From 3ce9a0ab2c56964821dcecf0a61b30651aa94123 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Pawe=C5=82=20K=C4=99dzia?= <Pawel.Kedzia@pwr.wroc.pl>
Date: Mon, 18 Feb 2013 14:46:26 +0100
Subject: [PATCH] Single and multi line comment handling added to lexicon
grammar
---
libwccl/lexicon/lexfilegrammar.g | 35 +++++++++++++++++++++++++++++++-
tests/data/indecl.lex | 4 ++++
2 files changed, 38 insertions(+), 1 deletion(-)
diff --git a/libwccl/lexicon/lexfilegrammar.g b/libwccl/lexicon/lexfilegrammar.g
index ce36648..2efc4b1 100644
--- a/libwccl/lexicon/lexfilegrammar.g
+++ b/libwccl/lexicon/lexfilegrammar.g
@@ -28,7 +28,7 @@ options {
// ----------------------------------------------------------------------------
class ANTLRLexiconParser extends Parser;
options {
- k = 1;
+ k = 2;
buildAST = false;
exportVocab = ANTLRLexicons;
defaultErrorHandler = false;
@@ -113,3 +113,36 @@ NEWLINE
| '\n'
) { newline(); $setType(antlr::Token::SKIP); }
;
+
+COMMENT
+options {
+ paraphrase = "Single line comment";
+}
+ : "//" (~('\n'|'\r'))* { $setType(antlr::Token::SKIP); }
+;
+
+ML_COMMENT
+options {
+ paraphrase = "Multi line comment";
+}
+ : "/*"
+ ( // TODO: test it and add reference to the site it's taken from!
+ /* This actually works OK despite the ambiguity that
+ '\r' '\n' can be matched in one alternative or by matching
+ '\r' in one iteration and '\n' in another.. But
+ this is really matched just by one rule per (...)*
+ loop iteration, so it's OK.
+ This is exactly how they do it all over the web - just
+ turn off the warning for this particular token.*/
+ options {
+ generateAmbigWarnings = false;
+ }
+ : { LA(2)!='/' }? '*'
+ | '\r' '\n' { newline(); }
+ | '\r' { newline(); }
+ | '\n' { newline(); }
+ | ~('*'|'\n'|'\r')
+ )*
+ "*/"
+ { $setType(antlr::Token::SKIP); }
+;
diff --git a/tests/data/indecl.lex b/tests/data/indecl.lex
index 97474c1..e4a533c 100644
--- a/tests/data/indecl.lex
+++ b/tests/data/indecl.lex
@@ -1,3 +1,7 @@
+/**
+ * ML COMMENT
+ */
+// Single line comment
by part
och interj
ach interj
--
GitLab