[PATCH 0/1] MR585: mshtml: Unescape path before converting it to wide character string in is_gecko_path().

[Jactry Zeng (＠jactry)]

This fixes an issue when the path includes non-ASCII characters.

Signed-off-by: Jactry Zeng <jzeng(a)codeweavers.com>

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/585

[PATCH 1/1] mshtml: Unescape path before converting it to wide character string in is_gecko_path().

[Jactry Zeng]

From: Jactry Zeng <jzeng(a)codeweavers.com>

This fixes an issue when the path includes non-ASCII characters.

Signed-off-by: Jactry Zeng <jzeng(a)codeweavers.com>
---
 dlls/mshtml/nsembed.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/dlls/mshtml/nsembed.c b/dlls/mshtml/nsembed.c
index c7683a6d06d..2918362c9c0 100644
--- a/dlls/mshtml/nsembed.c
+++ b/dlls/mshtml/nsembed.c
@@ -1285,19 +1285,21 @@ void close_gecko(void)
 
 BOOL is_gecko_path(const char *path)
 {
-    WCHAR *buf, *ptr;
+    char unescaped_path[INTERNET_MAX_URL_LENGTH], *ptr;
+    WCHAR *buf;
     BOOL ret;
 
-    buf = heap_strdupUtoW(path);
-    if(!buf || lstrlenW(buf) < gecko_path_len)
-        return FALSE;
-
-    for(ptr = buf; *ptr; ptr++) {
+    strcpy(unescaped_path, path);
+    UrlUnescapeA(unescaped_path, NULL, NULL, URL_UNESCAPE_INPLACE);
+    for(ptr = unescaped_path; *ptr; ptr++) {
         if(*ptr == '\\')
             *ptr = '/';
     }
 
-    UrlUnescapeW(buf, NULL, NULL, URL_UNESCAPE_INPLACE);
+    buf = heap_strdupUtoW(unescaped_path);
+    if(!buf || lstrlenW(buf) < gecko_path_len)
+        return FALSE;
+
     buf[gecko_path_len] = 0;
 
     ret = !wcsicmp(buf, gecko_path);
-- 
GitLab

https://gitlab.winehq.org/wine/wine/-/merge_requests/585

[Jacek Caban (＠jacek)]

Jacek Caban (@jacek) commented about dlls/mshtml/nsembed.c:
> BOOL is_gecko_path(const char *path)
>  {
> -    WCHAR *buf, *ptr;
> +    char unescaped_path[INTERNET_MAX_URL_LENGTH], *ptr;
> +    WCHAR *buf;
>      BOOL ret;
>
> -    buf = heap_strdupUtoW(path);
> -    if(!buf || lstrlenW(buf) < gecko_path_len)
> -        return FALSE;
> -
> -    for(ptr = buf; *ptr; ptr++) {
> +    strcpy(unescaped_path, path);
> +    UrlUnescapeA(unescaped_path, NULL, NULL, URL_UNESCAPE_INPLACE);

I think that path is in UTF8 (unless it's specified otherwise by NewURI caller; we don't handle that yet, but I think Gecko doesn't use it another way anyway), while UrlUnescapeA will use CP_ACP.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/585#note_5684