Skip to content

A few minor tweaks to Regex source generator rendering #63276

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
stephentoub merged 7 commits into from
Jan 4, 2022
Merged

A few minor tweaks to Regex source generator rendering #63276

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
47eabf7
Do addition for EndZ matching at compile time
stephentoub Dec 24, 2021
a3108ee
Tweak rendering of optional loops to say "Optional" rather than "Loop…
stephentoub Dec 24, 2021
3310154
Remove "at least X" from loop description when X is 0
stephentoub Dec 24, 2021
264da06
Add a missing blank line at the beginning of a back reference
stephentoub Dec 24, 2021
215e7a6
Rename ReturnFalse to NoStartingPositionFound
stephentoub Dec 24, 2021
9c2a580
Delete stale comments
stephentoub Dec 29, 2021
11e10e7
Address PR feedback
stephentoub Jan 3, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 15 additions & 11 deletions src/libraries/System.Text.RegularExpressions/gen/RegexGenerator.Emitter.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -376,8 +376,9 @@ private static RequiredHelperFunctions EmitFindFirstChar(IndentedTextWriter writ
}
writer.WriteLine();

const string NoStartingPositionFound = "NoStartingPositionFound";
writer.WriteLine("// No starting position found");
writer.WriteLine("ReturnFalse:");
writer.WriteLine($"{NoStartingPositionFound}:");
writer.WriteLine("base.runtextpos = end;");
writer.WriteLine("return false;");

Expand All @@ -399,7 +400,7 @@ bool EmitAnchors()
additionalDeclarations.Add("int beginning = base.runtextbeg;");
using (EmitBlock(writer, "if (pos > beginning)"))
{
writer.WriteLine("goto ReturnFalse;");
writer.WriteLine($"goto {NoStartingPositionFound};");
}
writer.WriteLine("return true;");
return true;
Expand All @@ -408,7 +409,7 @@ bool EmitAnchors()
writer.WriteLine("// Start \\G anchor");
using (EmitBlock(writer, "if (pos > base.runtextstart)"))
{
writer.WriteLine("goto ReturnFalse;");
writer.WriteLine($"goto {NoStartingPositionFound};");
}
writer.WriteLine("return true;");
return true;
Expand Down Expand Up @@ -444,7 +445,7 @@ bool EmitAnchors()
writer.WriteLine("int newlinePos = global::System.MemoryExtensions.IndexOf(inputSpan.Slice(pos), '\\n');");
using (EmitBlock(writer, "if (newlinePos < 0 || newlinePos + pos + 1 > end)"))
{
writer.WriteLine("goto ReturnFalse;");
writer.WriteLine($"goto {NoStartingPositionFound};");
}
writer.WriteLine("pos = newlinePos + pos + 1;");
}
Expand Down Expand Up @@ -517,7 +518,7 @@ void EmitFixedSet()
writer.WriteLine($"int indexOfPos = {indexOf};");
using (EmitBlock(writer, "if (indexOfPos < 0)"))
{
writer.WriteLine("goto ReturnFalse;");
writer.WriteLine($"goto {NoStartingPositionFound};");
}
writer.WriteLine("i += indexOfPos;");
writer.WriteLine();
Expand All @@ -526,7 +527,7 @@ void EmitFixedSet()
{
using (EmitBlock(writer, $"if (i >= span.Length - {minRequiredLength - 1})"))
{
writer.WriteLine("goto ReturnFalse;");
writer.WriteLine($"goto {NoStartingPositionFound};");
}
writer.WriteLine();
}
Expand Down Expand Up @@ -1128,7 +1129,11 @@ void EmitBackreference(RegexNode node)

int capnum = RegexParser.MapCaptureNumber(node.M, rm.Code.Caps);

TransferSliceStaticPosToPos();
if (sliceStaticPos > 0)
{
TransferSliceStaticPosToPos();
writer.WriteLine();
}

// If the specified capture hasn't yet captured anything, fail to match... except when using RegexOptions.ECMAScript,
// in which case per ECMA 262 section 21.2.2.9 the backreference should succeed.
Expand Down Expand Up @@ -1999,7 +2004,7 @@ void EmitAnchors(RegexNode node)
break;

case RegexNode.EndZ:
writer.WriteLine($"if ({sliceSpan}.Length - 1 > {sliceStaticPos} || ({IsSliceLengthGreaterThanSliceStaticPos()} && {sliceSpan}[{sliceStaticPos}] != '\\n'))");
writer.WriteLine($"if ({sliceSpan}.Length > {sliceStaticPos + 1} || ({IsSliceLengthGreaterThanSliceStaticPos()} && {sliceSpan}[{sliceStaticPos}] != '\\n'))");
using (EmitBlock(writer, null))
{
writer.WriteLine($"goto {doneLabel};");
Expand Down Expand Up @@ -3385,7 +3390,7 @@ private static string DescribeNode(RegexNode node) =>
RegexNode.End => "Match if at the end of the string.",
RegexNode.EndZ => "Match if at the end of the string or if before an ending newline.",
RegexNode.Eol => "Match if at the end of a line.",
RegexNode.Loop or RegexNode.Lazyloop => $"Loop {DescribeLoop(node)}.",
RegexNode.Loop or RegexNode.Lazyloop => node.M == 0 && node.N == 1 ? $"Optional ({(node.Type is RegexNode.Loop ? "greedy" : "lazy")})." : $"Loop {DescribeLoop(node)}.",
RegexNode.Multi => $"Match the string {Literal(node.Str!)}.",
RegexNode.NonBoundary => $"Match if at anything other than a word boundary.",
RegexNode.NonECMABoundary => $"Match if at anything other than a word boundary (according to ECMAScript rules).",
Expand All @@ -3404,8 +3409,6 @@ private static string DescribeNode(RegexNode node) =>
RegexNode.Testref => $"Conditionally match {(node.ChildCount() == 1 ? "an expression" : "one of two expressions")} depending on whether the {DescribeNonNegative(node.M)} capture group matched.",
RegexNode.UpdateBumpalong => $"Advance the next matching position.",
_ => $"Unknown node type {node.Type}",

// Concatenation
};

/// <summary>Writes a textual description of the node tree fit for rending in source.</summary>
Expand Down Expand Up @@ -3488,6 +3491,7 @@ private static string DescribeLoop(RegexNode node)
(2, int.MaxValue) => " at least twice",
(_, int.MaxValue) => $" at least {node.M} times",
(0, 1) => ", optionally",
(0, _) => $"at most {node.N} times",
_ => $" at least {node.M} and at most {node.N} times"
};

Expand Down