1 // Copyright 2014 The Rust Project Developers. See the COPYRIGHT
2 // file at the top-level directory of this distribution and at
3 // http://rust-lang.org/COPYRIGHT.
5 // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6 // http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7 // <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8 // option. This file may not be copied, modified, or distributed
9 // except according to those terms.
11 #![feature(globs, phase, macro_rules)]
19 #[phase(link, plugin)]
22 #[phase(plugin)] extern crate regex_macros;
24 use std::collections::HashMap;
28 use syntax::parse::lexer;
29 use rustc::session::{mod, config};
32 use syntax::ast::Name;
33 use syntax::parse::token;
34 use syntax::parse::lexer::TokenAndSpan;
36 fn parse_token_list(file: &str) -> HashMap<String, token::Token> {
37 fn id() -> token::Token {
38 token::Ident(ast::Ident { name: Name(0), ctxt: 0, }, token::Plain)
41 let mut res = HashMap::new();
43 res.insert("-1".to_string(), token::Eof);
45 for line in file.split('\n') {
46 let eq = match line.trim().rfind('=') {
51 let val = line.slice_to(eq);
52 let num = line.slice_from(eq + 1);
55 "SHR" => token::BinOp(token::Shr),
56 "DOLLAR" => token::Dollar,
58 "STAR" => token::BinOp(token::Star),
59 "FLOAT_SUFFIX" => id(),
61 "SHL" => token::BinOp(token::Shl),
62 "LBRACE" => token::OpenDelim(token::Brace),
63 "RARROW" => token::RArrow,
64 "LIT_STR" => token::Literal(token::Str_(Name(0)), None),
65 "DOTDOT" => token::DotDot,
66 "MOD_SEP" => token::ModSep,
67 "DOTDOTDOT" => token::DotDotDot,
69 "AND" => token::BinOp(token::And),
70 "LPAREN" => token::OpenDelim(token::Paren),
71 "ANDAND" => token::AndAnd,
73 "LBRACKET" => token::OpenDelim(token::Bracket),
74 "LIT_STR_RAW" => token::Literal(token::StrRaw(Name(0), 0), None),
75 "RPAREN" => token::CloseDelim(token::Paren),
76 "SLASH" => token::BinOp(token::Slash),
77 "COMMA" => token::Comma,
78 "LIFETIME" => token::Lifetime(ast::Ident { name: Name(0), ctxt: 0 }),
79 "CARET" => token::BinOp(token::Caret),
80 "TILDE" => token::Tilde,
82 "PLUS" => token::BinOp(token::Plus),
83 "LIT_CHAR" => token::Literal(token::Char(Name(0)), None),
84 "LIT_BYTE" => token::Literal(token::Byte(Name(0)), None),
86 "RBRACKET" => token::CloseDelim(token::Bracket),
87 "COMMENT" => token::Comment,
88 "DOC_COMMENT" => token::DocComment(Name(0)),
90 "EQEQ" => token::EqEq,
93 "PERCENT" => token::BinOp(token::Percent),
94 "RBRACE" => token::CloseDelim(token::Brace),
95 "BINOP" => token::BinOp(token::Plus),
96 "POUND" => token::Pound,
97 "OROR" => token::OrOr,
98 "LIT_INTEGER" => token::Literal(token::Integer(Name(0)), None),
99 "BINOPEQ" => token::BinOpEq(token::Plus),
100 "LIT_FLOAT" => token::Literal(token::Float(Name(0)), None),
101 "WHITESPACE" => token::Whitespace,
102 "UNDERSCORE" => token::Underscore,
103 "MINUS" => token::BinOp(token::Minus),
104 "SEMI" => token::Semi,
105 "COLON" => token::Colon,
106 "FAT_ARROW" => token::FatArrow,
107 "OR" => token::BinOp(token::Or),
110 "LIT_BINARY" => token::Literal(token::Binary(Name(0)), None),
111 "LIT_BINARY_RAW" => token::Literal(token::BinaryRaw(Name(0), 0), None),
115 res.insert(num.to_string(), tok);
118 debug!("Token map: {}", res);
122 fn str_to_binop(s: &str) -> token::BinOpToken {
128 "%" => token::Percent,
134 _ => panic!("Bad binop str `{}`", s),
138 /// Assuming a string/binary literal, strip out the leading/trailing
139 /// hashes and surrounding quotes/raw/binary prefix.
140 fn fix(mut lit: &str) -> ast::Name {
141 if lit.char_at(0) == 'r' {
142 if lit.char_at(1) == 'b' {
143 lit = lit.slice_from(2)
145 lit = lit.slice_from(1);
147 } else if lit.char_at(0) == 'b' {
148 lit = lit.slice_from(1);
151 let leading_hashes = count(lit);
153 // +1/-1 to adjust for single quotes
154 parse::token::intern(lit.slice(leading_hashes + 1, lit.len() - leading_hashes - 1))
157 /// Assuming a char/byte literal, strip the 'b' prefix and the single quotes.
158 fn fixchar(mut lit: &str) -> ast::Name {
159 if lit.char_at(0) == 'b' {
160 lit = lit.slice_from(1);
163 parse::token::intern(lit.slice(1, lit.len() - 1))
166 fn count(lit: &str) -> uint {
167 lit.chars().take_while(|c| *c == '#').count()
170 fn parse_antlr_token(s: &str, tokens: &HashMap<String, token::Token>) -> TokenAndSpan {
172 r"\[@(?P<seq>\d+),(?P<start>\d+):(?P<end>\d+)='(?P<content>.+?)',<(?P<toknum>-?\d+)>,\d+:\d+]"
175 let m = re.captures(s).expect(format!("The regex didn't match {}", s).as_slice());
176 let start = m.name("start").unwrap_or("");
177 let end = m.name("end").unwrap_or("");
178 let toknum = m.name("toknum").unwrap_or("");
179 let content = m.name("content").unwrap_or("");
181 let proto_tok = tokens.get(toknum).expect(format!("didn't find token {} in the map",
184 let nm = parse::token::intern(content);
186 debug!("What we got: content (`{}`), proto: {}", content, proto_tok);
188 let real_tok = match *proto_tok {
189 token::BinOp(..) => token::BinOp(str_to_binop(content)),
190 token::BinOpEq(..) => token::BinOpEq(str_to_binop(content.slice_to(
191 content.len() - 1))),
192 token::Literal(token::Str_(..), n) => token::Literal(token::Str_(fix(content)), n),
193 token::Literal(token::StrRaw(..), n) => token::Literal(token::StrRaw(fix(content),
195 token::Literal(token::Char(..), n) => token::Literal(token::Char(fixchar(content)), n),
196 token::Literal(token::Byte(..), n) => token::Literal(token::Byte(fixchar(content)), n),
197 token::DocComment(..) => token::DocComment(nm),
198 token::Literal(token::Integer(..), n) => token::Literal(token::Integer(nm), n),
199 token::Literal(token::Float(..), n) => token::Literal(token::Float(nm), n),
200 token::Literal(token::Binary(..), n) => token::Literal(token::Binary(nm), n),
201 token::Literal(token::BinaryRaw(..), n) => token::Literal(token::BinaryRaw(fix(content),
203 token::Ident(..) => token::Ident(ast::Ident { name: nm, ctxt: 0 },
205 token::Lifetime(..) => token::Lifetime(ast::Ident { name: nm, ctxt: 0 }),
209 let offset = if real_tok == token::Eof
216 let sp = syntax::codemap::Span {
217 lo: syntax::codemap::BytePos(start.parse::<u32>().unwrap() - offset),
218 hi: syntax::codemap::BytePos(end.parse::<u32>().unwrap() + 1),
219 expn_id: syntax::codemap::NO_EXPANSION
228 fn tok_cmp(a: &token::Token, b: &token::Token) -> bool {
230 &token::Ident(id, _) => match b {
231 &token::Ident(id2, _) => id == id2,
239 fn next(r: &mut lexer::StringReader) -> TokenAndSpan {
240 use syntax::parse::lexer::Reader;
244 let args = std::os::args();
246 let mut token_file = File::open(&Path::new(args[2].as_slice()));
247 let token_map = parse_token_list(token_file.read_to_string().unwrap().as_slice());
249 let mut stdin = std::io::stdin();
250 let mut lock = stdin.lock();
251 let lines = lock.lines();
252 let mut antlr_tokens = lines.map(|l| parse_antlr_token(l.unwrap().as_slice().trim(),
255 let code = File::open(&Path::new(args[1].as_slice())).unwrap().read_to_string().unwrap();
256 let options = config::basic_options();
257 let session = session::build_session(options, None,
258 syntax::diagnostics::registry::Registry::new(&[]));
259 let filemap = parse::string_to_filemap(&session.parse_sess,
261 String::from_str("<n/a>"));
262 let mut lexer = lexer::StringReader::new(session.diagnostic(), filemap);
264 for antlr_tok in antlr_tokens {
265 let rustc_tok = next(&mut lexer);
266 if rustc_tok.tok == token::Eof && antlr_tok.tok == token::Eof {
270 assert!(rustc_tok.sp == antlr_tok.sp, "{} and {} have different spans", rustc_tok,
273 macro_rules! matches (
275 match rustc_tok.tok {
276 $($x => match antlr_tok.tok {
278 if !tok_cmp(&rustc_tok.tok, &antlr_tok.tok) {
279 // FIXME #15677: needs more robust escaping in
281 warn!("Different names for {} and {}", rustc_tok, antlr_tok);
284 _ => panic!("{} is not {}", antlr_tok, rustc_tok)
286 ref c => assert!(c == &antlr_tok.tok, "{} is not {}", rustc_tok, antlr_tok)
292 token::Literal(token::Byte(..), _),
293 token::Literal(token::Char(..), _),
294 token::Literal(token::Integer(..), _),
295 token::Literal(token::Float(..), _),
296 token::Literal(token::Str_(..), _),
297 token::Literal(token::StrRaw(..), _),
298 token::Literal(token::Binary(..), _),
299 token::Literal(token::BinaryRaw(..), _),
302 token::Interpolated(..),
303 token::DocComment(..),