[RegexDiff X64] [stephentoub] Remove capture groups from negative lookarounds #1285
Description
Job completed in 17 minutes 34 seconds (remote runner delay: 44 seconds).
dotnet/runtime#118084
Using arguments: regexdiff
70 out of 18857 patterns have generated source code changes.
Examples of GeneratedRegex source diffs
"&(?!(amp;)|(lt;)|(gt;)|(quot;))" (1847 uses)
[GeneratedRegex("&(?!(amp;)|(lt;)|(gt;)|(quot;))", RegexOptions.IgnoreCase)] /// ○ Match '&'.<br/>
/// ○ Zero-width negative lookahead.<br/>
/// ○ Match with 4 alternative expressions, atomically.<br/>
- /// ○ 1st capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Aa].<br/>
/// ○ Match a character in the set [Mm].<br/>
/// ○ Match a character in the set [Pp].<br/>
/// ○ Match ';'.<br/>
- /// ○ 2nd capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Ll].<br/>
/// ○ Match a character in the set [Tt].<br/>
/// ○ Match ';'.<br/>
- /// ○ 3rd capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Gg].<br/>
/// ○ Match a character in the set [Tt].<br/>
/// ○ Match ';'.<br/>
- /// ○ 4th capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Qq].<br/>
/// ○ Match a character in the set [Uu].<br/>
/// ○ Match a character in the set [Oo].<br/>
{
int pos = base.runtextpos;
int matchStart = pos;
- int capture_starting_pos = 0;
- int capture_starting_pos1 = 0;
- int capture_starting_pos2 = 0;
- int capture_starting_pos3 = 0;
- int negativelookahead__capture_pos = 0;
ReadOnlySpan<char> slice = inputSpan.Slice(pos);
// Match '&'.
if (slice.IsEmpty || slice[0] != '&')
{
- UncaptureUntil(0);
return false; // The input didn't match.
}
base.CheckTimeout();
}
- negativelookahead__capture_pos = base.Crawlpos();
// Match with 4 alternative expressions, atomically.
{
if ((uint)slice.Length < 2)
switch (slice[1])
{
case 'A' or 'a':
- // 1st capture group.
+
+ if ((uint)slice.Length < 5 ||
+ !slice.Slice(2).StartsWith("mp;", StringComparison.OrdinalIgnoreCase)) // Match the string "mp;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos = pos;
-
- if ((uint)slice.Length < 4 ||
- !slice.StartsWith("amp;", StringComparison.OrdinalIgnoreCase)) // Match the string "amp;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 4;
- slice = inputSpan.Slice(pos);
- base.Capture(1, capture_starting_pos, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 5;
+ slice = inputSpan.Slice(pos);
break;
case 'L' or 'l':
- // 2nd capture group.
+
+ if ((uint)slice.Length < 4 ||
+ !slice.Slice(2).StartsWith("t;", StringComparison.OrdinalIgnoreCase)) // Match the string "t;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos1 = pos;
-
- if ((uint)slice.Length < 3 ||
- !slice.StartsWith("lt;", StringComparison.OrdinalIgnoreCase)) // Match the string "lt;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 3;
- slice = inputSpan.Slice(pos);
- base.Capture(2, capture_starting_pos1, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 4;
+ slice = inputSpan.Slice(pos);
break;
case 'G' or 'g':
- // 3rd capture group.
+
+ if ((uint)slice.Length < 4 ||
+ !slice.Slice(2).StartsWith("t;", StringComparison.OrdinalIgnoreCase)) // Match the string "t;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos2 = pos;
-
- if ((uint)slice.Length < 3 ||
- !slice.StartsWith("gt;", StringComparison.OrdinalIgnoreCase)) // Match the string "gt;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 3;
- slice = inputSpan.Slice(pos);
- base.Capture(3, capture_starting_pos2, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 4;
+ slice = inputSpan.Slice(pos);
break;
case 'Q' or 'q':
- // 4th capture group.
+
+ if ((uint)slice.Length < 6 ||
+ !slice.Slice(2).StartsWith("uot;", StringComparison.OrdinalIgnoreCase)) // Match the string "uot;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos3 = pos;
-
- if ((uint)slice.Length < 5 ||
- !slice.StartsWith("quot;", StringComparison.OrdinalIgnoreCase)) // Match the string "quot;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 5;
- slice = inputSpan.Slice(pos);
- base.Capture(4, capture_starting_pos3, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 6;
+ slice = inputSpan.Slice(pos);
break;
default:
}
}
- UncaptureUntil(0);
return false; // The input didn't match.
NegativeLookaroundMatch:
pos = negativelookahead__starting_pos;
slice = inputSpan.Slice(pos);
- UncaptureUntil(negativelookahead__capture_pos);
}
// The input matched.
base.runtextpos = pos;
base.Capture(0, matchStart, pos);
return true;
-
- // <summary>Undo captures until it reaches the specified capture position.</summary>
- [MethodImpl(MethodImplOptions.AggressiveInlining)]
- void UncaptureUntil(int capturePosition)
- {
- while (base.Crawlpos() > capturePosition)
- {
- base.Uncapture();
- }
- }
}
}
}"&(?!(amp;)|(lt;)|(gt;)|(quot;)|(nbsp;)|(reg;))" (783 uses)
[GeneratedRegex("&(?!(amp;)|(lt;)|(gt;)|(quot;)|(nbsp;)|(reg;))", RegexOptions.IgnoreCase)] /// ○ Match '&'.<br/>
/// ○ Zero-width negative lookahead.<br/>
/// ○ Match with 6 alternative expressions, atomically.<br/>
- /// ○ 1st capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Aa].<br/>
/// ○ Match a character in the set [Mm].<br/>
/// ○ Match a character in the set [Pp].<br/>
/// ○ Match ';'.<br/>
- /// ○ 2nd capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Ll].<br/>
/// ○ Match a character in the set [Tt].<br/>
/// ○ Match ';'.<br/>
- /// ○ 3rd capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Gg].<br/>
/// ○ Match a character in the set [Tt].<br/>
/// ○ Match ';'.<br/>
- /// ○ 4th capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Qq].<br/>
/// ○ Match a character in the set [Uu].<br/>
/// ○ Match a character in the set [Oo].<br/>
/// ○ Match a character in the set [Tt].<br/>
/// ○ Match ';'.<br/>
- /// ○ 5th capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Nn].<br/>
/// ○ Match a character in the set [Bb].<br/>
/// ○ Match a character in the set [Ss].<br/>
/// ○ Match a character in the set [Pp].<br/>
/// ○ Match ';'.<br/>
- /// ○ 6th capture group.<br/>
+ /// ○ Match a sequence of expressions.<br/>
/// ○ Match a character in the set [Rr].<br/>
/// ○ Match a character in the set [Ee].<br/>
/// ○ Match a character in the set [Gg].<br/>
{
int pos = base.runtextpos;
int matchStart = pos;
- int capture_starting_pos = 0;
- int capture_starting_pos1 = 0;
- int capture_starting_pos2 = 0;
- int capture_starting_pos3 = 0;
- int capture_starting_pos4 = 0;
- int capture_starting_pos5 = 0;
- int negativelookahead__capture_pos = 0;
ReadOnlySpan<char> slice = inputSpan.Slice(pos);
// Match '&'.
if (slice.IsEmpty || slice[0] != '&')
{
- UncaptureUntil(0);
return false; // The input didn't match.
}
base.CheckTimeout();
}
- negativelookahead__capture_pos = base.Crawlpos();
// Match with 6 alternative expressions, atomically.
{
if ((uint)slice.Length < 2)
switch (slice[1])
{
case 'A' or 'a':
- // 1st capture group.
+
+ if ((uint)slice.Length < 5 ||
+ !slice.Slice(2).StartsWith("mp;", StringComparison.OrdinalIgnoreCase)) // Match the string "mp;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos = pos;
-
- if ((uint)slice.Length < 4 ||
- !slice.StartsWith("amp;", StringComparison.OrdinalIgnoreCase)) // Match the string "amp;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 4;
- slice = inputSpan.Slice(pos);
- base.Capture(1, capture_starting_pos, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 5;
+ slice = inputSpan.Slice(pos);
break;
case 'L' or 'l':
- // 2nd capture group.
+
+ if ((uint)slice.Length < 4 ||
+ !slice.Slice(2).StartsWith("t;", StringComparison.OrdinalIgnoreCase)) // Match the string "t;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos1 = pos;
-
- if ((uint)slice.Length < 3 ||
- !slice.StartsWith("lt;", StringComparison.OrdinalIgnoreCase)) // Match the string "lt;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 3;
- slice = inputSpan.Slice(pos);
- base.Capture(2, capture_starting_pos1, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 4;
+ slice = inputSpan.Slice(pos);
break;
case 'G' or 'g':
- // 3rd capture group.
+
+ if ((uint)slice.Length < 4 ||
+ !slice.Slice(2).StartsWith("t;", StringComparison.OrdinalIgnoreCase)) // Match the string "t;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos2 = pos;
-
- if ((uint)slice.Length < 3 ||
- !slice.StartsWith("gt;", StringComparison.OrdinalIgnoreCase)) // Match the string "gt;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 3;
- slice = inputSpan.Slice(pos);
- base.Capture(3, capture_starting_pos2, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 4;
+ slice = inputSpan.Slice(pos);
break;
case 'Q' or 'q':
- // 4th capture group.
+
+ if ((uint)slice.Length < 6 ||
+ !slice.Slice(2).StartsWith("uot;", StringComparison.OrdinalIgnoreCase)) // Match the string "uot;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos3 = pos;
-
- if ((uint)slice.Length < 5 ||
- !slice.StartsWith("quot;", StringComparison.OrdinalIgnoreCase)) // Match the string "quot;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 5;
- slice = inputSpan.Slice(pos);
- base.Capture(4, capture_starting_pos3, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 6;
+ slice = inputSpan.Slice(pos);
break;
case 'N' or 'n':
- // 5th capture group.
+
+ if ((uint)slice.Length < 6 ||
+ !slice.Slice(2).StartsWith("bsp;", StringComparison.OrdinalIgnoreCase)) // Match the string "bsp;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos4 = pos;
-
- if ((uint)slice.Length < 5 ||
- !slice.StartsWith("nbsp;", StringComparison.OrdinalIgnoreCase)) // Match the string "nbsp;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 5;
- slice = inputSpan.Slice(pos);
- base.Capture(5, capture_starting_pos4, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 6;
+ slice = inputSpan.Slice(pos);
break;
case 'R' or 'r':
- // 6th capture group.
+
+ if ((uint)slice.Length < 5 ||
+ !slice.Slice(2).StartsWith("eg;", StringComparison.OrdinalIgnoreCase)) // Match the string "eg;" (ordinal case-insensitive)
{
- pos++;
- slice = inputSpan.Slice(pos);
- capture_starting_pos5 = pos;
-
- if ((uint)slice.Length < 4 ||
- !slice.StartsWith("reg;", StringComparison.OrdinalIgnoreCase)) // Match the string "reg;" (ordinal case-insensitive)
- {
- goto NegativeLookaroundMatch;
- }
-
- pos += 4;
- slice = inputSpan.Slice(pos);
- base.Capture(6, capture_starting_pos5, pos);
+ goto NegativeLookaroundMatch;
}
+ pos += 5;
+ slice = inputSpan.Slice(pos);
break;
default:
}
}
- UncaptureUntil(0);
return false; // The input didn't match.
NegativeLookaroundMatch:
pos = negativelookahead__starting_pos;
slice = inputSpan.Slice(pos);
- UncaptureUntil(negativelookahead__capture_pos);
}
// The input matched.
base.runtextpos = pos;
base.Capture(0, matchStart, pos);
return true;
-
- // <summary>Undo captures until it reaches the specified capture position.</summary>
- [MethodImpl(MethodImplOptions.AggressiveInlining)]
- void UncaptureUntil(int capturePosition)
- {
- while (base.Crawlpos() > capturePosition)
- {
- base.Uncapture();
- }
- }
}
}
}"(?!(^[A-Z]))([A-Z])" (70 uses)
[GeneratedRegex("(?!(^[A-Z]))([A-Z])")] /// Explanation:<br/>
/// <code>
/// ○ Zero-width negative lookahead.<br/>
- /// ○ 1st capture group.<br/>
- /// ○ Match if at the beginning of the string.<br/>
- /// ○ Match a character in the set [A-Z].<br/>
+ /// ○ Match if at the beginning of the string.<br/>
+ /// ○ Match a character in the set [A-Z].<br/>
/// ○ 2nd capture group.<br/>
/// ○ Match a character in the set [A-Z].<br/>
/// </code>
int pos = base.runtextpos;
int matchStart = pos;
int capture_starting_pos = 0;
- int capture_starting_pos1 = 0;
- int negativelookahead__capture_pos = 0;
ReadOnlySpan<char> slice = inputSpan.Slice(pos);
// Zero-width negative lookahead.
base.CheckTimeout();
}
- negativelookahead__capture_pos = base.Crawlpos();
- // 1st capture group.
+ // Match if at the beginning of the string.
+ if (pos != 0)
{
- capture_starting_pos = pos;
-
- // Match if at the beginning of the string.
- if (pos != 0)
- {
- goto NegativeLookaroundMatch;
- }
-
- // Match a character in the set [A-Z].
- if (slice.IsEmpty || !char.IsAsciiLetterUpper(slice[0]))
- {
- goto NegativeLookaroundMatch;
- }
-
- pos++;
- slice = inputSpan.Slice(pos);
- base.Capture(1, capture_starting_pos, pos);
+ goto NegativeLookaroundMatch;
+ }
+
+ // Match a character in the set [A-Z].
+ if (slice.IsEmpty || !char.IsAsciiLetterUpper(slice[0]))
+ {
+ goto NegativeLookaroundMatch;
}
UncaptureUntil(0);
NegativeLookaroundMatch:
pos = negativelookahead__starting_pos;
slice = inputSpan.Slice(pos);
- UncaptureUntil(negativelookahead__capture_pos);
}
// 2nd capture group.
{
- capture_starting_pos1 = pos;
+ capture_starting_pos = pos;
// Match a character in the set [A-Z].
if (slice.IsEmpty || !char.IsAsciiLetterUpper(slice[0]))
pos++;
slice = inputSpan.Slice(pos);
- base.Capture(2, capture_starting_pos1, pos);
+ base.Capture(2, capture_starting_pos, pos);
}
// The input matched.For more diff examples, see https://gist.github.com/MihuBot/92ee92ddd485ed9c74d00540c589475f
Total bytes of base: 54274946
Total bytes of diff: 54227335
Total bytes of delta: -47611 (-0.09 % of base)
Total relative delta: -14.88
diff is an improvement.
relative diff is an improvement.
For a list of JIT diff improvements, see Improvements.md
Sample source code for further analysis
const string JsonPath = "RegexResults-1285.json";
if (!File.Exists(JsonPath))
{
await using var archiveStream = await new HttpClient().GetStreamAsync("https://mihubot.xyz/r/E2kAZabA");
using var archive = new ZipArchive(archiveStream, ZipArchiveMode.Read);
archive.Entries.First(e => e.Name == "Results.json").ExtractToFile(JsonPath);
}
using FileStream jsonFileStream = File.OpenRead(JsonPath);
RegexEntry[] entries = JsonSerializer.Deserialize<RegexEntry[]>(jsonFileStream, new JsonSerializerOptions { IncludeFields = true })!;
Console.WriteLine($"Working with {entries.Length} patterns");
record KnownPattern(string Pattern, RegexOptions Options, int Count);
sealed class RegexEntry
{
public required KnownPattern Regex { get; set; }
public required string MainSource { get; set; }
public required string PrSource { get; set; }
public string? FullDiff { get; set; }
public string? ShortDiff { get; set; }
public (string Name, string Values)[]? SearchValuesOfChar { get; set; }
public (string[] Values, StringComparison ComparisonType)[]? SearchValuesOfString { get; set; }
}Artifacts:
- ShortExampleDiffs.md (27 KB)
- LongExampleDiffs.md (901 KB)
- Results.zip (42 MB)
- JitAnalyzeSummary.txt (46 KB)
- jit-diffs.zip (391 MB)
- JitDiffImprovements.md (896 KB)
- LongJitDiffImprovements.md (10 MB)