From: Mauri de Souza Meneguzzo <mauri870@gmail.com>
Date: Sat, 29 Jul 2023 20:26:00 +0000 (+0000)
Subject: regexp/syntax: accept (?<name>...) syntax as valid capture
X-Git-Tag: go1.22rc1~1467
X-Git-Url: http://www.git.cypherpunks.su/?a=commitdiff_plain;h=ee61186b3301bb1c8610c0925fffd89e061909bb;p=gostls13.git

regexp/syntax: accept (?<name>...) syntax as valid capture

Currently the only named capture supported by regexp is (?P<name>a).

The syntax (?<name>a) is also widely used and there is currently an effort from
 the Rust regex and RE2 teams to also accept this syntax.

Fixes #58458

Change-Id: If22d44d3a5c4e8133ec68238ab130c151ca7c5c5
GitHub-Last-Rev: 31b50e6ab40cfb0f36df6f570525657d4680017f
GitHub-Pull-Request: golang/go#61624
Reviewed-on: https://go-review.googlesource.com/c/go/+/513838
Auto-Submit: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
---

diff --git a/src/regexp/syntax/doc.go b/src/regexp/syntax/doc.go
index f6a4b43f7a..eb8a971c73 100644
--- a/src/regexp/syntax/doc.go
+++ b/src/regexp/syntax/doc.go
@@ -56,6 +56,7 @@ Grouping:
 
 	(re)           numbered capturing group (submatch)
 	(?P<name>re)   named & numbered capturing group (submatch)
+	(?<name>re)    named & numbered capturing group (submatch)
 	(?:re)         non-capturing group
 	(?flags)       set flags within current group; non-capturing
 	(?flags:re)    set flags during re; non-capturing
diff --git a/src/regexp/syntax/parse.go b/src/regexp/syntax/parse.go
index accee9ab08..a4ccfe3bdb 100644
--- a/src/regexp/syntax/parse.go
+++ b/src/regexp/syntax/parse.go
@@ -1159,9 +1159,18 @@ func (p *parser) parsePerlFlags(s string) (rest string, err error) {
 	// support all three as well. EcmaScript 4 uses only the Python form.
 	//
 	// In both the open source world (via Code Search) and the
-	// Google source tree, (?P<expr>name) is the dominant form,
-	// so that's the one we implement. One is enough.
-	if len(t) > 4 && t[2] == 'P' && t[3] == '<' {
+	// Google source tree, (?P<expr>name) and (?<expr>name) are the
+	// dominant forms of named captures and both are supported.
+	startsWithP := len(t) > 4 && t[2] == 'P' && t[3] == '<'
+	startsWithName := len(t) > 3 && t[2] == '<'
+
+	if startsWithP || startsWithName {
+		// position of expr start
+		exprStartPos := 4
+		if startsWithName {
+			exprStartPos = 3
+		}
+
 		// Pull out name.
 		end := strings.IndexRune(t, '>')
 		if end < 0 {
@@ -1171,8 +1180,8 @@ func (p *parser) parsePerlFlags(s string) (rest string, err error) {
 			return "", &Error{ErrInvalidNamedCapture, s}
 		}
 
-		capture := t[:end+1] // "(?P<name>"
-		name := t[4:end]     // "name"
+		capture := t[:end+1]        // "(?P<name>" or "(?<name>"
+		name := t[exprStartPos:end] // "name"
 		if err = checkUTF8(name); err != nil {
 			return "", err
 		}
diff --git a/src/regexp/syntax/parse_test.go b/src/regexp/syntax/parse_test.go
index 67e3c5622a..d7999046e0 100644
--- a/src/regexp/syntax/parse_test.go
+++ b/src/regexp/syntax/parse_test.go
@@ -160,6 +160,7 @@ var parseTests = []parseTest{
 
 	// Test named captures
 	{`(?P<name>a)`, `cap{name:lit{a}}`},
+	{`(?<name>a)`, `cap{name:lit{a}}`},
 
 	// Case-folded literals
 	{`[Aa]`, `litfold{A}`},
@@ -482,6 +483,11 @@ var invalidRegexps = []string{
 	`(?P<name`,
 	`(?P<x y>a)`,
 	`(?P<>a)`,
+	`(?<name>a`,
+	`(?<name>`,
+	`(?<name`,
+	`(?<x y>a)`,
+	`(?<>a)`,
 	`[a-Z]`,
 	`(?i)[a-Z]`,
 	`\Q\E*`,