Avoid concatenating adjacent regex loops+strings under right-to-left (#…

…103591) The pattern gets reversed but the actual strings in multis aren't (instead the evaluation just compares them in reverse), which means optimizations based on comparing nodes in a sequence and the text of such a string either need to take that into account or be disabled for right-to-left.
dotnet · Jun 19, 2024 · 24b59d8 · 24b59d8
1 parent 83031f1
commit 24b59d8
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 1 deletion.
diff --git a/src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/RegexNode.cs b/src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/RegexNode.cs
@@ -1702,7 +1702,10 @@ static bool CanCombineCounts(int nodeMin, int nodeMax, int nextMin, int nextMax)
                             break;
 
                         // Coalescing a loop with a subsequent string
-                        case RegexNodeKind.Oneloop or RegexNodeKind.Onelazy when nextNode.Kind == RegexNodeKind.Multi && currentNode.Ch == nextNode.Str![0]:
+                        case RegexNodeKind.Oneloop or RegexNodeKind.Onelazy when
+                                nextNode.Kind == RegexNodeKind.Multi &&
+                                (nextNode.Options & RegexOptions.RightToLeft) == 0 && // RTL multi nodes don't have their text reversed, and it's not worth the code to optimize further
+                                currentNode.Ch == nextNode.Str![0]:
                             {
                                 // Determine how many of the multi's characters can be combined.
                                 // We already checked for the first, so we know it's at least one.

diff --git a/src/libraries/System.Text.RegularExpressions/tests/FunctionalTests/Regex.Match.Tests.cs b/src/libraries/System.Text.RegularExpressions/tests/FunctionalTests/Regex.Match.Tests.cs
@@ -870,6 +870,11 @@ public static IEnumerable<object[]> Match_MemberData()
                 yield return (@"\s+\d+", " asdf12 ", RegexOptions.RightToLeft, 0, 6, false, string.Empty);
                 yield return ("aaa", "aaabbb", RegexOptions.None, 3, 3, false, string.Empty);
                 yield return ("abc|def", "123def456", RegexOptions.RightToLeft | RegexOptions.IgnoreCase | RegexOptions.CultureInvariant, 0, 9, true, "def");
+                yield return (@"^says?$", "says", RegexOptions.RightToLeft, 0, 4, true, "says");
+                yield return (@"^says?$", "say", RegexOptions.RightToLeft, 0, 3, true, "say");
+                yield return (@"^say(s?)$", "says", RegexOptions.RightToLeft, 0, 4, true, "says");
+                yield return (@"^(say)s?$", "says", RegexOptions.RightToLeft, 0, 4, true, "says");
+                yield return (@"^(.+?) (says?),\s'(.+)'$", "User says, 'adventure'", RegexOptions.RightToLeft, 0, 22, true, "User says, 'adventure'");
 
                 // .* : RTL, Case-sensitive
                 yield return (@".*\nfoo", "This shouldn't match", RegexOptions.None | RegexOptions.RightToLeft, 0, 20, false, "");