From 19646015efe70e5decdceb659d5268430d91eff1 Mon Sep 17 00:00:00 2001
From: Mauri de Souza Meneguzzo <mauri870@gmail.com>
Date: Thu, 27 Jul 2023 22:25:16 -0300
Subject: [PATCH 1/2] regexp/syntax: accept (?<name>...) 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
---
 src/regexp/syntax/parse.go      | 19 ++++++++++++++-----
 src/regexp/syntax/parse_test.go |  6 ++++++
 2 files changed, 20 insertions(+), 5 deletions(-)

diff --git a/src/regexp/syntax/parse.go b/src/regexp/syntax/parse.go
index accee9ab089edc..a4ccfe3bdbc75d 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 67e3c5622a0fef..d7999046e0eaaa 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*`,

From 31b50e6ab40cfb0f36df6f570525657d4680017f Mon Sep 17 00:00:00 2001
From: Mauri de Souza Meneguzzo <mauri870@gmail.com>
Date: Sat, 29 Jul 2023 17:22:33 -0300
Subject: [PATCH 2/2] add new regex syntax to doc

---
 src/regexp/syntax/doc.go | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/regexp/syntax/doc.go b/src/regexp/syntax/doc.go
index f6a4b43f7aebb1..eb8a971c734874 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