From: Ulya Trofimovich <skvadrik@gmail.com>
Date: Tue, 16 Jun 2015 11:19:03 +0000 (+0100)
Subject: Partial fix for bug #61 "empty character class [] matches empty string".
X-Git-Tag: 0.15~206
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=7157354c453d3f4ba538bf73ac76d2a2d2311dee;p=re2c

Partial fix for bug #61 "empty character class [] matches empty string".

Given the following code:
    /*!re2c
        [] {}
    */

    /*!re2c
        [^\x00-\xFF] {}
    */

    /*!re2c
        [\x00-\xFF]\[\x00-\xFF] {}
    */
re2c versions <=0.13.6 and >=0.13.7 behaved differently.
0.13.6 consistently considered that empty range should match empty string.
Since 0.13.7 empty positive range [] and empty difference (e.g. [a-z][a-z])
still match empty string, but empty negative range (e.g. [^\x00-\xFF])
matches nothing (always fails). The faulty commit is
28ee7c95bca46ad3cdb965741c5c29e21c50df14
"Added UTF-8 encoding support and tests for it."

This commit brings back consistent behaviour of 0.13.6: empty range,
however it was constructed, always matches empty string. Whether this
behaviour is sane or not is another question.
---

diff --git a/re2c/src/ir/regexp/regexp.cc b/re2c/src/ir/regexp/regexp.cc
index 3eade194..d259531f 100644
--- a/re2c/src/ir/regexp/regexp.cc
+++ b/re2c/src/ir/regexp/regexp.cc
@@ -106,20 +106,6 @@ RegExp * doCat (RegExp * e1, RegExp * e2)
 	return new CatOp (e1, e2);
 }
 
-RegExp * mkDiff (RegExp * e1, RegExp * e2)
-{
-	MatchOp * m1 = dynamic_cast<MatchOp *> (e1);
-	MatchOp * m2 = dynamic_cast<MatchOp *> (e2);
-	if (m1 == NULL || m2 == NULL)
-	{
-		return NULL;
-	}
-	Range * r = range_diff (m1->match, m2->match);
-	return r
-		? (RegExp *) new MatchOp(r)
-		: (RegExp *) new NullOp;
-}
-
 Range * Scanner::getRange(SubStr &s) const
 {
 	uint32_t lb = unescape(s), ub;
@@ -223,6 +209,11 @@ Range * Scanner::mkRange(SubStr &s) const
 
 RegExp * Scanner::matchSymbolRange(Range * r) const
 {
+	if (!r)
+	{
+		return new NullOp;
+	}
+
 	if (encoding.is(Enc::UTF16))
 		return UTF16Range(r);
 	else if (encoding.is(Enc::UTF8))
@@ -236,10 +227,11 @@ RegExp * Scanner::ranToRE (SubStr & s) const
 	s.len -= 2;
 	s.str += 1;
 
-	if (s.len == 0)
-		return new NullOp;
+	Range * r = s.len == 0
+		? NULL
+		: mkRange(s);
 
-	return matchSymbolRange(mkRange(s));
+	return matchSymbolRange (r);
 }
 
 RegExp * Scanner::invToRE (SubStr & s) const
@@ -256,6 +248,19 @@ RegExp * Scanner::invToRE (SubStr & s) const
 	return matchSymbolRange(r);
 }
 
+RegExp * Scanner::mkDiff (RegExp * e1, RegExp * e2) const
+{
+	MatchOp * m1 = dynamic_cast<MatchOp *> (e1);
+	MatchOp * m2 = dynamic_cast<MatchOp *> (e2);
+	if (m1 == NULL || m2 == NULL)
+	{
+		fatal("can only difference char sets");
+	}
+	Range * r = range_diff (m1->match, m2->match);
+
+	return matchSymbolRange (r);
+}
+
 RegExp * Scanner::mkDot() const
 {
 	Range * full = encoding.fullRange();
diff --git a/re2c/src/ir/regexp/regexp.h b/re2c/src/ir/regexp/regexp.h
index 758d3b4f..a5323146 100644
--- a/re2c/src/ir/regexp/regexp.h
+++ b/re2c/src/ir/regexp/regexp.h
@@ -69,7 +69,6 @@ public:
 RegExp * doAlt (RegExp * e1, RegExp * e2);
 RegExp * mkAlt (RegExp * e1, RegExp * e2);
 RegExp * doCat (RegExp * e1, RegExp * e2);
-RegExp * mkDiff (RegExp * e1, RegExp * e2);
 
 } // end namespace re2c
 
diff --git a/re2c/src/parse/parser.ypp b/re2c/src/parse/parser.ypp
index 114ad940..14c3f941 100644
--- a/re2c/src/parse/parser.ypp
+++ b/re2c/src/parse/parser.ypp
@@ -404,11 +404,7 @@ diff:
 		}
 	|	diff '\\' term
 		{
-			$$ = mkDiff($1, $3);
-			if(!$$)
-			{
-				in->fatal("can only difference char sets");
-			}
+			$$ = in->mkDiff($1, $3);
 		}
 ;
 
diff --git a/re2c/src/parse/scanner.h b/re2c/src/parse/scanner.h
index 51236b5a..21e98e91 100644
--- a/re2c/src/parse/scanner.h
+++ b/re2c/src/parse/scanner.h
@@ -93,6 +93,7 @@ public:
 	RegExp * strToCaseInsensitiveRE (SubStr & s) const;
 	RegExp * ranToRE (SubStr & s) const;
 	RegExp * invToRE (SubStr & s) const;
+	RegExp * mkDiff (RegExp * e1, RegExp * e2) const;
 	RegExp * mkDot () const;
 	RegExp * mkDefault () const;
 
diff --git a/re2c/test/bug61.i.c b/re2c/test/bug61.i.c
new file mode 100644
index 00000000..453370d6
--- /dev/null
+++ b/re2c/test/bug61.i.c
@@ -0,0 +1,21 @@
+/* Generated by re2c */
+
+{
+	YYCTYPE yych;
+	{}
+}
+
+
+
+{
+	YYCTYPE yych;
+	{}
+}
+
+
+
+{
+	YYCTYPE yych;
+	{}
+}
+
diff --git a/re2c/test/bug61.i.re b/re2c/test/bug61.i.re
new file mode 100644
index 00000000..4937d7da
--- /dev/null
+++ b/re2c/test/bug61.i.re
@@ -0,0 +1,11 @@
+/*!re2c
+    [] {}
+*/
+
+/*!re2c
+    [^\x00-\xFF] {}
+*/
+
+/*!re2c
+    [\x00-\xFF]\[\x00-\xFF] {}
+*/
diff --git a/re2c/test/bug61a.i.c b/re2c/test/bug61a.i.c
new file mode 100644
index 00000000..39becbe0
--- /dev/null
+++ b/re2c/test/bug61a.i.c
@@ -0,0 +1 @@
+re2c: error: line 14, column 31: can only difference char sets
diff --git a/re2c/test/bug61a.i.re b/re2c/test/bug61a.i.re
new file mode 100644
index 00000000..0e00561e
--- /dev/null
+++ b/re2c/test/bug61a.i.re
@@ -0,0 +1,15 @@
+/*!re2c
+    [] {}
+*/
+
+/*!re2c
+    [^\x00-\xFF] {}
+*/
+
+/*!re2c
+    [\x00-\xFF]\[\x00-\xFF] {}
+*/
+
+/*!re2c
+    [^\x00-\xFF]\[^\x00-\xFF] {}
+*/