net.urllib: fix parse of url relative address (fix vlang#21061) (vlan…

…g#23180)
kimshrier · Dec 16, 2024 · 03f3f7a · 03f3f7a
1 parent b1502d4
commit 03f3f7a
Show file tree

Hide file tree

Showing 2 changed files with 19 additions and 8 deletions.
diff --git a/vlib/net/urllib/urllib.v b/vlib/net/urllib/urllib.v
@@ -436,11 +436,15 @@ fn split(s string, sep u8, cutc bool) (string, string) {
 pub fn parse(rawurl string) !URL {
 	// Cut off #frag
 	u, frag := split(rawurl, `#`, true)
-	mut url := parse_url(u, false) or { return error(error_msg(err_msg_parse, u)) }
+	mut url := parse_url(u, false) or {
+		return error(error_msg(err_msg_parse + '[${err.msg()}]', u))
+	}
 	if frag == '' {
 		return url
 	}
-	f := unescape(frag, .encode_fragment) or { return error(error_msg(err_msg_parse, u)) }
+	f := unescape(frag, .encode_fragment) or {
+		return error(error_msg(err_msg_parse + '[${err.msg()}]', u))
+	}
 	url.fragment = f
 	return url
 }
@@ -503,12 +507,13 @@ fn parse_url(rawurl string, via_request bool) !URL {
 		// RFC 3986, §3.3:
 		// In addition, a URI reference (Section 4.1) may be a relative-path reference,
 		// in which case the first path segment cannot contain a colon (':') character.
-		colon := rest.index(':') or { return error('there should be a : in the URL') }
-		slash := rest.index('/') or { return error('there should be a / in the URL') }
-		if colon >= 0 && (slash < 0 || colon < slash) {
-			// First path segment has colon. Not allowed in relative URL.
-			return error(error_msg('parse_url: first path segment in URL cannot contain colon',
-				''))
+		if colon := rest.index(':') {
+			slash := rest.index('/') or { return error('there should be a / in the URL') }
+			if colon >= 0 && (slash < 0 || colon < slash) {
+				// First path segment has colon. Not allowed in relative URL.
+				return error(error_msg('parse_url: first path segment in URL cannot contain colon',
+					''))
+			}
 		}
 	}
 	if ((url.scheme != '' || !via_request) && !rest.starts_with('///')) && rest.starts_with('//')

diff --git a/vlib/net/urllib/urllib_test.v b/vlib/net/urllib/urllib_test.v
@@ -102,6 +102,9 @@ fn test_parse() {
 		'foo://example.com:8042/over/there?name=ferret#nose',
 		'ftp://2001:0db8:85a3:0000:0000:8a2e:0370:7334/path/file.txt',
 		'ws://[2001:0db8:85a3:0000:0000:8a2e:0370:7334]:4000',
+		'/2000k/hls/mixed.m3u8',
+		'2000k/hls/mixed.m3u8',
+		'./2000k:hls/mixed.m3u8',
 	]
 	for url in urls {
 		urllib.parse(url)!
@@ -112,6 +115,9 @@ fn test_parse() {
 	assert u.scheme == 'https' && u.hostname() == 'www.mydomain.com' && u.port() == '8080'
 		&& u.path == '/som/url' && u.fragment == 'testfragment' && u.user.username == 'joe'
 		&& u.user.password == 'pass'
+
+	v := urllib.parse('https://vip.ffzy-online4.com/20230205/6094_d2720761/index.m3u8')!.resolve_reference(urllib.parse('2000k/hls/mixed.m3u8')!)!
+	assert v.str() == 'https://vip.ffzy-online4.com/20230205/6094_d2720761/2000k/hls/mixed.m3u8'
 }
 
 fn test_parse_authority() {