Fix overaggressive CanBeMadeAtomic check for Set + Notone (#33409)

We're erroneously converting a set loop to be atomic when it's followed by a notone where the notone's character is in the set. But if we for example have `[ab]*[^a]`, we can't make the loop atomic, because the `[ab]*` can actually give back something (a `b`) that the `[^a]` will match. The fix is simply to delete the erroneous, overaggressive checks.
dotnet · Mar 10, 2020 · d12f79a · d12f79a
1 parent fb01e2c
commit d12f79a
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 4 deletions.
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
@@ -1541,10 +1541,6 @@ private static bool CanBeMadeAtomic(RegexNode node, RegexNode subsequent, uint m
                         case Onelazy when subsequent.M > 0 && !RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
                         case Oneloop when subsequent.M > 0 && !RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
                         case Oneloopatomic when subsequent.M > 0 && !RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
-                        case Notone when RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
-                        case Notonelazy when subsequent.M > 0 && RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
-                        case Notoneloop when subsequent.M > 0 && RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
-                        case Notoneloopatomic when subsequent.M > 0 && RegexCharClass.CharInClass(subsequent.Ch, node.Str!):
                         case Multi when !RegexCharClass.CharInClass(subsequent.Str![0], node.Str!):
                         case Set when !RegexCharClass.MayOverlap(node.Str!, subsequent.Str!):
                         case Setlazy when subsequent.M > 0 && !RegexCharClass.MayOverlap(node.Str!, subsequent.Str!):

diff --git a/src/libraries/System.Text.RegularExpressions/tests/Regex.Groups.Tests.cs b/src/libraries/System.Text.RegularExpressions/tests/Regex.Groups.Tests.cs
@@ -733,6 +733,8 @@ public static IEnumerable<object[]> Groups_Basic_TestData()
             yield return new object[] { null, @"(?:a{2}?){3}?", "aaaaaaaaa", RegexOptions.None, new string[] { "aaaaaa" } };
             yield return new object[] { null, @"(?:(?:[ab]c[de]f){3}){2}", "acdfbcdfacefbcefbcefbcdfacdef", RegexOptions.None, new string[] { "acdfbcdfacefbcefbcefbcdf" } };
             yield return new object[] { null, @"(?:(?:[ab]c[de]f){3}hello){2}", "aaaaaacdfbcdfacefhellobcefbcefbcdfhellooooo", RegexOptions.None, new string[] { "acdfbcdfacefhellobcefbcefbcdfhello" } };
+            yield return new object[] { null, @"CN=(.*[^,]+).*", "CN=localhost", RegexOptions.Singleline, new string[] { "CN=localhost", "localhost" } };
+
             // Nested atomic
             yield return new object[] { null, @"(?>abc[def]gh(i*))", "123abceghiii456", RegexOptions.None, new string[] { "abceghiii", "iii" } };
 

diff --git a/src/libraries/System.Text.RegularExpressions/tests/RegexReductionTests.cs b/src/libraries/System.Text.RegularExpressions/tests/RegexReductionTests.cs
@@ -375,6 +375,10 @@ public void PatternsReduceIdentically(string pattern1, string pattern2)
         [InlineData("a*a*?", "a*")]
         [InlineData("a*?a*", "a*")]
         [InlineData("a*[^a]*", "a*")]
+        [InlineData("[ab]*[^a]", "(?>[ab]*)[^a]")]
+        [InlineData("[ab]*[^a]*", "(?>[ab]*)[^a]*")]
+        [InlineData("[ab]*[^a]*?", "(?>[ab]*)[^a]*?")]
+        [InlineData("[ab]*(?>[^a]*)", "(?>[ab]*)(?>[^a]*)")]
         [InlineData("[^a]*a*", "a*")]
         [InlineData("a{2147483646}a", "a{2147483647}")]
         [InlineData("a{2147483647}a", "a{2147483647}")]