parse5 2.1.5 → 2.2.3

package/lib/common/html.js

@@ -84,7 +84,6 @@ var $ = exports.TAG_NAMES = {
     IMAGE: 'image',
     INPUT: 'input',
     IFRAME: 'iframe',
-    ISINDEX: 'isindex',
 
     KEYGEN: 'keygen',
 
@@ -215,14 +214,12 @@ SPECIAL_ELEMENTS[NS.HTML][$.HTML] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.IFRAME] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.IMG] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.INPUT] = true;
-SPECIAL_ELEMENTS[NS.HTML][$.ISINDEX] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.LI] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.LINK] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.LISTING] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.MAIN] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.MARQUEE] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.MENU] = true;
-SPECIAL_ELEMENTS[NS.HTML][$.MENUITEM] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.META] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.NAV] = true;
 SPECIAL_ELEMENTS[NS.HTML][$.NOEMBED] = true;

package/lib/index.js

@@ -1,108 +1,108 @@
-'use strict';
-
-var Parser = require('./parser'),
-    Serializer = require('./serializer');
-
-/** @namespace parse5 */
-
-/**
- * Parses an HTML string.
- * @function parse
- * @memberof parse5
- * @instance
- * @param {string} html - Input HTML string.
- * @param {ParserOptions} [options] - Parsing options.
- * @returns {ASTNode<Document>} document
- * @example
- * var parse5 = require('parse5');
- *
- * var document = parse5.parse('<!DOCTYPE html><html><head></head><body>Hi there!</body></html>');
- */
-exports.parse = function parse(html, options) {
-    var parser = new Parser(options);
-
-    return parser.parse(html);
-};
-
-/**
- * Parses an HTML fragment.
- * @function parseFragment
- * @memberof parse5
- * @instance
- * @param {ASTNode} [fragmentContext] - Parsing context element. If specified, given fragment
- * will be parsed as if it was set to the context element's `innerHTML` property.
- * @param {string} html - Input HTML fragment string.
- * @param {ParserOptions} [options] - Parsing options.
- * @returns {ASTNode<DocumentFragment>} documentFragment
- * @example
- * var parse5 = require('parse5');
- *
- * var documentFragment = parse5.parseFragment('<table></table>');
- *
- * // Parses the html fragment in the context of the parsed <table> element.
- * var trFragment = parser.parseFragment(documentFragment.childNodes[0], '<tr><td>Shake it, baby</td></tr>');
- */
-exports.parseFragment = function parseFragment(fragmentContext, html, options) {
-    if (typeof fragmentContext === 'string') {
-        options = html;
-        html = fragmentContext;
-        fragmentContext = null;
-    }
-
-    var parser = new Parser(options);
-
-    return parser.parseFragment(html, fragmentContext);
-};
-
-/**
- * Serializes an AST node to an HTML string.
- * @function serialize
- * @memberof parse5
- * @instance
- * @param {ASTNode} node - Node to serialize.
- * @param {SerializerOptions} [options] - Serialization options.
- * @returns {String} html
- * @example
- * var parse5 = require('parse5');
- *
- * var document = parse5.parse('<!DOCTYPE html><html><head></head><body>Hi there!</body></html>');
- *
- * // Serializes a document.
- * var html = parse5.serialize(document);
- *
- * // Serializes the <body> element content.
- * var bodyInnerHtml = parse5.serialize(document.childNodes[0].childNodes[1]);
- */
-exports.serialize = function (node, options) {
-    var serializer = new Serializer(node, options);
-
-    return serializer.serialize();
-};
-
-/**
- * Provides built-in tree adapters that can be used for parsing and serialization.
- * @var treeAdapters
- * @memberof parse5
- * @instance
- * @property {TreeAdapter} default - Default tree format for parse5.
- * @property {TreeAdapter} htmlparser2 - Quite popular [htmlparser2](https://github.com/fb55/htmlparser2) tree format
- * (e.g. used by [cheerio](https://github.com/MatthewMueller/cheerio) and [jsdom](https://github.com/tmpvar/jsdom)).
- * @example
- * var parse5 = require('parse5');
- *
- * // Uses the default tree adapter for parsing.
- * var document = parse5.parse('<div></div>', { treeAdapter: parse5.treeAdapters.default });
- *
- * // Uses the htmlparser2 tree adapter with the SerializerStream.
- * var serializer = new parse5.SerializerStream(node, { treeAdapter: parse5.treeAdapters.htmlparser2 });
- */
-exports.treeAdapters = {
-    default: require('./tree_adapters/default'),
-    htmlparser2: require('./tree_adapters/htmlparser2')
-};
-
-
-// Streaming
-exports.ParserStream = require('./parser/stream');
-exports.SerializerStream = require('./serializer/stream');
-exports.SAXParser = require('./sax');
+'use strict';
+
+var Parser = require('./parser'),
+    Serializer = require('./serializer');
+
+/** @namespace parse5 */
+
+/**
+ * Parses an HTML string.
+ * @function parse
+ * @memberof parse5
+ * @instance
+ * @param {string} html - Input HTML string.
+ * @param {ParserOptions} [options] - Parsing options.
+ * @returns {ASTNode<Document>} document
+ * @example
+ * var parse5 = require('parse5');
+ *
+ * var document = parse5.parse('<!DOCTYPE html><html><head></head><body>Hi there!</body></html>');
+ */
+exports.parse = function parse(html, options) {
+    var parser = new Parser(options);
+
+    return parser.parse(html);
+};
+
+/**
+ * Parses an HTML fragment.
+ * @function parseFragment
+ * @memberof parse5
+ * @instance
+ * @param {ASTNode} [fragmentContext] - Parsing context element. If specified, given fragment
+ * will be parsed as if it was set to the context element's `innerHTML` property.
+ * @param {string} html - Input HTML fragment string.
+ * @param {ParserOptions} [options] - Parsing options.
+ * @returns {ASTNode<DocumentFragment>} documentFragment
+ * @example
+ * var parse5 = require('parse5');
+ *
+ * var documentFragment = parse5.parseFragment('<table></table>');
+ *
+ * // Parses the html fragment in the context of the parsed <table> element.
+ * var trFragment = parser.parseFragment(documentFragment.childNodes[0], '<tr><td>Shake it, baby</td></tr>');
+ */
+exports.parseFragment = function parseFragment(fragmentContext, html, options) {
+    if (typeof fragmentContext === 'string') {
+        options = html;
+        html = fragmentContext;
+        fragmentContext = null;
+    }
+
+    var parser = new Parser(options);
+
+    return parser.parseFragment(html, fragmentContext);
+};
+
+/**
+ * Serializes an AST node to an HTML string.
+ * @function serialize
+ * @memberof parse5
+ * @instance
+ * @param {ASTNode} node - Node to serialize.
+ * @param {SerializerOptions} [options] - Serialization options.
+ * @returns {String} html
+ * @example
+ * var parse5 = require('parse5');
+ *
+ * var document = parse5.parse('<!DOCTYPE html><html><head></head><body>Hi there!</body></html>');
+ *
+ * // Serializes a document.
+ * var html = parse5.serialize(document);
+ *
+ * // Serializes the <body> element content.
+ * var bodyInnerHtml = parse5.serialize(document.childNodes[0].childNodes[1]);
+ */
+exports.serialize = function (node, options) {
+    var serializer = new Serializer(node, options);
+
+    return serializer.serialize();
+};
+
+/**
+ * Provides built-in tree adapters that can be used for parsing and serialization.
+ * @var treeAdapters
+ * @memberof parse5
+ * @instance
+ * @property {TreeAdapter} default - Default tree format for parse5.
+ * @property {TreeAdapter} htmlparser2 - Quite popular [htmlparser2](https://github.com/fb55/htmlparser2) tree format
+ * (e.g. used by [cheerio](https://github.com/MatthewMueller/cheerio) and [jsdom](https://github.com/tmpvar/jsdom)).
+ * @example
+ * var parse5 = require('parse5');
+ *
+ * // Uses the default tree adapter for parsing.
+ * var document = parse5.parse('<div></div>', { treeAdapter: parse5.treeAdapters.default });
+ *
+ * // Uses the htmlparser2 tree adapter with the SerializerStream.
+ * var serializer = new parse5.SerializerStream(node, { treeAdapter: parse5.treeAdapters.htmlparser2 });
+ */
+exports.treeAdapters = {
+    default: require('./tree_adapters/default'),
+    htmlparser2: require('./tree_adapters/htmlparser2')
+};
+
+
+// Streaming
+exports.ParserStream = require('./parser/stream');
+exports.SerializerStream = require('./serializer/stream');
+exports.SAXParser = require('./sax');

package/lib/location_info/parser_mixin.js

@@ -50,7 +50,10 @@ function setEndLocation(element, closingToken, treeAdapter) {
             };
         }
 
-        loc.endOffset = ctLocation.endOffset;
+        if (isClosingEndTag)
+            loc.endOffset = ctLocation.endOffset;
+        else
+            loc.endOffset = ctLocation.startOffset;
     }
 }
 

package/lib/location_info/tokenizer_mixin.js

@@ -1,168 +1,169 @@
-'use strict';
-
-var UNICODE = require('../common/unicode');
-
-//Aliases
-var $ = UNICODE.CODE_POINTS;
-
-
-exports.assign = function (tokenizer) {
-    //NOTE: obtain Tokenizer proto this way to avoid module circular references
-    var tokenizerProto = Object.getPrototypeOf(tokenizer),
-        tokenStartOffset = -1,
-        tokenCol = -1,
-        tokenLine = 1,
-        isEol = false,
-        lineStartPosStack = [0],
-        lineStartPos = 0,
-        col = -1,
-        line = 1;
-
-    function attachLocationInfo(token) {
-        /**
-         * @typedef {Object} LocationInfo
-         *
-         * @property {Number} line - One-based line index
-         * @property {Number} col - One-based column index
-         * @property {Number} startOffset - Zero-based first character index
-         * @property {Number} endOffset - Zero-based last character index
-         */
-        token.location = {
-            line: tokenLine,
-            col: tokenCol,
-            startOffset: tokenStartOffset,
-            endOffset: -1
-        };
-    }
-
-    //NOTE: patch consumption method to track line/col information
-    tokenizer._consume = function () {
-        var cp = tokenizerProto._consume.call(this);
-
-        //NOTE: LF should be in the last column of the line
-        if (isEol) {
-            isEol = false;
-            line++;
-            lineStartPosStack.push(this.preprocessor.pos);
-            lineStartPos = this.preprocessor.pos;
-        }
-
-        if (cp === $.LINE_FEED)
-            isEol = true;
-
-        col = this.preprocessor.pos - lineStartPos + 1;
-
-        return cp;
-    };
-
-    tokenizer._unconsume = function () {
-        tokenizerProto._unconsume.call(this);
-
-        while (lineStartPos > this.preprocessor.pos && lineStartPosStack.length > 1) {
-            lineStartPos = lineStartPosStack.pop();
-            line--;
-        }
-
-        col = this.preprocessor.pos - lineStartPos + 1;
-    };
-
-    //NOTE: patch token creation methods and attach location objects
-    tokenizer._createStartTagToken = function (tagNameFirstCh) {
-        tokenizerProto._createStartTagToken.call(this, tagNameFirstCh);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createEndTagToken = function (tagNameFirstCh) {
-        tokenizerProto._createEndTagToken.call(this, tagNameFirstCh);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createCommentToken = function () {
-        tokenizerProto._createCommentToken.call(this);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createDoctypeToken = function (doctypeNameFirstCh) {
-        tokenizerProto._createDoctypeToken.call(this, doctypeNameFirstCh);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createCharacterToken = function (type, ch) {
-        tokenizerProto._createCharacterToken.call(this, type, ch);
-        attachLocationInfo(this.currentCharacterToken);
-    };
-
-    tokenizer._createAttr = function (attrNameFirstCh) {
-        tokenizerProto._createAttr.call(this, attrNameFirstCh);
-        this.currentAttrLocation = {
-            line: line,
-            col: col,
-            startOffset: this.preprocessor.pos,
-            endOffset: -1
-        };
-    };
-
-    tokenizer._leaveAttrName = function (toState) {
-        tokenizerProto._leaveAttrName.call(this, toState);
-        this._attachCurrentAttrLocationInfo();
-    };
-
-    tokenizer._leaveAttrValue = function (toState) {
-        tokenizerProto._leaveAttrValue.call(this, toState);
-        this._attachCurrentAttrLocationInfo();
-    };
-
-    tokenizer._attachCurrentAttrLocationInfo = function () {
-        this.currentAttrLocation.endOffset = this.preprocessor.pos;
-
-        if (!this.currentToken.location.attrs)
-            this.currentToken.location.attrs = {};
-
-        /**
-         * @typedef {Object} StartTagLocationInfo
-         * @extends LocationInfo
-         *
-         * @property {Dictionary<String, LocationInfo>} attrs - Start tag attributes' location info.
-         */
-        this.currentToken.location.attrs[this.currentAttr.name] = this.currentAttrLocation;
-    };
-
-    //NOTE: patch token emission methods to determine end location
-    tokenizer._emitCurrentToken = function () {
-        //NOTE: if we have pending character token make it's end location equal to the
-        //current token's start location.
-        if (this.currentCharacterToken)
-            this.currentCharacterToken.location.endOffset = this.currentToken.location.startOffset;
-
-        this.currentToken.location.endOffset = this.preprocessor.pos + 1;
-        tokenizerProto._emitCurrentToken.call(this);
-    };
-
-    tokenizer._emitCurrentCharacterToken = function () {
-        //NOTE: if we have character token and it's location wasn't set in the _emitCurrentToken(),
-        //then set it's location at the current preprocessor position.
-        //We don't need to increment preprocessor position, since character token
-        //emission is always forced by the start of the next character token here.
-        //So, we already have advanced position.
-        if (this.currentCharacterToken && this.currentCharacterToken.location.endOffset === -1)
-            this.currentCharacterToken.location.endOffset = this.preprocessor.pos;
-
-        tokenizerProto._emitCurrentCharacterToken.call(this);
-    };
-
-    //NOTE: patch initial states for each mode to obtain token start position
-    Object.keys(tokenizerProto.MODE)
-
-        .map(function (modeName) {
-            return tokenizerProto.MODE[modeName];
-        })
-
-        .forEach(function (state) {
-            tokenizer[state] = function (cp) {
-                tokenStartOffset = this.preprocessor.pos;
-                tokenLine = line;
-                tokenCol = col;
-                tokenizerProto[state].call(this, cp);
-            };
-        });
-};
+'use strict';
+
+var UNICODE = require('../common/unicode');
+
+//Aliases
+var $ = UNICODE.CODE_POINTS;
+
+
+exports.assign = function (tokenizer) {
+    //NOTE: obtain Tokenizer proto this way to avoid module circular references
+    var tokenizerProto = Object.getPrototypeOf(tokenizer),
+        tokenStartOffset = -1,
+        tokenCol = -1,
+        tokenLine = 1,
+        isEol = false,
+        lineStartPosStack = [0],
+        lineStartPos = 0,
+        col = -1,
+        line = 1;
+
+    function attachLocationInfo(token) {
+        /**
+         * @typedef {Object} LocationInfo
+         *
+         * @property {Number} line - One-based line index
+         * @property {Number} col - One-based column index
+         * @property {Number} startOffset - Zero-based first character index
+         * @property {Number} endOffset - Zero-based last character index
+         */
+        token.location = {
+            line: tokenLine,
+            col: tokenCol,
+            startOffset: tokenStartOffset,
+            endOffset: -1
+        };
+    }
+
+    //NOTE: patch consumption method to track line/col information
+    tokenizer._consume = function () {
+        var cp = tokenizerProto._consume.call(this);
+
+        //NOTE: LF should be in the last column of the line
+        if (isEol) {
+            isEol = false;
+            line++;
+            lineStartPosStack.push(this.preprocessor.sourcePos);
+            lineStartPos = this.preprocessor.sourcePos;
+        }
+
+        if (cp === $.LINE_FEED)
+            isEol = true;
+
+        col = this.preprocessor.sourcePos - lineStartPos + 1;
+
+        return cp;
+    };
+
+    tokenizer._unconsume = function () {
+        tokenizerProto._unconsume.call(this);
+        isEol = false;
+
+        while (lineStartPos > this.preprocessor.sourcePos && lineStartPosStack.length > 1) {
+            lineStartPos = lineStartPosStack.pop();
+            line--;
+        }
+
+        col = this.preprocessor.sourcePos - lineStartPos + 1;
+    };
+
+    //NOTE: patch token creation methods and attach location objects
+    tokenizer._createStartTagToken = function () {
+        tokenizerProto._createStartTagToken.call(this);
+        attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createEndTagToken = function () {
+        tokenizerProto._createEndTagToken.call(this);
+        attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createCommentToken = function () {
+        tokenizerProto._createCommentToken.call(this);
+        attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createDoctypeToken = function (initialName) {
+        tokenizerProto._createDoctypeToken.call(this, initialName);
+        attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createCharacterToken = function (type, ch) {
+        tokenizerProto._createCharacterToken.call(this, type, ch);
+        attachLocationInfo(this.currentCharacterToken);
+    };
+
+    tokenizer._createAttr = function (attrNameFirstCh) {
+        tokenizerProto._createAttr.call(this, attrNameFirstCh);
+        this.currentAttrLocation = {
+            line: line,
+            col: col,
+            startOffset: this.preprocessor.sourcePos,
+            endOffset: -1
+        };
+    };
+
+    tokenizer._leaveAttrName = function (toState) {
+        tokenizerProto._leaveAttrName.call(this, toState);
+        this._attachCurrentAttrLocationInfo();
+    };
+
+    tokenizer._leaveAttrValue = function (toState) {
+        tokenizerProto._leaveAttrValue.call(this, toState);
+        this._attachCurrentAttrLocationInfo();
+    };
+
+    tokenizer._attachCurrentAttrLocationInfo = function () {
+        this.currentAttrLocation.endOffset = this.preprocessor.sourcePos;
+
+        if (!this.currentToken.location.attrs)
+            this.currentToken.location.attrs = {};
+
+        /**
+         * @typedef {Object} StartTagLocationInfo
+         * @extends LocationInfo
+         *
+         * @property {Dictionary<String, LocationInfo>} attrs - Start tag attributes' location info.
+         */
+        this.currentToken.location.attrs[this.currentAttr.name] = this.currentAttrLocation;
+    };
+
+    //NOTE: patch token emission methods to determine end location
+    tokenizer._emitCurrentToken = function () {
+        //NOTE: if we have pending character token make it's end location equal to the
+        //current token's start location.
+        if (this.currentCharacterToken)
+            this.currentCharacterToken.location.endOffset = this.currentToken.location.startOffset;
+
+        this.currentToken.location.endOffset = this.preprocessor.sourcePos + 1;
+        tokenizerProto._emitCurrentToken.call(this);
+    };
+
+    tokenizer._emitCurrentCharacterToken = function () {
+        //NOTE: if we have character token and it's location wasn't set in the _emitCurrentToken(),
+        //then set it's location at the current preprocessor position.
+        //We don't need to increment preprocessor position, since character token
+        //emission is always forced by the start of the next character token here.
+        //So, we already have advanced position.
+        if (this.currentCharacterToken && this.currentCharacterToken.location.endOffset === -1)
+            this.currentCharacterToken.location.endOffset = this.preprocessor.sourcePos;
+
+        tokenizerProto._emitCurrentCharacterToken.call(this);
+    };
+
+    //NOTE: patch initial states for each mode to obtain token start position
+    Object.keys(tokenizerProto.MODE)
+
+        .map(function (modeName) {
+            return tokenizerProto.MODE[modeName];
+        })
+
+        .forEach(function (state) {
+            tokenizer[state] = function (cp) {
+                tokenStartOffset = this.preprocessor.sourcePos;
+                tokenLine = line;
+                tokenCol = col;
+                tokenizerProto[state].call(this, cp);
+            };
+        });
+};