From dcab79146b5a6a8092c5b9a730e331bb6222ea81 Mon Sep 17 00:00:00 2001
From: Delyan Angelov <delian66@gmail.com>
Date: Tue, 30 Mar 2021 20:14:10 +0300
Subject: [PATCH] net.urllib: allow `urllib.parse_query("format=%l")`

---
 vlib/net/urllib/urllib.v      | 12 +++++++++---
 vlib/net/urllib/urllib_test.v | 29 ++++++++++++++++++-----------
 2 files changed, 27 insertions(+), 14 deletions(-)

diff --git a/vlib/net/urllib/urllib.v b/vlib/net/urllib/urllib.v
index 17e354a9e7..4960c26f3d 100644
--- a/vlib/net/urllib/urllib.v
+++ b/vlib/net/urllib/urllib.v
@@ -53,8 +53,9 @@ fn should_escape(c byte, mode EncodingMode) bool {
 		// we could possibly allow, and parse will reject them if we
 		// escape them (because hosts can`t use %-encoding for
 		// ASCII bytes).
-		if
-			c in [`!`, `$`, `&`, `\\`, `(`, `)`, `*`, `+`, `,`, `;`, `=`, `:`, `[`, `]`, `<`, `>`, `"`] {
+		if c in [`!`, `$`, `&`, `\\`, `(`, `)`, `*`, `+`, `,`, `;`, `=`, `:`, `[`, `]`, `<`, `>`,
+			`"`,
+		] {
 			return false
 		}
 	}
@@ -158,6 +159,11 @@ fn unescape(s_ string, mode EncodingMode) ?string {
 				}
 				n++
 				if i + 2 >= s.len || !ishex(s[i + 1]) || !ishex(s[i + 2]) {
+					if mode == .encode_query_component && i + 1 < s.len {
+						s = s[..i] + '%25' + s[(i + 1)..]
+						i += 4 // skip the %25 and the next character
+						continue
+					}
 					s = s[i..]
 					if s.len > 3 {
 						s = s[..3]
@@ -789,7 +795,7 @@ pub fn parse_query(query string) ?Values {
 // but any errors will be silent
 fn parse_query_silent(query string) Values {
 	mut m := new_values()
-	parse_query_values(mut m, query) or { }
+	parse_query_values(mut m, query) or {}
 	return m
 }
 
diff --git a/vlib/net/urllib/urllib_test.v b/vlib/net/urllib/urllib_test.v
index 947b88cf12..2d354efda7 100644
--- a/vlib/net/urllib/urllib_test.v
+++ b/vlib/net/urllib/urllib_test.v
@@ -1,7 +1,6 @@
 // Copyright (c) 2019-2021 Alexander Medvednikov. All rights reserved.
 // Use of this source code is governed by an MIT license
 // that can be found in the LICENSE file.
-
 import net.urllib
 
 fn test_net_urllib() {
@@ -13,18 +12,14 @@ fn test_net_urllib() {
 		assert false
 		return
 	}
-	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'
+	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'
 }
 
 fn test_str() {
-	url := urllib.parse("https://en.wikipedia.org/wiki/Brazil_(1985_film)") or {
-		panic("unable to parse URL")
+	url := urllib.parse('https://en.wikipedia.org/wiki/Brazil_(1985_film)') or {
+		panic('unable to parse URL')
 	}
 	assert url.str() == 'https://en.wikipedia.org/wiki/Brazil_(1985_film)'
 }
@@ -33,6 +28,18 @@ fn test_escape_unescape() {
 	original := 'те ст: т\\%'
 	escaped := urllib.query_escape(original)
 	assert escaped == '%D1%82%D0%B5+%D1%81%D1%82%3A+%D1%82%5C%25'
-	unescaped := urllib.query_unescape(escaped) or { assert false return }
+	unescaped := urllib.query_unescape(escaped) or {
+		assert false
+		return
+	}
 	assert unescaped == original
 }
+
+fn test_parse_query() ? {
+	q1 := urllib.parse_query('format=%22%25l%3A+%25c+%25t%22') ?
+	q2 := urllib.parse_query('format="%l:+%c+%t"') ?
+	// dump(q1)
+	// dump(q2)
+	assert q1.data['format'].data == ['"%l: %c %t"']
+	assert q2.data['format'].data == ['"%l: %c %t"']
+}