Title: [255584] trunk
- Revision
- 255584
- Author
- [email protected]
- Date
- 2020-02-03 13:23:20 -0800 (Mon, 03 Feb 2020)
Log Message
\0 identity escapes should be syntax errors in Unicode patterns only
https://bugs.webkit.org/show_bug.cgi?id=207114
Reviewed by Darin Adler.
JSTests:
* test262/expectations.yaml: Mark 2 test cases as passing.
Source/_javascript_Core:
This change adds a separate check for null character because `strchr`
always returns a non-null pointer when called with '\0' as second argument.
Grammar: https://tc39.es/ecma262/#prod-annexB-IdentityEscape
(/u flag precludes the use of SourceCharacterIdentityEscape)
* yarr/YarrParser.h:
(JSC::Yarr::Parser::isIdentityEscapeAnError):
Modified Paths
Diff
Modified: trunk/JSTests/ChangeLog (255583 => 255584)
--- trunk/JSTests/ChangeLog 2020-02-03 21:12:43 UTC (rev 255583)
+++ trunk/JSTests/ChangeLog 2020-02-03 21:23:20 UTC (rev 255584)
@@ -1,3 +1,12 @@
+2020-02-03 Alexey Shvayka <[email protected]>
+
+ \0 identity escapes should be syntax errors in Unicode patterns only
+ https://bugs.webkit.org/show_bug.cgi?id=207114
+
+ Reviewed by Darin Adler.
+
+ * test262/expectations.yaml: Mark 2 test cases as passing.
+
2020-02-01 Alexey Shvayka <[email protected]>
Non-alphabetical \c escapes should be syntax errors in Unicode patterns only
Modified: trunk/JSTests/test262/expectations.yaml (255583 => 255584)
--- trunk/JSTests/test262/expectations.yaml 2020-02-03 21:12:43 UTC (rev 255583)
+++ trunk/JSTests/test262/expectations.yaml 2020-02-03 21:23:20 UTC (rev 255584)
@@ -1573,9 +1573,6 @@
test/built-ins/RegExp/prototype/unicode/cross-realm.js:
default: 'Test262Error: cross-realm RegExp.prototype Expected a TypeError to be thrown but no exception was thrown at all'
strict mode: 'Test262Error: cross-realm RegExp.prototype Expected a TypeError to be thrown but no exception was thrown at all'
-test/built-ins/RegExp/unicode_restricted_identity_escape.js:
- default: "Test262Error: Invalid IdentityEscape in AtomEscape: '\\"
- strict mode: "Test262Error: Invalid IdentityEscape in AtomEscape: '\\"
test/built-ins/RegExp/unicode_restricted_identity_escape_alpha.js:
default: "Test262Error: IdentityEscape in AtomEscape: 'k' Expected a SyntaxError to be thrown but no exception was thrown at all"
strict mode: "Test262Error: IdentityEscape in AtomEscape: 'k' Expected a SyntaxError to be thrown but no exception was thrown at all"
Modified: trunk/Source/_javascript_Core/ChangeLog (255583 => 255584)
--- trunk/Source/_javascript_Core/ChangeLog 2020-02-03 21:12:43 UTC (rev 255583)
+++ trunk/Source/_javascript_Core/ChangeLog 2020-02-03 21:23:20 UTC (rev 255584)
@@ -1,3 +1,19 @@
+2020-02-03 Alexey Shvayka <[email protected]>
+
+ \0 identity escapes should be syntax errors in Unicode patterns only
+ https://bugs.webkit.org/show_bug.cgi?id=207114
+
+ Reviewed by Darin Adler.
+
+ This change adds a separate check for null character because `strchr`
+ always returns a non-null pointer when called with '\0' as second argument.
+
+ Grammar: https://tc39.es/ecma262/#prod-annexB-IdentityEscape
+ (/u flag precludes the use of SourceCharacterIdentityEscape)
+
+ * yarr/YarrParser.h:
+ (JSC::Yarr::Parser::isIdentityEscapeAnError):
+
2020-02-01 Alexey Shvayka <[email protected]>
Non-alphabetical \c escapes should be syntax errors in Unicode patterns only
Modified: trunk/Source/_javascript_Core/yarr/YarrParser.h (255583 => 255584)
--- trunk/Source/_javascript_Core/yarr/YarrParser.h 2020-02-03 21:12:43 UTC (rev 255583)
+++ trunk/Source/_javascript_Core/yarr/YarrParser.h 2020-02-03 21:23:20 UTC (rev 255584)
@@ -231,7 +231,7 @@
// For non-unicode patterns, most any character can be escaped.
bool isIdentityEscapeAnError(int ch)
{
- if (m_isUnicode && !strchr("^$\\.*+?()[]{}|/", ch)) {
+ if (m_isUnicode && (!strchr("^$\\.*+?()[]{}|/", ch) || !ch)) {
m_errorCode = ErrorCode::InvalidIdentityEscape;
return true;
}
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes
