Improve urlskip= filter option

gorhill · gorhill · commit a7aa755f1884 · 2024-12-24T08:59:28.000-05:00
New step: `#`, to extract the hash part of a URL. Example, URL: https://example.com/#aHR0cHM6Ly9naXRodWIuY29tL3VCbG9ja09yaWdpbi8= Filter: ||example.com^$urlskip=# -base64 As a result, navigate to https://github.com/uBlockOrigin/
diff --git a/src/js/urlskip.js b/src/js/urlskip.js
@@ -39,6 +39,8 @@
  * `&i`: extract the name of the parameter at position `i` as the current
  *   string. The position is 1-based.
  * 
+ * `#`: extract the hash as the current string.
+ * 
  * `/.../`: extract the first capture group of a regex as the current string.
  * 
  * `+https`: prepend the current string with `https://`.
@@ -77,6 +79,12 @@ export function urlSkip(url, blocked, steps, directive = {}) {
         for ( const step of steps ) {
             const urlin = urlout;
             const c0 = step.charCodeAt(0);
+            // Extract from hash
+            if ( c0 === 0x23 && step === '#' ) { // #
+                const pos = urlin.indexOf('#');
+                urlout = pos !== -1 ? urlin.slice(pos+1) : '';
+                continue;
+            }
             // Extract from URL parameter name at position i
             if ( c0 === 0x26 ) { // &
                 const i = (parseInt(step.slice(1)) || 0) - 1;
@@ -88,14 +96,14 @@ export function urlSkip(url, blocked, steps, directive = {}) {
                 continue;
             }
             // Enforce https
-            if ( c0 === 0x2B && step === '+https' ) {
+            if ( c0 === 0x2B && step === '+https' ) { // +
                 const s = urlin.replace(/^https?:\/\//, '');
                 if ( /^[\w-]:\/\//.test(s) ) { return; }
                 urlout = `https://${s}`;
                 continue;
             }
             // Decode
-            if ( c0 === 0x2D ) {
+            if ( c0 === 0x2D ) { // -
                 // Base64
                 if ( step === '-base64' ) {
                     urlout = self.atob(urlin);
