sp == antlr_tok.sp, "{} and {} have different spans", rustc_tok,
|
|
|
|
antlr_tok);
|
2014-07-14 03:52:18 -05:00
|
|
|
|
|
|
|
macro_rules! matches (
|
|
|
|
( $($x:pat),+ ) => (
|
|
|
|
match rustc_tok.tok {
|
|
|
|
$($x => match antlr_tok.tok {
|
2014-07-14 19:27:28 -05:00
|
|
|
$x => {
|
|
|
|
if !tok_cmp(&rustc_tok.tok, &antlr_tok.tok) {
|
|
|
|
// FIXME #15677: needs more robust escaping in
|
|
|
|
// antlr
|
|
|
|
warn!("Different names for {} and {}", rustc_tok, antlr_tok);
|
|
|
|
}
|
|
|
|
}
|
2014-10-09 14:17:22 -05:00
|
|
|
_ => panic!("{} is not {}", antlr_tok, rustc_tok)
|
2014-07-14 03:52:18 -05:00
|
|
|
},)*
|
2014-07-14 19:27:28 -05:00
|
|
|
ref c => assert!(c == &antlr_tok.tok, "{} is not {}", rustc_tok, antlr_tok)
|
2014-07-14 03:52:18 -05:00
|
|
|
}
|
|
|
|
)
|
2014-12-26 13:49:36 -06:00
|
|
|
);
|
2014-07-14 03:52:18 -05:00
|
|
|
|
2014-10-27 03:22:52 -05:00
|
|
|
matches!(
|
2014-12-26 13:49:36 -06:00
|
|
|
token::Literal(token::Byte(..), _),
|
|
|
|
token::Literal(token::Char(..), _),
|
|
|
|
token::Literal(token::Integer(..), _),
|
|
|
|
token::Literal(token::Float(..), _),
|
|
|
|
token::Literal(token::Str_(..), _),
|
|
|
|
token::Literal(token::StrRaw(..), _),
|
|
|
|
token::Literal(token::Binary(..), _),
|
|
|
|
token::Literal(token::BinaryRaw(..), _),
|
2014-11-18 19:52:44 -06:00
|
|
|
token::Ident(..),
|
|
|
|
token::Lifetime(..),
|
|
|
|
token::Interpolated(..),
|
|
|
|
token::DocComment(..),
|
|
|
|
token::Shebang(..)
|
2014-07-14 03:52:18 -05:00
|
|
|
);
|
|
|
|
}
|
|
|
|
}
|