Issue 3143 - Filter elements with :-abp-has()

chrome/content/elemHideEmulation.js

 /*
  * This file is part of Adblock Plus <https://adblockplus.org/>,
  * Copyright (C) 2006-2017 eyeo GmbH
  *
  * Adblock Plus is free software: you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 3 as
  * published by the Free Software Foundation.
  *
  * Adblock Plus is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
  * along with Adblock Plus.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 /* globals filterToRegExp */
 
 "use strict";
 
-let propertySelectorRegExp = /\[-abp-properties=(["'])([^"']+)\1\]/;
-let pseudoClassHasSelectorRegExp = /:-abp-has\((.*)\)/;
-let pseudoClassPropsSelectorRegExp = /:-abp-properties\((["'])([^"']+)\1\)/;
+const abpSelectorRegexp = /:-abp-([\w-]+)\(/i;
+
+let reportError = () => {};
 
 function splitSelector(selector)
 {
   if (selector.indexOf(",") == -1)
     return [selector];
 
   let selectors = [];
   let start = 0;
   let level = 0;
   let sep = "";
@@ skipping 19 matching lines @@
         selectors.push(selector.substring(start, i));
         start = i + 1;
       }
     }
   }
 
   selectors.push(selector.substring(start));
   return selectors;
 }
 
-// 1 base index like for :nth-child()
+/** Return position of node from parent.
+ * @param {Node} node the node to find the position of.
+ * @return {number} One-based index like for :nth-child(), or 0 on error.
+ */
 function positionInParent(node)
 {
-  let parentNode = node ? node.parentNode : null;
-  if (parentNode == null)
-    return 0;
-
-  let {children} = parentNode;
-  if (!children)
-    return 0;
-  let i = 0;
-  for (i = 0; i < children.length; i++)
+  let {children} = node.parentNode;
+  for (let i = 0; i < children.length; i++)
     if (children[i] == node)
-      break;
-  return i + 1;
+      return i + 1;
+  return 0;
 }
 
 function makeSelector(node, selector)
 {
+  if (!node.parentElement)
+  {
+    let newSelector = ":root";
+    if (selector)
+      newSelector += " > " + selector;
+    return newSelector;
+  }
   let idx = positionInParent(node);
   if (idx > 0)
   {
-    let newSelector = `:nth-child(${idx}) `;
-    if (selector != "")
-      newSelector += "> ";
-    return makeSelector(node.parentNode, newSelector + selector);
+    let newSelector = `${node.tagName}:nth-child(${idx})`;
+    if (selector)
+      newSelector += " > " + selector;
+    return makeSelector(node.parentElement, newSelector);
   }
 
   return selector;
 }
 
-function matchChildren(e, selector)
-{
-  let newSelector = makeSelector(e, "");
-  newSelector += selector;
-
-  return document.querySelector(newSelector) != null;
-}
-
-function selectChildren(e, selector)
-{
-  let newSelector = makeSelector(e, "");
-  newSelector += selector;
-
-  return document.querySelectorAll(newSelector);
-}
-
-function parsePattern(pattern)
-{
-  let {selector} = pattern;
-  let match = pseudoClassHasSelectorRegExp.exec(selector);
-  if (match)
-    return {
-      type: "has",
-      text: pattern.text,
-      elementMatcher: new PseudoHasMatcher(match[1]),
-      prefix: selector.substr(0, match.index).trim(),
-      suffix: selector.substr(match.index + match[0].length).trim()
-    };
-  match = pseudoClassPropsSelectorRegExp.exec(selector);
-
+function parseSelectorContent(content, startIndex)
+{
+  let parens = 1;
+  let quote = null;
+  let i = startIndex;
+  for (; i < content.length; i++)
+  {
+    let c = content[i];
+    if (c == "\\")
+    {
+      // Ignore escaped characters
+      i++;
+    }
+    else if (quote)
+    {
+      if (c == quote)
+        quote = null;
+    }
+    else if (c == "'" || c == '"')
+      quote = c;
+    else if (c == "(")
+      parens++;
+    else if (c == ")")
+    {
+      parens--;
+      if (parens == 0)
+        break;
+    }
+  }
+
+  if (parens > 0)
+    return null;
+  return {text: content.substring(startIndex, i), end: i};
+}
+
+/** Parse the selector
+ * @param {string} selector the selector to parse
+ * @return {Object} selectors is an array of objects,
+ * or null in case of errors. hide is true if we'll hide
+ * elements instead of styles..
+ */
+function parseSelector(selector)
+{
+  if (selector.length == 0)
+    return [];
+
+  let match = abpSelectorRegexp.exec(selector);
   if (!match)
-    match = propertySelectorRegExp.exec(selector);
-  if (match)
-  {
-    let regexpString;
-    let propertyExpression = match[2];
-    if (propertyExpression.length >= 2 && propertyExpression[0] == "/" &&
-        propertyExpression[propertyExpression.length - 1] == "/")
-      regexpString = propertyExpression.slice(1, -1)
+    return [new PlainSelector(selector)];
+
+  let selectors = [];
+  if (match.index > 0)
+    selectors.push(new PlainSelector(selector.substr(0, match.index)));
+
+  let startIndex = match.index + match[0].length;
+  let content = parseSelectorContent(selector, startIndex);
+  if (!content)
+  {
+    reportError(new SyntaxError("Failed to parse Adblock Plus " +
+                                `selector ${selector}, ` +
+                                "due to unmatched parentheses."));
+    return null;
+  }
+  if (match[1] == "properties")
+    selectors.push(new PropsSelector(content.text));
+  else if (match[1] == "has")
+  {
+    let hasSelector = new HasSelector(content.text);
+    if (!hasSelector.valid())
+      return null;
+    selectors.push(hasSelector);
+  }
+  else
+  {
+    // this is an error, can't parse selector.
+    reportError(new SyntaxError("Failed to parse Adblock Plus " +
+                                `selector ${selector}, invalid ` +
+                                `pseudo-class :-abp-${match[1]}().`));
+    return null;
+  }
+
+  let suffix = parseSelector(selector.substr(content.end + 1));
+  if (suffix == null)
+    return null;
+
+  selectors.push(...suffix);
+
+  return selectors;
+}
+
+/** Stringified style objects
+ * @typedef {Object} StringifiedStyle
+ * @property {string} style CSS style represented by a string.
+ * @property {string[]} subSelectors selectors the CSS properties apply to.
+ */
+
+/**
+ * Produce a string representation of the stylesheet entry.
+ * @param {CSSStyleRule} rule the CSS style rule.
+ * @return {StringifiedStyle} the stringified style.
+ */
+function stringifyStyle(rule)
+{
+  let styles = [];
+  for (let i = 0; i < rule.style.length; i++)
+  {
+    let property = rule.style.item(i);
+    let value = rule.style.getPropertyValue(property);
+    let priority = rule.style.getPropertyPriority(property);
+    styles.push(`${property}: ${value}${priority ? " !" + priority : ""};`);
+  }
+  styles.sort();
+  return {
+    style: styles.join(" "),
+    subSelectors: splitSelector(rule.selectorText)
+  };
+}
+
+function* evaluate(chain, index, prefix, subtree, styles)
+{
+  if (index >= chain.length)
+  {
+    yield prefix;
+    return;
+  }
+  for (let [selector, element] of
+       chain[index].getSelectors(prefix, subtree, styles))
+    yield* evaluate(chain, index + 1, selector, element, styles);
+}
+
+function PlainSelector(selector)
+{
+  this._selector = selector;
+}
+
+PlainSelector.prototype = {
+  /**
+   * Generator function returning a pair of selector
+   * string and subtree.
+   * @param {string} prefix the prefix for the selector.
+   * @param {Node} subtree the subtree we work on.
+   * @param {StringifiedStyle[]} styles the stringified style objects.
+   */
+  *getSelectors(prefix, subtree, styles)
+  {
+    yield [prefix + this._selector, subtree];
+  }
+};
+
+const incompletePrefixRegexp = /[\s>+~]$/;
+
+function HasSelector(selector)
+{
+  this._innerSelectors = parseSelector(selector);
+}
+
+HasSelector.prototype = {
+  requiresHiding: true,
+
+  valid()
+  {
+    return this._innerSelectors != null;
+  },
+
+  *getSelectors(prefix, subtree, styles)
+  {
+    for (let element of this.getElements(prefix, subtree, styles))
+      yield [makeSelector(element, ""), element];
+  },
+
+  /**
+   * Generator function returning selected elements.
+   * @param {string} prefix the prefix for the selector.
+   * @param {Node} subtree the subtree we work on.
+   * @param {StringifiedStyle[]} styles the stringified style objects.
+   */
+  *getElements(prefix, subtree, styles)
+  {
+    let actualPrefix = (!prefix || incompletePrefixRegexp.test(prefix)) ?
+        prefix + "*" : prefix;
+    let elements = subtree.querySelectorAll(actualPrefix);
+    for (let element of elements)
+    {
+      let newPrefix = makeSelector(element, "");
+      let iter = evaluate(this._innerSelectors, 0, newPrefix + " ",
+                          element, styles);
+      for (let selector of iter)
+        // we insert a space between the two. It becomes a no-op if selector
+        // doesn't have a combinator
+        if (subtree.querySelector(selector))
+          yield element;
+    }
+  }
+};
+
+function PropsSelector(propertyExpression)
+{
+  let regexpString;
+  if (propertyExpression.length >= 2 && propertyExpression[0] == "/" &&
+      propertyExpression[propertyExpression.length - 1] == "/")
+  {
+    regexpString = propertyExpression.slice(1, -1)
       .replace("\\x7B ", "{").replace("\\x7D ", "}");
-    else
-      regexpString = filterToRegExp(propertyExpression);
-    return {
-      type: "props",
-      text: pattern.text,
-      regexp: new RegExp(regexpString, "i"),
-      prefix: selector.substr(0, match.index),
-      suffix: selector.substr(match.index + match[0].length)
-    };
-  }
-}
-
-function matchStyleProps(style, rule, pattern, selectors, filters)
-{
-  if (pattern.regexp.test(style))
-  {
-    let subSelectors = splitSelector(rule.selectorText);
-    for (let i = 0; i < subSelectors.length; i++)
-    {
-      let subSelector = subSelectors[i];
-      selectors.push(pattern.prefix + subSelector + pattern.suffix);
-      filters.push(pattern.text);
-    }
-  }
-}
-
-function findPropsSelectors(stylesheet, patterns, selectors, filters)
-{
-  let rules = stylesheet.cssRules;
-  if (!rules)
-    return;
-
-  for (let rule of rules)
-  {
-    if (rule.type != rule.STYLE_RULE)
-      continue;
-
-    let style = stringifyStyle(rule.style);
-    for (let pattern of patterns)
-    {
-      matchStyleProps(style, rule, pattern, selectors, filters);
-    }
-  }
-}
-
-/**
- * Match the selector @pattern containing :-abp-has() starting from @node.
- * @param {string} pattern - the pattern to match
- * @param {object} node - the top level node.
- * @param {object} stylesheets - the stylesheets to check from.
- * @param {array} elements - elements that match
- * @param {array} filters - filters that match
- */
-function pseudoClassHasMatch(pattern, node, stylesheets, elements, filters)
-{
-  // select element for the prefix pattern. Or just use node.
-  let haveElems = pattern.prefix ? node.querySelectorAll(pattern.prefix) :
-      [node];
-  for (let elem of haveElems)
-  {
-    let matched = pattern.elementMatcher.match(elem, stylesheets);
-    if (!matched)
-      continue;
-
-    if (pattern.suffix)
-    {
-      let subElements = selectChildren(elem, pattern.suffix);
-      if (subElements)
-      {
-        for (let subElement of subElements)
-        {
-          elements.push(subElement);
-          filters.push(pattern.text);
-        }
-      }
-    }
-    else
-    {
-      elements.push(elem);
-      filters.push(pattern.text);
-    }
-  }
-}
-
-function stringifyStyle(style)
-{
-  let styles = [];
-  for (let i = 0; i < style.length; i++)
-  {
-    let property = style.item(i);
-    let value = style.getPropertyValue(property);
-    let priority = style.getPropertyPriority(property);
-    styles.push(property + ": " + value + (priority ? " !" + priority : "") +
-                ";");
-  }
-  styles.sort();
-  return styles.join(" ");
-}
-
-/** matcher for the pseudo class :-abp-has()
- * For those browser that don't have it yet.
- * @param {string} selector - the inner selector.
- */
-function PseudoHasMatcher(selector)
-{
-  this.hasSelector = selector;
-  this.parsed = parsePattern({selector});
-}
-
-PseudoHasMatcher.prototype = {
-  match(elem, stylesheets)
-  {
-    let selectors = [];
-
-    if (this.parsed)
-    {
-      let filters = []; // don't need this, we have a partial filter.
-      if (this.parsed.type == "has")
-      {
-        let child = elem.firstChild;
-        while (child)
-        {
-          if (child.nodeType === Node.ELEMENT_NODE)
-          {
-            let matches = [];
-            pseudoClassHasMatch(this.parsed, child, stylesheets, matches,
-                                filters);
-            if (matches.length > 0)
-              return true;
-          }
-          child = child.nextSibling;
-        }
-        return false;
-      }
-      if (this.parsed.type == "props")
-        for (let stylesheet of stylesheets)
-          findPropsSelectors(stylesheet, [this.parsed], selectors, filters);
-    }
-    else
-      selectors = [this.hasSelector];
-
-    let matched = false;
-    // look up for all elements that match the :-abp-has().
-    for (let selector of selectors)
-      try
-      {
-        matched = matchChildren(elem, selector);
-        if (matched)
-          break;
-      }
-      catch (e)
-      {
-        console.error("Exception with querySelector()", selector, e);
-      }
-
-    return matched;
+  }
+  else
+    regexpString = filterToRegExp(propertyExpression);
+
+  this._regexp = new RegExp(regexpString, "i");
+}
+
+PropsSelector.prototype = {
+  *findPropsSelectors(styles, prefix, regexp)
+  {
+    for (let style of styles)
+      if (regexp.test(style.style))
+        for (let subSelector of style.subSelectors)
+          yield prefix + subSelector;
+  },
+
+  *getSelectors(prefix, subtree, styles)
+  {
+    for (let selector of this.findPropsSelectors(styles, prefix, this._regexp))
+      yield [selector, subtree];
   }
 };
 
 function ElemHideEmulation(window, getFiltersFunc, addSelectorsFunc,
-                           hideElementsFunc)
+                           hideElemsFunc)
 {
   this.window = window;
   this.getFiltersFunc = getFiltersFunc;
   this.addSelectorsFunc = addSelectorsFunc;
-  this.hideElementsFunc = hideElementsFunc;
+  this.hideElemsFunc = hideElemsFunc;
 }
 
 ElemHideEmulation.prototype = {
-
   isSameOrigin(stylesheet)
   {
     try
     {
       return new URL(stylesheet.href).origin == this.window.location.origin;
     }
     catch (e)
     {
       // Invalid URL, assume that it is first-party.
       return true;
     }
   },
 
-  findPseudoClassHasElements(node, stylesheets, elements, filters)
-  {
-    for (let pattern of this.pseudoHasPatterns)
-      pseudoClassHasMatch(pattern, node, stylesheets, elements, filters);
-  },
-
   addSelectors(stylesheets)
   {
     let selectors = [];
-    let filters = [];
+    let selectorFilters = [];
+
+    let elements = [];
+    let elementFilters = [];
+
+    let cssStyles = [];
+
     for (let stylesheet of stylesheets)
     {
       // Explicitly ignore third-party stylesheets to ensure consistent behavior
       // between Firefox and Chrome.
       if (!this.isSameOrigin(stylesheet))
         continue;
-      findPropsSelectors(stylesheet, this.propSelPatterns, selectors, filters);
-    }
-    this.addSelectorsFunc(selectors, filters);
-  },
-
-  hideElements(stylesheets)
-  {
-    let elements = [];
-    let filters = [];
-    this.findPseudoClassHasElements(document, stylesheets, elements, filters);
-    this.hideElementsFunc(elements, filters);
+
+      let rules = stylesheet.cssRules;
+      if (!rules)
+        continue;
+
+      for (let rule of rules)
+      {
+        if (rule.type != rule.STYLE_RULE)
+          continue;
+
+        cssStyles.push(stringifyStyle(rule));
+      }
+    }
+
+    let {document} = this.window;
+    for (let pattern of this.patterns)
+    {
+      for (let selector of evaluate(pattern.selectors,
+                                    0, "", document, cssStyles))
+      {
+        if (!pattern.selectors.some(s => s.requiresHiding))
+        {
+          selectors.push(selector);
+          selectorFilters.push(pattern.text);
+        }
+        else
+        {
+          for (let element of document.querySelectorAll(selector))
+          {
+            elements.push(element);
+            elementFilters.push(pattern.text);
+          }
+        }
+      }
+    }
+
+    this.addSelectorsFunc(selectors, selectorFilters);
+    this.hideElemsFunc(elements, elementFilters);
   },
 
   onLoad(event)
   {
     let stylesheet = event.target.sheet;
     if (stylesheet)
       this.addSelectors([stylesheet]);
-    this.hideElements([stylesheet]);
   },
 
   apply()
   {
     this.getFiltersFunc(patterns =>
     {
-      this.propSelPatterns = [];
-      this.pseudoHasPatterns = [];
-
+      let oldReportError = reportError;
+      reportError = error => this.window.console.error(error);
+
+      this.patterns = [];
       for (let pattern of patterns)
       {
-        let parsed = parsePattern(pattern);
-        if (parsed == undefined)
-          continue;
-
-        if (parsed.type == "props")
-        {
-          this.propSelPatterns.push(parsed);
-        }
-        else if (parsed.type == "has")
-        {
-          this.pseudoHasPatterns.push(parsed);
-        }
+        let selectors = parseSelector(pattern.selector);
+        if (selectors != null && selectors.length > 0)
+          this.patterns.push({selectors, text: pattern.text});
       }
 
-      if (this.pseudoHasPatterns.length > 0 || this.propSelPatterns.length > 0)
+      if (this.patterns.length > 0)
       {
         let {document} = this.window;
         this.addSelectors(document.styleSheets);
-        this.hideElements(document.styleSheets);
         document.addEventListener("load", this.onLoad.bind(this), true);
       }
+      reportError = oldReportError;
     });
   }
 };