Refactor Adblock to use simpler checks with optional pattern matching
The style is only added once. Match checks are simpler so they can be a lot faster. And slow pattern matching is optional, albeit it currently is hardcoded in the source.
This commit is contained in:
parent
7a2fad7578
commit
12ed2da768
1 changed files with 269 additions and 165 deletions
|
@ -22,18 +22,14 @@
|
||||||
|
|
||||||
#define HAVE_WEBKIT_RESOURCE_REQUEST WEBKIT_CHECK_VERSION (1, 1, 14)
|
#define HAVE_WEBKIT_RESOURCE_REQUEST WEBKIT_CHECK_VERSION (1, 1, 14)
|
||||||
|
|
||||||
|
#define SIGNATURE_SIZE 8
|
||||||
|
#define USE_PATTERN_MATCHING 1
|
||||||
#define ADBLOCK_FILTER_VALID(__filter) \
|
#define ADBLOCK_FILTER_VALID(__filter) \
|
||||||
(__filter && (g_str_has_prefix (__filter, "http") \
|
(__filter && (g_str_has_prefix (__filter, "http") \
|
||||||
|| g_str_has_prefix (__filter, "file")))
|
|| g_str_has_prefix (__filter, "file")))
|
||||||
|
|
||||||
typedef struct
|
|
||||||
{
|
|
||||||
const gchar* page_uri;
|
|
||||||
const gchar* uri;
|
|
||||||
const gchar* query;
|
|
||||||
} Matcher;
|
|
||||||
|
|
||||||
static GHashTable* pattern;
|
static GHashTable* pattern;
|
||||||
|
static GHashTable* keys;
|
||||||
static gchar* blockcss = NULL;
|
static gchar* blockcss = NULL;
|
||||||
static gchar* blockcssprivate = NULL;
|
static gchar* blockcssprivate = NULL;
|
||||||
static gchar* blockscript = NULL;
|
static gchar* blockscript = NULL;
|
||||||
|
@ -48,6 +44,11 @@ adblock_build_js (const gchar* style,
|
||||||
return g_strdup_printf (
|
return g_strdup_printf (
|
||||||
"window.addEventListener ('DOMContentLoaded',"
|
"window.addEventListener ('DOMContentLoaded',"
|
||||||
"function () {"
|
"function () {"
|
||||||
|
" var styles = document.getElementsByTagName('style');"
|
||||||
|
" for (i=0; i<styles.length; i++) {"
|
||||||
|
" if (styles[i].getAttribute('title') == 'madblock')"
|
||||||
|
" return;"
|
||||||
|
" }"
|
||||||
" var URL = location.href;"
|
" var URL = location.href;"
|
||||||
" var sites = new Array(); %s;"
|
" var sites = new Array(); %s;"
|
||||||
" var public = '%s';"
|
" var public = '%s';"
|
||||||
|
@ -59,6 +60,7 @@ adblock_build_js (const gchar* style,
|
||||||
" public += ' {display: none !important;}';"
|
" public += ' {display: none !important;}';"
|
||||||
" var mystyle = document.createElement('style');"
|
" var mystyle = document.createElement('style');"
|
||||||
" mystyle.setAttribute('type', 'text/css');"
|
" mystyle.setAttribute('type', 'text/css');"
|
||||||
|
" mystyle.setAttribute('title', 'madblock');"
|
||||||
" mystyle.appendChild(document.createTextNode(public));"
|
" mystyle.appendChild(document.createTextNode(public));"
|
||||||
" var head = document.getElementsByTagName('head')[0];"
|
" var head = document.getElementsByTagName('head')[0];"
|
||||||
" if (head) head.appendChild(mystyle);"
|
" if (head) head.appendChild(mystyle);"
|
||||||
|
@ -68,45 +70,19 @@ adblock_build_js (const gchar* style,
|
||||||
}
|
}
|
||||||
|
|
||||||
static gchar *
|
static gchar *
|
||||||
adblock_fixup_regexp (gchar* src)
|
adblock_fixup_regexp (gchar* src);
|
||||||
{
|
|
||||||
gchar* dst;
|
|
||||||
gchar* s;
|
|
||||||
|
|
||||||
if (!(src && *src))
|
static void
|
||||||
return g_strdup ("");
|
adblock_init_db ()
|
||||||
|
|
||||||
/* FIXME: Avoid always allocating twice the string */
|
|
||||||
s = dst = g_malloc (strlen (src) * 2);
|
|
||||||
|
|
||||||
while (*src)
|
|
||||||
{
|
{
|
||||||
switch (*src)
|
pattern = g_hash_table_new_full (g_str_hash, g_str_equal,
|
||||||
{
|
(GDestroyNotify)g_free,
|
||||||
case '*':
|
(GDestroyNotify)g_regex_unref);
|
||||||
*s++ = '.';
|
keys = g_hash_table_new_full (g_str_hash, g_str_equal,
|
||||||
break;
|
(GDestroyNotify)g_free,
|
||||||
case '.':
|
(GDestroyNotify)g_regex_unref);
|
||||||
*s++ = '\\';
|
katze_assign (blockcss, g_strdup ("z-non-exist"));
|
||||||
break;
|
katze_assign (blockcssprivate, g_strdup (""));
|
||||||
case '?':
|
|
||||||
*s++ = '\\';
|
|
||||||
break;
|
|
||||||
case '|':
|
|
||||||
*s++ = '\\';
|
|
||||||
break;
|
|
||||||
/* FIXME: We actually need to match :[0-9]+ or '/'. Sign means
|
|
||||||
"here could be port number or nothing". So bla.com^ will match
|
|
||||||
bla.com/ or bla.com:8080/ but not bla.com.au/ */
|
|
||||||
case '^':
|
|
||||||
*src = '?';
|
|
||||||
break;
|
|
||||||
}
|
|
||||||
*s++ = *src;
|
|
||||||
src++;
|
|
||||||
}
|
|
||||||
*s = 0;
|
|
||||||
return dst;
|
|
||||||
}
|
}
|
||||||
|
|
||||||
static void
|
static void
|
||||||
|
@ -135,11 +111,7 @@ adblock_reload_rules (MidoriExtension* extension)
|
||||||
if (!filters)
|
if (!filters)
|
||||||
return;
|
return;
|
||||||
|
|
||||||
pattern = g_hash_table_new_full (g_str_hash, g_str_equal,
|
adblock_init_db ();
|
||||||
(GDestroyNotify)g_free,
|
|
||||||
(GDestroyNotify)g_regex_unref);
|
|
||||||
katze_assign (blockcss, g_strdup ("z-non-exist"));
|
|
||||||
katze_assign (blockcssprivate, g_strdup (""));
|
|
||||||
|
|
||||||
while (filters[i++] != NULL)
|
while (filters[i++] != NULL)
|
||||||
{
|
{
|
||||||
|
@ -554,40 +526,97 @@ adblock_browser_populate_tool_menu_cb (MidoriBrowser* browser,
|
||||||
}
|
}
|
||||||
|
|
||||||
static gboolean
|
static gboolean
|
||||||
adblock_is_matched (const gchar* opts,
|
adblock_check_filter_options (GRegex* regex,
|
||||||
const GRegex* regex,
|
const gchar* opts,
|
||||||
Matcher* data)
|
const gchar* req_uri,
|
||||||
{
|
const gchar* page_uri)
|
||||||
gchar* patt;
|
|
||||||
|
|
||||||
patt = g_strdup (data->uri);
|
|
||||||
/* TODO: To figure out
|
|
||||||
if (g_regex_match_simple ("type=fulluri,", opts, G_REGEX_UNGREEDY, G_REGEX_MATCH_NOTEMPTY))
|
|
||||||
patt = g_strdup (data->uri);
|
|
||||||
else
|
|
||||||
patt = g_strdup (data->query);
|
|
||||||
*/
|
|
||||||
if (g_regex_match_full (regex, patt, -1, 0, 0, NULL, NULL))
|
|
||||||
{
|
{
|
||||||
if (g_regex_match_simple (",third-party", opts,
|
if (g_regex_match_simple (",third-party", opts,
|
||||||
G_REGEX_CASELESS, G_REGEX_MATCH_NOTEMPTY))
|
G_REGEX_CASELESS, G_REGEX_MATCH_NOTEMPTY))
|
||||||
{
|
{
|
||||||
if (data->page_uri && g_regex_match_full (regex, data->page_uri, -1, 0, 0, NULL, NULL))
|
if (page_uri && g_regex_match_full (regex, page_uri, -1, 0, 0, NULL, NULL))
|
||||||
{
|
|
||||||
g_free (patt);
|
|
||||||
return FALSE;
|
|
||||||
}
|
|
||||||
}
|
|
||||||
/* TODO: Domain opt check */
|
|
||||||
g_free (patt);
|
|
||||||
return TRUE;
|
return TRUE;
|
||||||
}
|
}
|
||||||
else
|
/* TODO: Domain opt check */
|
||||||
{
|
|
||||||
g_free (patt);
|
|
||||||
return FALSE;
|
return FALSE;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
static gboolean
|
||||||
|
adblock_is_matched_by_pattern (const gchar* req_uri,
|
||||||
|
const gchar* page_uri)
|
||||||
|
{
|
||||||
|
GHashTableIter iter;
|
||||||
|
gpointer opts, regex;
|
||||||
|
|
||||||
|
if (USE_PATTERN_MATCHING == 0)
|
||||||
|
return FALSE;
|
||||||
|
|
||||||
|
g_hash_table_iter_init (&iter, pattern);
|
||||||
|
while (g_hash_table_iter_next (&iter, &opts, ®ex))
|
||||||
|
{
|
||||||
|
if (g_regex_match_full (regex, req_uri, -1, 0, 0, NULL, NULL))
|
||||||
|
{
|
||||||
|
if (opts && adblock_check_filter_options (regex, opts, req_uri, page_uri) == TRUE)
|
||||||
|
return FALSE;
|
||||||
|
else
|
||||||
|
{
|
||||||
|
/* g_debug("blocked by pattern regexp=%s -- %s", g_regex_get_pattern (regex), req_uri); */
|
||||||
|
return TRUE;
|
||||||
}
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return FALSE;
|
||||||
|
}
|
||||||
|
|
||||||
|
static gboolean
|
||||||
|
adblock_is_matched_by_key (const gchar* opts,
|
||||||
|
const gchar* req_uri,
|
||||||
|
const gchar* page_uri)
|
||||||
|
{
|
||||||
|
int pos = 0;
|
||||||
|
gchar *sig = NULL;
|
||||||
|
GRegex* regex;
|
||||||
|
gchar* uri;
|
||||||
|
|
||||||
|
uri = adblock_fixup_regexp ((gchar*)req_uri);
|
||||||
|
|
||||||
|
int len = strlen (uri);
|
||||||
|
for (pos = len - SIGNATURE_SIZE; pos >= 0; pos--) {
|
||||||
|
sig = strndup (uri+pos, SIGNATURE_SIZE);
|
||||||
|
regex = g_hash_table_lookup (keys, sig);
|
||||||
|
if (regex)
|
||||||
|
{
|
||||||
|
if (g_regex_match_full (regex, req_uri, -1, 0, 0, NULL, NULL))
|
||||||
|
{
|
||||||
|
g_free (uri);
|
||||||
|
g_free (sig);
|
||||||
|
if (opts && adblock_check_filter_options (regex, opts, req_uri, page_uri) == TRUE)
|
||||||
|
return FALSE;
|
||||||
|
else
|
||||||
|
{
|
||||||
|
/* g_debug("blocked by key sig=%s regexp=%s -- %s", sig, g_regex_get_pattern (regex), req_uri); */
|
||||||
|
return TRUE;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
g_free (sig);
|
||||||
|
}
|
||||||
|
g_free (uri);
|
||||||
|
return FALSE;
|
||||||
|
}
|
||||||
|
|
||||||
|
static gboolean
|
||||||
|
adblock_is_matched (const gchar* opts,
|
||||||
|
const gchar* req_uri,
|
||||||
|
const gchar* page_uri)
|
||||||
|
{
|
||||||
|
|
||||||
|
if (adblock_is_matched_by_key (opts, req_uri, page_uri) == TRUE)
|
||||||
|
return TRUE;
|
||||||
|
else
|
||||||
|
return adblock_is_matched_by_pattern (req_uri, page_uri);
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
#if HAVE_WEBKIT_RESOURCE_REQUEST
|
#if HAVE_WEBKIT_RESOURCE_REQUEST
|
||||||
static void
|
static void
|
||||||
|
@ -598,91 +627,66 @@ adblock_resource_request_starting_cb (WebKitWebView* web_view,
|
||||||
WebKitNetworkResponse* response,
|
WebKitNetworkResponse* response,
|
||||||
GtkWidget* image)
|
GtkWidget* image)
|
||||||
{
|
{
|
||||||
Matcher data;
|
|
||||||
const char *page_uri;
|
|
||||||
const gchar* uri;
|
|
||||||
SoupMessage* msg;
|
SoupMessage* msg;
|
||||||
SoupURI* soup_uri;
|
const gchar* req_uri;
|
||||||
|
const char *page_uri;
|
||||||
|
|
||||||
uri = webkit_network_request_get_uri (request);
|
req_uri = webkit_network_request_get_uri (request);
|
||||||
if (!strncmp (uri, "data", 4) || !strncmp (uri, "file", 4))
|
|
||||||
|
if (!req_uri)
|
||||||
|
return;
|
||||||
|
if (!strncmp (req_uri, "data", 4) || !strncmp (req_uri, "file", 4))
|
||||||
return;
|
return;
|
||||||
|
|
||||||
msg = webkit_network_request_get_message (request);
|
msg = webkit_network_request_get_message (request);
|
||||||
if (!msg)
|
if (!msg)
|
||||||
return;
|
return;
|
||||||
|
|
||||||
if (msg->method && !strncmp (msg->method, "POST", 4))
|
if (msg->method && !strncmp (msg->method, "POST", 4))
|
||||||
return;
|
return;
|
||||||
|
|
||||||
soup_uri = soup_uri_new (uri);
|
|
||||||
if (soup_uri->query)
|
|
||||||
data.query = g_strdup_printf ("%s?%s", soup_uri->path, soup_uri->query);
|
|
||||||
else
|
|
||||||
data.query = g_strdup (soup_uri->path);
|
|
||||||
soup_uri_free (soup_uri);
|
|
||||||
|
|
||||||
data.uri = uri;
|
|
||||||
page_uri = webkit_web_view_get_uri (web_view);
|
page_uri = webkit_web_view_get_uri (web_view);
|
||||||
|
|
||||||
if (!page_uri || !strcmp (page_uri, "about:blank"))
|
if (!page_uri || !strcmp (page_uri, "about:blank"))
|
||||||
page_uri = uri;
|
page_uri = req_uri;
|
||||||
data.page_uri = page_uri;
|
|
||||||
|
|
||||||
if (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched, &data))
|
/* gdouble elapsed = 0.0;
|
||||||
{
|
g_test_timer_start (); */
|
||||||
#if 0
|
|
||||||
gchar* text;
|
|
||||||
|
|
||||||
if (gtk_widget_get_has_tooltip (image))
|
|
||||||
text = g_strdup_printf ("%s\n%s", gtk_widget_get_tooltip_text (image), uri);
|
|
||||||
else
|
|
||||||
text = g_strdup_printf ("Blocked content: \n%s", uri);
|
|
||||||
gtk_widget_set_tooltip_text (image, text);
|
|
||||||
g_free (text);
|
|
||||||
#endif
|
|
||||||
|
|
||||||
|
/* TODO: opts should be defined */
|
||||||
|
if (adblock_is_matched (NULL, req_uri, page_uri))
|
||||||
webkit_network_request_set_uri (request, "about:blank");
|
webkit_network_request_set_uri (request, "about:blank");
|
||||||
}
|
/* elapsed += g_test_timer_elapsed ();
|
||||||
|
g_debug ("%f", elapsed); */
|
||||||
|
|
||||||
}
|
}
|
||||||
#else
|
#else
|
||||||
static void
|
static void
|
||||||
adblock_session_request_queued_cb (SoupSession* session,
|
adblock_session_request_queued_cb (SoupSession* session,
|
||||||
SoupMessage* msg)
|
SoupMessage* msg)
|
||||||
{
|
{
|
||||||
Matcher data;
|
|
||||||
SoupURI* soup_uri;
|
SoupURI* soup_uri;
|
||||||
gchar* uri;
|
gchar* req_uri;
|
||||||
gchar* page_uri;
|
gchar* page_uri;
|
||||||
|
|
||||||
if (msg->method && !strncmp (msg->method, "POST", 4))
|
if (msg->method && !strncmp (msg->method, "POST", 4))
|
||||||
return;
|
return;
|
||||||
|
|
||||||
/* FIXME: There is a crasher somewhere introduced with the refactoring */
|
|
||||||
|
|
||||||
soup_uri = soup_message_get_uri (msg);
|
soup_uri = soup_message_get_uri (msg);
|
||||||
uri = soup_uri_to_string (soup_uri, FALSE);
|
req_uri = soup_uri_to_string (soup_uri, FALSE);
|
||||||
if (soup_uri->query)
|
|
||||||
data.query = g_strdup_printf ("%s?%s", soup_uri->path, soup_uri->query);
|
|
||||||
else
|
|
||||||
data.query = g_strdup (soup_uri->path);
|
|
||||||
|
|
||||||
data.uri = uri;
|
|
||||||
page_uri = NULL; /* FIXME */
|
page_uri = NULL; /* FIXME */
|
||||||
|
|
||||||
if (!page_uri || !strcmp (page_uri, "about:blank"))
|
if (!page_uri || !strcmp (page_uri, "about:blank"))
|
||||||
page_uri = uri;
|
page_uri = req_uri;
|
||||||
data.page_uri = page_uri;
|
|
||||||
|
|
||||||
if (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched, &data))
|
if (adblock_is_matched (NULL, req_uri, page_uri))
|
||||||
{
|
{
|
||||||
/* FIXME: Update image tooltip */
|
|
||||||
|
|
||||||
soup_uri = soup_uri_new ("http://.invalid");
|
soup_uri = soup_uri_new ("http://.invalid");
|
||||||
soup_message_set_uri (msg, soup_uri);
|
soup_message_set_uri (msg, soup_uri);
|
||||||
soup_uri_free (soup_uri);
|
soup_uri_free (soup_uri);
|
||||||
|
g_free (req_uri);
|
||||||
|
return;
|
||||||
}
|
}
|
||||||
g_free (uri);
|
g_free (req_uri);
|
||||||
}
|
}
|
||||||
#endif
|
#endif
|
||||||
|
|
||||||
|
@ -700,6 +704,7 @@ adblock_add_tab_cb (MidoriBrowser* browser,
|
||||||
MidoriView* view,
|
MidoriView* view,
|
||||||
GtkWidget* image)
|
GtkWidget* image)
|
||||||
{
|
{
|
||||||
|
|
||||||
GtkWidget* web_view = gtk_bin_get_child (GTK_BIN (view));
|
GtkWidget* web_view = gtk_bin_get_child (GTK_BIN (view));
|
||||||
g_signal_connect (web_view, "window-object-cleared",
|
g_signal_connect (web_view, "window-object-cleared",
|
||||||
G_CALLBACK (adblock_window_object_cleared_cb), 0);
|
G_CALLBACK (adblock_window_object_cleared_cb), 0);
|
||||||
|
@ -749,28 +754,118 @@ adblock_app_add_browser_cb (MidoriApp* app,
|
||||||
g_object_unref (statusbar);
|
g_object_unref (statusbar);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
static gchar *
|
||||||
|
adblock_fixup_regexp (gchar* src)
|
||||||
|
{
|
||||||
|
gchar* dst;
|
||||||
|
GString* str;
|
||||||
|
int len;
|
||||||
|
|
||||||
|
if (!src)
|
||||||
|
return NULL;
|
||||||
|
|
||||||
|
str = g_string_new ("");
|
||||||
|
|
||||||
|
/* lets strip first .* */
|
||||||
|
if (src[0] == '*')
|
||||||
|
{
|
||||||
|
(void)*src++;
|
||||||
|
}
|
||||||
|
|
||||||
|
do
|
||||||
|
{
|
||||||
|
switch (*src)
|
||||||
|
{
|
||||||
|
case '*':
|
||||||
|
g_string_append (str, ".*");
|
||||||
|
break;
|
||||||
|
case '.':
|
||||||
|
g_string_append (str, "\\.");
|
||||||
|
break;
|
||||||
|
case '?':
|
||||||
|
g_string_append (str, "\\?");
|
||||||
|
break;
|
||||||
|
case '|':
|
||||||
|
g_string_append (str, "");
|
||||||
|
break;
|
||||||
|
/* FIXME: We actually need to match :[0-9]+ or '/'. Sign means
|
||||||
|
"here could be port number or nothing". So bla.com^ will match
|
||||||
|
bla.com/ or bla.com:8080/ but not bla.com.au/ */
|
||||||
|
case '^':
|
||||||
|
g_string_append (str, "");
|
||||||
|
break;
|
||||||
|
default:
|
||||||
|
g_string_append_printf (str,"%c", *src);
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
src++;
|
||||||
|
}
|
||||||
|
while (*src);
|
||||||
|
|
||||||
|
dst = g_strdup (str->str);
|
||||||
|
g_string_free (str, TRUE);
|
||||||
|
/* We dont need .* in the end of url. Thats stupid */
|
||||||
|
len = strlen (dst);
|
||||||
|
if (dst && dst[len-1] == '*' && dst[len-2] == '.')
|
||||||
|
{
|
||||||
|
dst[len-2] = '\0';
|
||||||
|
}
|
||||||
|
return dst;
|
||||||
|
}
|
||||||
|
|
||||||
static void
|
static void
|
||||||
adblock_compile_regexp (GHashTable* tbl,
|
adblock_compile_regexp (GHashTable* tbl,
|
||||||
|
GHashTable* keystbl,
|
||||||
gchar* patt,
|
gchar* patt,
|
||||||
gchar* opts)
|
gchar* opts)
|
||||||
{
|
{
|
||||||
GRegex* regex;
|
GRegex* regex;
|
||||||
GError* error = NULL;
|
GError* error = NULL;
|
||||||
|
int pos = 0;
|
||||||
|
gchar *sig;
|
||||||
|
|
||||||
/* TODO: Play with optimization flags */
|
/* TODO: Play with optimization flags */
|
||||||
regex = g_regex_new (patt, G_REGEX_OPTIMIZE,
|
regex = g_regex_new (patt, G_REGEX_OPTIMIZE,
|
||||||
G_REGEX_MATCH_NOTEMPTY, &error);
|
G_REGEX_MATCH_NOTEMPTY, &error);
|
||||||
|
if (error)
|
||||||
if (!error)
|
|
||||||
g_hash_table_insert (tbl, opts, regex);
|
|
||||||
else
|
|
||||||
{
|
{
|
||||||
g_warning ("%s: %s", G_STRFUNC, error->message);
|
g_warning ("%s: %s", G_STRFUNC, error->message);
|
||||||
g_error_free (error);
|
g_error_free (error);
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
if (!g_regex_match_simple ("^/.*[\\^\\$\\*].*/$", patt, G_REGEX_UNGREEDY, G_REGEX_MATCH_NOTEMPTY))
|
||||||
|
{
|
||||||
|
int len = strlen (patt);
|
||||||
|
for (pos = len - SIGNATURE_SIZE; pos >= 0; pos--) {
|
||||||
|
sig = strndup(patt+pos, SIGNATURE_SIZE);
|
||||||
|
if (!g_regex_match_simple ("[\\*]", sig, G_REGEX_UNGREEDY, G_REGEX_MATCH_NOTEMPTY) &&
|
||||||
|
!g_hash_table_lookup (keystbl, sig))
|
||||||
|
{
|
||||||
|
/* g_debug ("sig: %s %s", sig, patt); */
|
||||||
|
g_hash_table_insert (keystbl, sig, regex);
|
||||||
|
}
|
||||||
|
else
|
||||||
|
{
|
||||||
|
if (g_regex_match_simple ("^\\*", sig, G_REGEX_UNGREEDY, G_REGEX_MATCH_NOTEMPTY) &&
|
||||||
|
!g_hash_table_lookup (tbl, opts))
|
||||||
|
{
|
||||||
|
/* g_debug ("patt2: %s %s", sig, patt); */
|
||||||
|
g_hash_table_insert (tbl, opts, regex);
|
||||||
|
}
|
||||||
|
g_free (sig);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
else
|
||||||
|
{
|
||||||
|
/*g_debug ("patt: %s", patt); */
|
||||||
|
/* Pattern is a regexp chars */
|
||||||
|
g_hash_table_insert (tbl, opts, regex);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
static void
|
static gchar*
|
||||||
adblock_add_url_pattern (gchar* format,
|
adblock_add_url_pattern (gchar* format,
|
||||||
gchar* type,
|
gchar* type,
|
||||||
gchar* line)
|
gchar* line)
|
||||||
|
@ -802,12 +897,12 @@ adblock_add_url_pattern (gchar* format,
|
||||||
format_patt = g_strdup_printf (format, fixed_patt);
|
format_patt = g_strdup_printf (format, fixed_patt);
|
||||||
|
|
||||||
/* g_debug ("got: %s opts %s", format_patt, opts); */
|
/* g_debug ("got: %s opts %s", format_patt, opts); */
|
||||||
adblock_compile_regexp (pattern, format_patt, opts);
|
adblock_compile_regexp (pattern, keys, format_patt, opts);
|
||||||
|
|
||||||
g_strfreev (data);
|
g_strfreev (data);
|
||||||
g_free (patt);
|
g_free (patt);
|
||||||
g_free (fixed_patt);
|
g_free (fixed_patt);
|
||||||
g_free (format_patt);
|
return format_patt;
|
||||||
}
|
}
|
||||||
|
|
||||||
static void
|
static void
|
||||||
|
@ -896,17 +991,14 @@ adblock_parse_line (gchar* line)
|
||||||
{
|
{
|
||||||
(void)*line++;
|
(void)*line++;
|
||||||
(void)*line++;
|
(void)*line++;
|
||||||
adblock_add_url_pattern ("^https?://([a-z0-9\\.]+)?%s", "fulluri", line);
|
return adblock_add_url_pattern ("%s", "fulluri", line);
|
||||||
return NULL;
|
|
||||||
}
|
}
|
||||||
if (line[0] == '|')
|
if (line[0] == '|')
|
||||||
{
|
{
|
||||||
(void)*line++;
|
(void)*line++;
|
||||||
adblock_add_url_pattern ("^%s", "fulluri", line);
|
return adblock_add_url_pattern ("^%s", "fulluri", line);
|
||||||
return NULL;
|
|
||||||
}
|
}
|
||||||
adblock_add_url_pattern ("%s", "uri", line);
|
return adblock_add_url_pattern ("%s", "uri", line);
|
||||||
return line;
|
|
||||||
}
|
}
|
||||||
|
|
||||||
static void
|
static void
|
||||||
|
@ -914,11 +1006,16 @@ adblock_parse_file (gchar* path)
|
||||||
{
|
{
|
||||||
FILE* file;
|
FILE* file;
|
||||||
gchar line[500];
|
gchar line[500];
|
||||||
|
gchar *out;
|
||||||
|
|
||||||
if ((file = g_fopen (path, "r")))
|
if ((file = g_fopen (path, "r")))
|
||||||
{
|
{
|
||||||
while (fgets (line, 500, file))
|
while (fgets (line, 500, file))
|
||||||
adblock_parse_line (line);
|
{
|
||||||
|
out = adblock_parse_line (line);
|
||||||
|
if (out)
|
||||||
|
g_free (out);
|
||||||
|
}
|
||||||
fclose (file);
|
fclose (file);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
@ -999,19 +1096,20 @@ adblock_activate_cb (MidoriExtension* extension,
|
||||||
static void
|
static void
|
||||||
test_adblock_parse (void)
|
test_adblock_parse (void)
|
||||||
{
|
{
|
||||||
|
adblock_init_db ();
|
||||||
g_assert (!adblock_parse_line (NULL));
|
g_assert (!adblock_parse_line (NULL));
|
||||||
g_assert (!adblock_parse_line ("!"));
|
g_assert (!adblock_parse_line ("!"));
|
||||||
g_assert (!adblock_parse_line ("@@"));
|
g_assert (!adblock_parse_line ("@@"));
|
||||||
g_assert (!adblock_parse_line ("##"));
|
g_assert (!adblock_parse_line ("##"));
|
||||||
g_assert (!adblock_parse_line ("["));
|
g_assert (!adblock_parse_line ("["));
|
||||||
|
|
||||||
g_assert_cmpstr (adblock_parse_line ("*foo"), ==, ".*foo");
|
g_assert_cmpstr (adblock_parse_line ("*foo"), ==, "foo");
|
||||||
|
g_assert_cmpstr (adblock_parse_line ("f*oo"), ==, "f.*oo");
|
||||||
g_assert_cmpstr (adblock_parse_line ("?foo"), ==, "\\?foo");
|
g_assert_cmpstr (adblock_parse_line ("?foo"), ==, "\\?foo");
|
||||||
g_assert_cmpstr (adblock_parse_line ("foo*"), ==, "foo.*");
|
|
||||||
g_assert_cmpstr (adblock_parse_line ("foo?"), ==, "foo\\?");
|
g_assert_cmpstr (adblock_parse_line ("foo?"), ==, "foo\\?");
|
||||||
|
|
||||||
g_assert_cmpstr (adblock_parse_line (".*foo/bar"), ==, "\\..*foo/bar");
|
g_assert_cmpstr (adblock_parse_line (".*foo/bar"), ==, "\\..*foo/bar");
|
||||||
g_assert_cmpstr (adblock_parse_line ("http://bla.blub/*"), ==, "http://bla\\.blub/.*");
|
g_assert_cmpstr (adblock_parse_line ("http://bla.blub/*"), ==, "http://bla\\.blub/");
|
||||||
}
|
}
|
||||||
|
|
||||||
static void
|
static void
|
||||||
|
@ -1020,33 +1118,42 @@ test_adblock_pattern (void)
|
||||||
gint temp;
|
gint temp;
|
||||||
gchar* filename;
|
gchar* filename;
|
||||||
|
|
||||||
pattern = g_hash_table_new_full (g_str_hash, g_str_equal,
|
|
||||||
(GDestroyNotify)g_free,
|
|
||||||
(GDestroyNotify)g_regex_unref);
|
|
||||||
|
|
||||||
temp = g_file_open_tmp ("midori_adblock_match_test_XXXXXX", &filename, NULL);
|
temp = g_file_open_tmp ("midori_adblock_match_test_XXXXXX", &filename, NULL);
|
||||||
|
|
||||||
/* TODO: Update some tests and add new ones. */
|
/* TODO: Update some tests and add new ones. */
|
||||||
g_file_set_contents (filename,
|
g_file_set_contents (filename,
|
||||||
"*ads.foo.bar*\n"
|
"*ads.foo.bar*\n"
|
||||||
"*ads.bogus.name*\n"
|
"*ads.bogus.name*\n"
|
||||||
"http://ads.bla.blub/*\n"
|
"||^http://ads.bla.blub/*\n"
|
||||||
"http://ads.blub.boing/*",
|
"|http://ads.blub.boing/*$domain=xxx.com\n"
|
||||||
|
"engine.adct.ru/*?\n"
|
||||||
|
"/addyn|*|adtech;\n"
|
||||||
|
"doubleclick.net/pfadx/*.mtvi\n"
|
||||||
|
"objects.tremormedia.com/embed/xml/*.xml?r=\n"
|
||||||
|
"videostrip.com^*/admatcherclient.\n"
|
||||||
|
"test.dom/test?var\n"
|
||||||
|
"br.gcl.ru/cgi-bin/br/",
|
||||||
-1, NULL);
|
-1, NULL);
|
||||||
|
|
||||||
adblock_parse_file (filename);
|
adblock_parse_file (filename);
|
||||||
|
|
||||||
g_assert (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
g_assert (adblock_is_matched (NULL, "http://test.dom/test?var=1", ""));
|
||||||
"http://ads.foo.bar/teddy"));
|
g_assert (adblock_is_matched (NULL, "http://ads.foo.bar/teddy", ""));
|
||||||
g_assert (!g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
g_assert (!adblock_is_matched (NULL, "http://ads.fuu.bar/teddy", ""));
|
||||||
"http://ads.fuu.bar/teddy"));
|
g_assert (adblock_is_matched (NULL, "https://ads.bogus.name/blub", ""));
|
||||||
g_assert (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
g_assert (adblock_is_matched (NULL, "http://ads.bla.blub/kitty", ""));
|
||||||
"https://ads.bogus.name/blub"));
|
g_assert (adblock_is_matched (NULL, "http://ads.blub.boing/soda", ""));
|
||||||
g_assert (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
g_assert (!adblock_is_matched (NULL, "http://ads.foo.boing/beer", ""));
|
||||||
"http://ads.bla.blub/kitty"));
|
g_assert (adblock_is_matched (NULL, "https://testsub.engine.adct.ru/test?id=1", ""));
|
||||||
g_assert (g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
if (USE_PATTERN_MATCHING)
|
||||||
"http://ads.blub.boing/soda"));
|
g_assert (adblock_is_matched (NULL, "http://test.ltd/addyn/test/test?var=adtech;&var2=1", ""));
|
||||||
g_assert (!g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,
|
g_assert (adblock_is_matched (NULL, "http://add.doubleclick.net/pfadx/aaaa.mtvi", ""));
|
||||||
"http://ads.foo.boing/beer"));
|
g_assert (!adblock_is_matched (NULL, "http://add.doubleclick.net/pfadx/aaaa.mtv", ""));
|
||||||
|
g_assert (adblock_is_matched (NULL, "http://objects.tremormedia.com/embed/xml/list.xml?r=", ""));
|
||||||
|
g_assert (!adblock_is_matched (NULL, "http://qq.videostrip.c/sub/admatcherclient.php", ""));
|
||||||
|
g_assert (adblock_is_matched (NULL, "http://qq.videostrip.com/sub/admatcherclient.php", ""));
|
||||||
|
g_assert (adblock_is_matched (NULL, "http://qq.videostrip.com/sub/admatcherclient.php", ""));
|
||||||
|
g_assert (adblock_is_matched (NULL, "http://br.gcl.ru/cgi-bin/br/test", ""));
|
||||||
|
|
||||||
close (temp);
|
close (temp);
|
||||||
g_unlink (filename);
|
g_unlink (filename);
|
||||||
|
@ -1057,10 +1164,7 @@ test_adblock_pattern (void)
|
||||||
static void
|
static void
|
||||||
test_adblock_count (void)
|
test_adblock_count (void)
|
||||||
{
|
{
|
||||||
pattern = g_hash_table_new_full (g_str_hash, g_str_equal,
|
adblock_init_db ();
|
||||||
(GDestroyNotify)g_free,
|
|
||||||
(GDestroyNotify)g_regex_unref);
|
|
||||||
|
|
||||||
gchar* urls[6] = {
|
gchar* urls[6] = {
|
||||||
"https://bugs.webkit.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&keywords_type=allwords&keywords=&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailassigned_to1=1&emailtype1=substring&email1=&emailassigned_to2=1&emailreporter2=1&emailcc2=1&emailtype2=substring&email2=&bugidtype=include&bug_id=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&query_based_on=gtkport&field0-0-0=keywords&type0-0-0=anywordssubstr&value0-0-0=Gtk%20Cairo%20soup&field0-0-1=short_desc&type0-0-1=anywordssubstr&value0-0-1=Gtk%20Cairo%20soup%20autoconf%20automake%20autotool&field0-0-2=component&type0-0-2=equals&value0-0-2=WebKit%20Gtk",
|
"https://bugs.webkit.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&keywords_type=allwords&keywords=&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailassigned_to1=1&emailtype1=substring&email1=&emailassigned_to2=1&emailreporter2=1&emailcc2=1&emailtype2=substring&email2=&bugidtype=include&bug_id=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&query_based_on=gtkport&field0-0-0=keywords&type0-0-0=anywordssubstr&value0-0-0=Gtk%20Cairo%20soup&field0-0-1=short_desc&type0-0-1=anywordssubstr&value0-0-1=Gtk%20Cairo%20soup%20autoconf%20automake%20autotool&field0-0-2=component&type0-0-2=equals&value0-0-2=WebKit%20Gtk",
|
||||||
"http://www.engadget.com/2009/09/24/google-hits-android-rom-modder-with-a-cease-and-desist-letter/",
|
"http://www.engadget.com/2009/09/24/google-hits-android-rom-modder-with-a-cease-and-desist-letter/",
|
||||||
|
@ -1079,7 +1183,7 @@ test_adblock_count (void)
|
||||||
{
|
{
|
||||||
str = urls[i];
|
str = urls[i];
|
||||||
g_test_timer_start ();
|
g_test_timer_start ();
|
||||||
g_hash_table_find (pattern, (GHRFunc) adblock_is_matched,str);
|
adblock_is_matched (NULL, str, "");
|
||||||
elapsed += g_test_timer_elapsed ();
|
elapsed += g_test_timer_elapsed ();
|
||||||
}
|
}
|
||||||
g_print ("Search took %f seconds\n", elapsed);
|
g_print ("Search took %f seconds\n", elapsed);
|
||||||
|
|
Loading…
Reference in a new issue