[release-branch.go1.20] html/template: handle all JS whitespace characters

Rather than just a small set. Character class as defined by \s [0].

Thanks to Juho Nurminen of Mattermost for reporting this.

For #59721
Fixes #59814
Fixes CVE-2023-24540

[0] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions/Character_Classes

Change-Id: I56d4fa1ef08125b417106ee7dbfb5b0923b901ba
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/1821459
Reviewed-by: Julie Qiu <julieqiu@google.com>
Run-TryBot: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/1851493
TryBot-Result: Security TryBots <security-trybots@go-security-trybots.iam.gserviceaccount.com>
Run-TryBot: Damien Neil <dneil@google.com>
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/491356
Run-TryBot: Carlos Amedee <carlos@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

2 files changed, 14 insertions, 5 deletions

diff --git a/src/html/template/js.go b/src/html/template/js.go
index fe7054efe5..4e05c14557 100644
--- a/src/html/template/js.go
+++ b/src/html/template/js.go
@@ -13,6 +13,11 @@ import (
 	"unicode/utf8"
 )
 
+// jsWhitespace contains all of the JS whitespace characters, as defined
+// by the \s character class.
+// See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_expressions/Character_classes.
+const jsWhitespace = "\f\n\r\t\v\u0020\u00a0\u1680\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u2028\u2029\u202f\u205f\u3000\ufeff"
+
 // nextJSCtx returns the context that determines whether a slash after the
 // given run of tokens starts a regular expression instead of a division
 // operator: / or /=.
@@ -26,7 +31,8 @@ import (
 // JavaScript 2.0 lexical grammar and requires one token of lookbehind:
 // https://www.mozilla.org/js/language/js20-2000-07/rationale/syntax.html
 func nextJSCtx(s []byte, preceding jsCtx) jsCtx {
-	s = bytes.TrimRight(s, "\t\n\f\r \u2028\u2029")
+	// Trim all JS whitespace characters
+	s = bytes.TrimRight(s, jsWhitespace)
 	if len(s) == 0 {
 		return preceding
 	}
diff --git a/src/html/template/js_test.go b/src/html/template/js_test.go
index eee7eb2bef..259dcfbdc5 100644
--- a/src/html/template/js_test.go
+++ b/src/html/template/js_test.go
@@ -80,14 +80,17 @@ func TestNextJsCtx(t *testing.T) {
 		{jsCtxDivOp, "0"},
 		// Dots that are part of a number are div preceders.
 		{jsCtxDivOp, "0."},
+		// Some JS interpreters treat NBSP as a normal space, so
+		// we must too in order to properly escape things.
+		{jsCtxRegexp, "=\u00A0"},
 	}
 
 	for _, test := range tests {
-		if nextJSCtx([]byte(test.s), jsCtxRegexp) != test.jsCtx {
-			t.Errorf("want %s got %q", test.jsCtx, test.s)
+		if ctx := nextJSCtx([]byte(test.s), jsCtxRegexp); ctx != test.jsCtx {
+			t.Errorf("%q: want %s got %s", test.s, test.jsCtx, ctx)
 		}
-		if nextJSCtx([]byte(test.s), jsCtxDivOp) != test.jsCtx {
-			t.Errorf("want %s got %q", test.jsCtx, test.s)
+		if ctx := nextJSCtx([]byte(test.s), jsCtxDivOp); ctx != test.jsCtx {
+			t.Errorf("%q: want %s got %s", test.s, test.jsCtx, ctx)
 		}
 	}