Fix parsing of AdGuard's #$?#-based cosmetic filters

As reported in the following commit: - AdguardTeam/AdguardFilters@4fe02d73cee6
gorhill · Jun 29, 2020 · cf1e6fd · cf1e6fd
1 parent 00134a9
commit cf1e6fd
Show file tree

Hide file tree

Showing 2 changed files with 19 additions and 9 deletions.
diff --git a/src/js/codemirror/ubo-static-filtering.js b/src/js/codemirror/ubo-static-filtering.js
@@ -1,7 +1,7 @@
 /*******************************************************************************
 
     uBlock Origin - a browser extension to block requests.
-    Copyright (C) 2018 Raymond Hill
+    Copyright (C) 2018-present Raymond Hill
 
     This program is free software: you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
@@ -26,7 +26,7 @@
 CodeMirror.defineMode("ubo-static-filtering", function() {
     var reComment1 = /^\s*!/;
     var reComment2 = /^\s*#/;
-    var reExt = /^(\s*[^#]*)(#(?:#|@#|\$#|@\$#|\?#|@\?#))(.+)$/;
+    var reExt = /^(\s*[^#]*)(#@?(?:\$\??|\?)?#)(.+)$/;
     var reNet = /^(.*?)(?:(\$)([^$]+)?)?$/;
     var reNetAllow = /^\s*@@/;
     var lineStyle = null;

diff --git a/src/js/static-ext-filtering.js b/src/js/static-ext-filtering.js
@@ -678,11 +678,22 @@
             if ( rpos === -1 ) { return false; }
         }
 
+        // https://github.com/AdguardTeam/AdguardFilters/commit/4fe02d73cee6
+        //   AdGuard also uses `$?` to force inline-based style rather than
+        //   stylesheet-based style.
         // Coarse-check that the anchor is valid.
-        // `##`: l = 1
-        // `#@#`, `#$#`, `#%#`, `#?#`: l = 2
-        // `#@$#`, `#@%#`, `#@?#`: l = 3
-        if ( (rpos - lpos) > 3 ) { return false; }
+        // `##`: l === 1
+        // `#@#`, `#$#`, `#%#`, `#?#`: l === 2
+        // `#@$#`, `#@%#`, `#@?#`, `#$?#`: l === 3
+        // `#@$?#`: l === 4
+        const anchorLen = rpos - lpos;
+        if ( anchorLen > 4 ) { return false; }
+        if (
+            anchorLen > 1 &&
+            /^@?(?:\$\??|%|\?)$/.test(raw.slice(lpos + 1, rpos)) === false
+        ) {
+            return false;
+        }
 
         // Extract the selector.
         let suffix = raw.slice(rpos + 1).trim();
@@ -700,9 +711,8 @@
         if ( cCode !== 0x23 /* '#' */ && cCode !== 0x40 /* '@' */ ) {
             // Adguard's scriptlet injection: not supported.
             if ( cCode === 0x25 /* '%' */ ) { return true; }
-            // Not a known extended filter.
-            if ( cCode !== 0x24 /* '$' */ && cCode !== 0x3F /* '?' */ ) {
-                return false;
+            if ( cCode === 0x3F /* '?' */ && anchorLen > 2 ) {
+                cCode = raw.charCodeAt(rpos - 2);
             }
             // Adguard's style injection: translate to uBO's format.
             if ( cCode === 0x24 /* '$' */ ) {