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, plugin)]
21 use std::collections::HashMap;
26 use syntax::parse::lexer;
27 use rustc::session::{self, config};
30 use syntax::ast::Name;
31 use syntax::parse::token;
32 use syntax::parse::lexer::TokenAndSpan;
34 fn parse_token_list(file: &str) -> HashMap<String, token::Token> {
35 fn id() -> token::Token {
36 token::Ident(ast::Ident { name: Name(0), ctxt: 0, }, token::Plain)
39 let mut res = HashMap::new();
41 res.insert("-1".to_string(), token::Eof);
43 for line in file.split('\n') {
44 let eq = match line.trim().rfind('=') {
49 let val = line.slice_to(eq);
50 let num = line.slice_from(eq + 1);
53 "SHR" => token::BinOp(token::Shr),
54 "DOLLAR" => token::Dollar,
56 "STAR" => token::BinOp(token::Star),
57 "FLOAT_SUFFIX" => id(),
59 "SHL" => token::BinOp(token::Shl),
60 "LBRACE" => token::OpenDelim(token::Brace),
61 "RARROW" => token::RArrow,
62 "LIT_STR" => token::Literal(token::Str_(Name(0)), None),
63 "DOTDOT" => token::DotDot,
64 "MOD_SEP" => token::ModSep,
65 "DOTDOTDOT" => token::DotDotDot,
67 "AND" => token::BinOp(token::And),
68 "LPAREN" => token::OpenDelim(token::Paren),
69 "ANDAND" => token::AndAnd,
71 "LBRACKET" => token::OpenDelim(token::Bracket),
72 "LIT_STR_RAW" => token::Literal(token::StrRaw(Name(0), 0), None),
73 "RPAREN" => token::CloseDelim(token::Paren),
74 "SLASH" => token::BinOp(token::Slash),
75 "COMMA" => token::Comma,
76 "LIFETIME" => token::Lifetime(ast::Ident { name: Name(0), ctxt: 0 }),
77 "CARET" => token::BinOp(token::Caret),
78 "TILDE" => token::Tilde,
80 "PLUS" => token::BinOp(token::Plus),
81 "LIT_CHAR" => token::Literal(token::Char(Name(0)), None),
82 "LIT_BYTE" => token::Literal(token::Byte(Name(0)), None),
84 "RBRACKET" => token::CloseDelim(token::Bracket),
85 "COMMENT" => token::Comment,
86 "DOC_COMMENT" => token::DocComment(Name(0)),
88 "EQEQ" => token::EqEq,
91 "PERCENT" => token::BinOp(token::Percent),
92 "RBRACE" => token::CloseDelim(token::Brace),
93 "BINOP" => token::BinOp(token::Plus),
94 "POUND" => token::Pound,
95 "OROR" => token::OrOr,
96 "LIT_INTEGER" => token::Literal(token::Integer(Name(0)), None),
97 "BINOPEQ" => token::BinOpEq(token::Plus),
98 "LIT_FLOAT" => token::Literal(token::Float(Name(0)), None),
99 "WHITESPACE" => token::Whitespace,
100 "UNDERSCORE" => token::Underscore,
101 "MINUS" => token::BinOp(token::Minus),
102 "SEMI" => token::Semi,
103 "COLON" => token::Colon,
104 "FAT_ARROW" => token::FatArrow,
105 "OR" => token::BinOp(token::Or),
108 "LIT_BINARY" => token::Literal(token::Binary(Name(0)), None),
109 "LIT_BINARY_RAW" => token::Literal(token::BinaryRaw(Name(0), 0), None),
113 res.insert(num.to_string(), tok);
116 debug!("Token map: {}", res);
120 fn str_to_binop(s: &str) -> token::BinOpToken {
126 "%" => token::Percent,
132 _ => panic!("Bad binop str `{}`", s),
136 /// Assuming a string/binary literal, strip out the leading/trailing
137 /// hashes and surrounding quotes/raw/binary prefix.
138 fn fix(mut lit: &str) -> ast::Name {
139 if lit.char_at(0) == 'r' {
140 if lit.char_at(1) == 'b' {
141 lit = lit.slice_from(2)
143 lit = lit.slice_from(1);
145 } else if lit.char_at(0) == 'b' {
146 lit = lit.slice_from(1);
149 let leading_hashes = count(lit);
151 // +1/-1 to adjust for single quotes
152 parse::token::intern(lit.slice(leading_hashes + 1, lit.len() - leading_hashes - 1))
155 /// Assuming a char/byte literal, strip the 'b' prefix and the single quotes.
156 fn fixchar(mut lit: &str) -> ast::Name {
157 if lit.char_at(0) == 'b' {
158 lit = lit.slice_from(1);
161 parse::token::intern(lit.slice(1, lit.len() - 1))
164 fn count(lit: &str) -> uint {
165 lit.chars().take_while(|c| *c == '#').count()
168 fn parse_antlr_token(s: &str, tokens: &HashMap<String, token::Token>) -> TokenAndSpan {
170 r"\[@(?P<seq>\d+),(?P<start>\d+):(?P<end>\d+)='(?P<content>.+?)',<(?P<toknum>-?\d+)>,\d+:\d+]"
173 let m = re.captures(s).expect(format!("The regex didn't match {}", s).as_slice());
174 let start = m.name("start").unwrap_or("");
175 let end = m.name("end").unwrap_or("");
176 let toknum = m.name("toknum").unwrap_or("");
177 let content = m.name("content").unwrap_or("");
179 let proto_tok = tokens.get(toknum).expect(format!("didn't find token {} in the map",
182 let nm = parse::token::intern(content);
184 debug!("What we got: content (`{}`), proto: {}", content, proto_tok);
186 let real_tok = match *proto_tok {
187 token::BinOp(..) => token::BinOp(str_to_binop(content)),
188 token::BinOpEq(..) => token::BinOpEq(str_to_binop(content.slice_to(
189 content.len() - 1))),
190 token::Literal(token::Str_(..), n) => token::Literal(token::Str_(fix(content)), n),
191 token::Literal(token::StrRaw(..), n) => token::Literal(token::StrRaw(fix(content),
193 token::Literal(token::Char(..), n) => token::Literal(token::Char(fixchar(content)), n),
194 token::Literal(token::Byte(..), n) => token::Literal(token::Byte(fixchar(content)), n),
195 token::DocComment(..) => token::DocComment(nm),
196 token::Literal(token::Integer(..), n) => token::Literal(token::Integer(nm), n),
197 token::Literal(token::Float(..), n) => token::Literal(token::Float(nm), n),
198 token::Literal(token::Binary(..), n) => token::Literal(token::Binary(nm), n),
199 token::Literal(token::BinaryRaw(..), n) => token::Literal(token::BinaryRaw(fix(content),
201 token::Ident(..) => token::Ident(ast::Ident { name: nm, ctxt: 0 },
203 token::Lifetime(..) => token::Lifetime(ast::Ident { name: nm, ctxt: 0 }),
207 let offset = if real_tok == token::Eof
214 let sp = syntax::codemap::Span {
215 lo: syntax::codemap::BytePos(start.parse::<u32>().unwrap() - offset),
216 hi: syntax::codemap::BytePos(end.parse::<u32>().unwrap() + 1),
217 expn_id: syntax::codemap::NO_EXPANSION
226 fn tok_cmp(a: &token::Token, b: &token::Token) -> bool {
228 &token::Ident(id, _) => match b {
229 &token::Ident(id2, _) => id == id2,
237 fn next(r: &mut lexer::StringReader) -> TokenAndSpan {
238 use syntax::parse::lexer::Reader;
242 let args = std::os::args();
244 let mut token_file = File::open(&Path::new(args[2].as_slice()));
245 let token_map = parse_token_list(token_file.read_to_string().unwrap().as_slice());
247 let mut stdin = std::io::stdin();
248 let mut lock = stdin.lock();
249 let lines = lock.lines();
250 let mut antlr_tokens = lines.map(|l| parse_antlr_token(l.unwrap().as_slice().trim(),
253 let code = File::open(&Path::new(args[1].as_slice())).unwrap().read_to_string().unwrap();
254 let options = config::basic_options();
255 let session = session::build_session(options, None,
256 syntax::diagnostics::registry::Registry::new(&[]));
257 let filemap = parse::string_to_filemap(&session.parse_sess,
259 String::from_str("<n/a>"));
260 let mut lexer = lexer::StringReader::new(session.diagnostic(), filemap);
262 for antlr_tok in antlr_tokens {
263 let rustc_tok = next(&mut lexer);
264 if rustc_tok.tok == token::Eof && antlr_tok.tok == token::Eof {
268 assert!(rustc_tok.sp == antlr_tok.sp, "{} and {} have different spans", rustc_tok,
271 macro_rules! matches {
273 match rustc_tok.tok {
274 $($x => match antlr_tok.tok {
276 if !tok_cmp(&rustc_tok.tok, &antlr_tok.tok) {
277 // FIXME #15677: needs more robust escaping in
279 warn!("Different names for {} and {}", rustc_tok, antlr_tok);
282 _ => panic!("{} is not {}", antlr_tok, rustc_tok)
284 ref c => assert!(c == &antlr_tok.tok, "{} is not {}", rustc_tok, antlr_tok)
290 token::Literal(token::Byte(..), _),
291 token::Literal(token::Char(..), _),
292 token::Literal(token::Integer(..), _),
293 token::Literal(token::Float(..), _),
294 token::Literal(token::Str_(..), _),
295 token::Literal(token::StrRaw(..), _),
296 token::Literal(token::Binary(..), _),
297 token::Literal(token::BinaryRaw(..), _),
300 token::Interpolated(..),
301 token::DocComment(..),