urlmon: Implemented a function which removes dot segments from paths of hierarchical URIs.
This commit is contained in:
Thomas Mullaly
Alexandre Julliard
parent
300d0e1ad4
commit
5758b068ae
|
|
@ -2454,6 +2454,211 @@ static const uri_properties uri_tests[] = {
|
|||
{URL_SCHEME_HTTP,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
{ "http://gov.uk/test/../test2/././../test3/.././././", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_HOST|
|
||||
Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_PORT|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"http://gov.uk/test/../test2/././../test3/.././././",S_OK,FALSE},
|
||||
{"http",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{80,S_OK,FALSE},
|
||||
{URL_SCHEME_HTTP,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
{ "http://gov.uk/test/test2/../../..", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_HOST|
|
||||
Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_PORT|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"http://gov.uk/test/test2/../../..",S_OK,FALSE},
|
||||
{"http",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{80,S_OK,FALSE},
|
||||
{URL_SCHEME_HTTP,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
{ "http://gov.uk/test/test2/../../.", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_HOST|
|
||||
Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_PORT|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"http://gov.uk/",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"http://gov.uk/test/test2/../../.",S_OK,FALSE},
|
||||
{"http",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{80,S_OK,FALSE},
|
||||
{URL_SCHEME_HTTP,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
{ "file://c:\\tests\\../tests\\./.\\..\\foo%20bar.mp3", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_DISPLAY_URI|Uri_HAS_EXTENSION|Uri_HAS_PATH|
|
||||
Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|Uri_HAS_HOST_TYPE|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"file:///c:/foo%2520bar.mp3",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"file:///c:/foo%2520bar.mp3",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{".mp3",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/c:/foo%2520bar.mp3",S_OK,TRUE},
|
||||
{"/c:/foo%2520bar.mp3",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"file://c:\\tests\\../tests\\./.\\..\\foo%20bar.mp3",S_OK,FALSE},
|
||||
{"file",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_UNKNOWN,S_OK,FALSE},
|
||||
{0,S_FALSE,FALSE},
|
||||
{URL_SCHEME_FILE,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
/* Dot removal happens for unknown scheme types. */
|
||||
{ "zip://gov.uk/test/test2/../../.", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_HOST|
|
||||
Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"zip://gov.uk/",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"zip://gov.uk/",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"/",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"zip://gov.uk/test/test2/../../.",S_OK,FALSE},
|
||||
{"zip",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{0,S_FALSE,FALSE},
|
||||
{URL_SCHEME_UNKNOWN,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
/* Dot removal doesn't happen if NO_CANONICALIZE is set. */
|
||||
{ "http://gov.uk/test/test2/../../.", Uri_CREATE_NO_CANONICALIZE, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_EXTENSION|
|
||||
Uri_HAS_HOST|Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_PORT|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"http://gov.uk/test/test2/../../.",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"http://gov.uk/test/test2/../../.",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{".",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/test/test2/../../.",S_OK,TRUE},
|
||||
{"/test/test2/../../.",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"http://gov.uk/test/test2/../../.",S_OK,FALSE},
|
||||
{"http",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{80,S_OK,FALSE},
|
||||
{URL_SCHEME_HTTP,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
},
|
||||
/* Dot removal doesn't happen for wildcard scheme types. */
|
||||
{ "*:gov.uk/test/test2/../../.", 0, S_OK, FALSE,
|
||||
Uri_HAS_ABSOLUTE_URI|Uri_HAS_AUTHORITY|Uri_HAS_DISPLAY_URI|Uri_HAS_EXTENSION|
|
||||
Uri_HAS_HOST|Uri_HAS_PATH|Uri_HAS_PATH_AND_QUERY|Uri_HAS_RAW_URI|Uri_HAS_SCHEME_NAME|
|
||||
Uri_HAS_HOST_TYPE|Uri_HAS_SCHEME,
|
||||
TRUE,
|
||||
{
|
||||
{"*:gov.uk/test/test2/../../.",S_OK,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"*:gov.uk/test/test2/../../.",S_OK,TRUE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{".",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"gov.uk",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"/test/test2/../../.",S_OK,TRUE},
|
||||
{"/test/test2/../../.",S_OK,TRUE},
|
||||
{"",S_FALSE,TRUE},
|
||||
{"*:gov.uk/test/test2/../../.",S_OK,FALSE},
|
||||
{"*",S_OK,FALSE},
|
||||
{"",S_FALSE,FALSE},
|
||||
{"",S_FALSE,FALSE}
|
||||
},
|
||||
{
|
||||
{Uri_HOST_DNS,S_OK,FALSE},
|
||||
{0,S_FALSE,FALSE},
|
||||
{URL_SCHEME_WILDCARD,S_OK,FALSE},
|
||||
{URLZONE_INVALID,E_NOTIMPL,FALSE}
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
|
|
|
|||
|
|
@ -471,6 +471,70 @@ static void find_domain_name(const WCHAR *host, DWORD host_len,
|
|||
(host+host_len)-(host+*domain_start)));
|
||||
}
|
||||
|
||||
/* Removes the dot segments from a heirarchical URIs path component. This
|
||||
* function performs the removal in place.
|
||||
*
|
||||
* This is a modified version of Qt's QUrl function "removeDotsFromPath".
|
||||
*
|
||||
* This function returns the new length of the path string.
|
||||
*/
|
||||
static DWORD remove_dot_segments(WCHAR *path, DWORD path_len) {
|
||||
WCHAR *out = path;
|
||||
const WCHAR *in = out;
|
||||
const WCHAR *end = out + path_len;
|
||||
DWORD len;
|
||||
|
||||
while(in < end) {
|
||||
/* A. if the input buffer begins with a prefix of "/./" or "/.",
|
||||
* where "." is a complete path segment, then replace that
|
||||
* prefix with "/" in the input buffer; otherwise,
|
||||
*/
|
||||
if(in <= end - 3 && in[0] == '/' && in[1] == '.' && in[2] == '/') {
|
||||
in += 2;
|
||||
continue;
|
||||
} else if(in == end - 2 && in[0] == '/' && in[1] == '.') {
|
||||
*out++ = '/';
|
||||
in += 2;
|
||||
break;
|
||||
}
|
||||
|
||||
/* B. if the input buffer begins with a prefix of "/../" or "/..",
|
||||
* where ".." is a complete path segment, then replace that
|
||||
* prefix with "/" in the input buffer and remove the last
|
||||
* segment and its preceding "/" (if any) from the output
|
||||
* buffer; otherwise,
|
||||
*/
|
||||
if(in <= end - 4 && in[0] == '/' && in[1] == '.' && in[2] == '.' && in[3] == '/') {
|
||||
while(out > path && *(--out) != '/');
|
||||
|
||||
in += 3;
|
||||
continue;
|
||||
} else if(in == end - 3 && in[0] == '/' && in[1] == '.' && in[2] == '.') {
|
||||
while(out > path && *(--out) != '/');
|
||||
|
||||
if(*out == '/')
|
||||
++out;
|
||||
|
||||
in += 3;
|
||||
break;
|
||||
}
|
||||
|
||||
/* C. move the first path segment in the input buffer to the end of
|
||||
* the output buffer, including the initial "/" character (if
|
||||
* any) and any subsequent characters up to, but not including,
|
||||
* the next "/" character or the end of the input buffer.
|
||||
*/
|
||||
*out++ = *in++;
|
||||
while(in < end && *in != '/')
|
||||
*out++ = *in++;
|
||||
}
|
||||
|
||||
len = out - path;
|
||||
TRACE("(%p %d): Path after dot segments removed %s len=%d\n", path, path_len,
|
||||
debugstr_wn(path, len), len);
|
||||
return len;
|
||||
}
|
||||
|
||||
/* Computes the location where the elision should occur in the IPv6
|
||||
* address using the numerical values of each component stored in
|
||||
* 'values'. If the address shouldn't contain an elision then 'index'
|
||||
|
|
@ -2370,6 +2434,20 @@ static BOOL canonicalize_path_hierarchical(const parse_data *data, Uri *uri,
|
|||
|
||||
uri->path_len = uri->canon_len - uri->path_start;
|
||||
|
||||
/* Removing the dot segments only happens when it's not in
|
||||
* computeOnly mode and it's not a wildcard scheme.
|
||||
*/
|
||||
if(!computeOnly && data->scheme_type != URL_SCHEME_WILDCARD) {
|
||||
if(!(flags & Uri_CREATE_NO_CANONICALIZE)) {
|
||||
/* Remove the dot segments (if any) and reset everything to the new
|
||||
* correct length.
|
||||
*/
|
||||
DWORD new_len = remove_dot_segments(uri->canon_uri+uri->path_start, uri->path_len);
|
||||
uri->canon_len -= uri->path_len-new_len;
|
||||
uri->path_len = new_len;
|
||||
}
|
||||
}
|
||||
|
||||
if(!computeOnly)
|
||||
TRACE("Canonicalized path %s len=%d\n",
|
||||
debugstr_wn(uri->canon_uri+uri->path_start, uri->path_len),
|
||||
|
|
@ -2515,6 +2593,7 @@ static HRESULT canonicalize_uri(const parse_data *data, Uri *uri, DWORD flags) {
|
|||
if(!uri->canon_uri)
|
||||
return E_OUTOFMEMORY;
|
||||
|
||||
uri->canon_size = len;
|
||||
if(!canonicalize_scheme(data, uri, flags, FALSE)) {
|
||||
ERR("(%p %p %x): Unable to canonicalize the scheme of the URI.\n", data, uri, flags);
|
||||
heap_free(uri->canon_uri);
|
||||
|
|
@ -2528,6 +2607,21 @@ static HRESULT canonicalize_uri(const parse_data *data, Uri *uri, DWORD flags) {
|
|||
return E_INVALIDARG;
|
||||
}
|
||||
|
||||
/* There's a possibility we didn't use all the space we allocated
|
||||
* earlier.
|
||||
*/
|
||||
if(uri->canon_len < uri->canon_size) {
|
||||
/* This happens if the URI is hierarchical and dot
|
||||
* segments were removed from it's path.
|
||||
*/
|
||||
WCHAR *tmp = heap_realloc(uri->canon_uri, (uri->canon_len+1)*sizeof(WCHAR));
|
||||
if(!tmp)
|
||||
return E_OUTOFMEMORY;
|
||||
|
||||
uri->canon_uri = tmp;
|
||||
uri->canon_size = uri->canon_len;
|
||||
}
|
||||
|
||||
uri->canon_uri[uri->canon_len] = '\0';
|
||||
TRACE("(%p %p %x): finished canonicalizing the URI. uri=%s\n", data, uri, flags, debugstr_w(uri->canon_uri));
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue