tree walk based replacement for .textContent. more efficient cleanStyles

vsiv · Nov 19, 2010 · b367442 · b367442
1 parent b7fa45d
commit b367442
Showing 1 changed file with 68 additions and 0 deletions.
diff --git a/lib/readability.js b/lib/readability.js
@@ -1916,6 +1916,74 @@ var jsdom = require('jsdom'),
         getArticleFooter: function () {
             return document.createElement("DIV");
         },
+
+        // hundredfold faster
+        // use native string.trim
+        // jsdom's implementation of textContent is innerHTML + strip tags + HTMLDecode
+        // here we replace it with an optimized tree walker
+        getInnerText: function (e, normalizeSpaces) {
+            if (normalizeSpaces === undefined) normalizeSpaces = true;
+
+            function TextWalker(node, func) {
+                function walk(cur) {
+                    var children, len, i;
+                    if (cur.nodeType == 3) {
+                        func(cur);
+                        return;
+                    } else if (cur.nodeType != 1) {
+                        return;
+                    }
+
+                    children = cur.childNodes;
+                    for (i = 0, len = children.length; i < len; i++) {
+                        walk(children[i]);
+                    }
+                }
+                walk(node);
+            }
+
+            var textContent = '';
+            TextWalker(e, function(cur) {
+                textContent += cur.nodeValue;
+            });
+            textContent = textContent.trim();
+            //var textContent = e.textContent.trim();
+
+            if(normalizeSpaces) {
+                return textContent.replace( readability.regexps.normalize, " "); }
+            else {
+                return textContent; 
+            }
+        },
+
+        cleanStyles: function (e) {
+            e = e || document;
+            // var all = e.getElementsByTagName('*'), i, len, node;
+            // for (i = 0, len = all.length; i < len; i++) {
+            //     node = all[i];
+            //     if (node.className != 'readability-styled') {
+            //         node.removeAttribute("style");
+            //     }
+            // }
+            // return;
+
+            function walk(cur) {
+                var children, i, l;
+
+                if (cur.nodeType == 1) {
+                    if (cur.className != 'readability-styled') {
+                        cur.removeAttribute("style");
+                    }
+
+                    children = cur.childNodes;
+                    for (i = 0, l = children.length; i < l; i++) {
+                        walk(children[i]);
+                    }
+                }
+            }
+            walk(e);
+        },
+
         //// new methods ///
         reset: function() {
             var z = this;