diff --git a/src/extract-uri.cpp b/src/extract-uri.cpp
index dd6549c8f8b9fd660e1fe93d401bd44b4d90a220..76778bacb93748ff655a9fb0840fdaed1b9a8904 100644 (file)
--- a/src/extract-uri.cpp
+++ b/src/extract-uri.cpp
#include <glib.h>
// FIXME: kill this ugliness when we have a proper CSS parser
-gchar *extract_uri(gchar const *s)
+
+// Functions as per 4.3.4 of CSS 2.1
+// http://www.w3.org/TR/CSS21/syndata.html#uri
+gchar *extract_uri( gchar const *s, gchar const** endptr )
{
+ if (!s)
+ return NULL;
+
+ gchar* result = NULL;
gchar const *sb = s;
- g_assert( strncmp(sb, "url", 3) == 0 );
+ if ( strlen(sb) < 4 || strncmp(sb, "url", 3) != 0 ) {
+ return NULL;
+ }
+
sb += 3;
+ if ( endptr ) {
+ *endptr = 0;
+ }
+
+ // This first whitespace technically is not allowed.
+ // Just left in for now for legacy behavior.
while ( ( *sb == ' ' ) ||
- ( *sb == '(' ) )
+ ( *sb == '\t' ) )
{
sb++;
}
- gchar const *se = sb + strlen(sb);
- while ( ( se[-1] == ' ' ) ||
- ( se[-1] == ')' ) )
- {
- se--;
- }
+ if ( *sb == '(' ) {
+ sb++;
+ while ( ( *sb == ' ' ) ||
+ ( *sb == '\t' ) )
+ {
+ sb++;
+ }
- if ( sb < se ) {
- return g_strndup(sb, se - sb);
- } else {
- return NULL;
+ gchar delim = ')';
+ if ( (*sb == '\'' || *sb == '"') ) {
+ delim = *sb;
+ sb++;
+ }
+ gchar const* se = sb + 1;
+ while ( *se && (*se != delim) ) {
+ se++;
+ }
+
+ // we found the delimiter
+ if ( *se ) {
+ if ( delim == ')' ) {
+ if ( endptr ) {
+ *endptr = se + 1;
+ }
+
+ // back up for any trailing whitespace
+ se--;
+ while ( ( se[-1] == ' ' ) ||
+ ( se[-1] == '\t' ) )
+ {
+ se--;
+ }
+
+ result = g_strndup(sb, se - sb + 1);
+ } else {
+ gchar const* tail = se + 1;
+ while ( ( *tail == ' ' ) ||
+ ( *tail == '\t' ) )
+ {
+ tail++;
+ }
+ if ( *tail == ')' ) {
+ if ( endptr ) {
+ *endptr = tail + 1;
+ }
+ result = g_strndup(sb, se - sb);
+ }
+ }
+ }
}
+
+ return result;
}
/*
fill-column:99
End:
*/
-// vim: filetype=cpp:expandtab:shiftwidth=4:tabstop=8:softtabstop=4:encoding=utf-8:textwidth=99 :
+// vim: filetype=cpp:expandtab:shiftwidth=4:tabstop=8:softtabstop=4:fileencoding=utf-8:textwidth=99 :