first commit

This commit is contained in:
2026-08-07 12:53:18 +09:00
commit 0b0415f0f4
1523 changed files with 496649 additions and 0 deletions

View File

@@ -0,0 +1,822 @@
/**
* DomParser.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/**
* This class parses HTML code into a DOM like structure of nodes it will remove redundant whitespace and make
* sure that the node tree is valid according to the specified schema.
* So for example: <p>a<p>b</p>c</p> will become <p>a</p><p>b</p><p>c</p>
*
* @example
* var parser = new tinymce.html.DomParser({validate: true}, schema);
* var rootNode = parser.parse('<h1>content</h1>');
*
* @class tinymce.html.DomParser
* @version 3.4
*/
define("tinymce/html/DomParser", [
"tinymce/html/Node",
"tinymce/html/Schema",
"tinymce/html/SaxParser",
"tinymce/util/Tools"
], function(Node, Schema, SaxParser, Tools) {
var makeMap = Tools.makeMap, each = Tools.each, explode = Tools.explode, extend = Tools.extend;
/**
* Constructs a new DomParser instance.
*
* @constructor
* @method DomParser
* @param {Object} settings Name/value collection of settings. comment, cdata, text, start and end are callbacks.
* @param {tinymce.html.Schema} schema HTML Schema class to use when parsing.
*/
return function(settings, schema) {
var self = this, nodeFilters = {}, attributeFilters = [], matchedNodes = {}, matchedAttributes = {};
settings = settings || {};
settings.validate = "validate" in settings ? settings.validate : true;
settings.root_name = settings.root_name || 'body';
self.schema = schema = schema || new Schema();
function fixInvalidChildren(nodes) {
var ni, node, parent, parents, newParent, currentNode, tempNode, childNode, i;
var nonEmptyElements, nonSplitableElements, textBlockElements, specialElements, sibling, nextNode;
nonSplitableElements = makeMap('tr,td,th,tbody,thead,tfoot,table');
nonEmptyElements = schema.getNonEmptyElements();
textBlockElements = schema.getTextBlockElements();
specialElements = schema.getSpecialElements();
for (ni = 0; ni < nodes.length; ni++) {
node = nodes[ni];
// Already removed or fixed
if (!node.parent || node.fixed) {
continue;
}
// If the invalid element is a text block and the text block is within a parent LI element
// Then unwrap the first text block and convert other sibling text blocks to LI elements similar to Word/Open Office
if (textBlockElements[node.name] && node.parent.name == 'li') {
// Move sibling text blocks after LI element
sibling = node.next;
while (sibling) {
if (textBlockElements[sibling.name]) {
sibling.name = 'li';
sibling.fixed = true;
node.parent.insert(sibling, node.parent);
} else {
break;
}
sibling = sibling.next;
}
// Unwrap current text block
node.unwrap(node);
continue;
}
// Get list of all parent nodes until we find a valid parent to stick the child into
parents = [node];
for (parent = node.parent; parent && !schema.isValidChild(parent.name, node.name) &&
!nonSplitableElements[parent.name]; parent = parent.parent) {
parents.push(parent);
}
// Found a suitable parent
if (parent && parents.length > 1) {
// Reverse the array since it makes looping easier
parents.reverse();
// Clone the related parent and insert that after the moved node
newParent = currentNode = self.filterNode(parents[0].clone());
// Start cloning and moving children on the left side of the target node
for (i = 0; i < parents.length - 1; i++) {
if (schema.isValidChild(currentNode.name, parents[i].name)) {
tempNode = self.filterNode(parents[i].clone());
currentNode.append(tempNode);
} else {
tempNode = currentNode;
}
for (childNode = parents[i].firstChild; childNode && childNode != parents[i + 1];) {
nextNode = childNode.next;
tempNode.append(childNode);
childNode = nextNode;
}
currentNode = tempNode;
}
if (!newParent.isEmpty(nonEmptyElements)) {
parent.insert(newParent, parents[0], true);
parent.insert(node, newParent);
} else {
parent.insert(node, parents[0], true);
}
// Check if the element is empty by looking through it's contents and special treatment for <p><br /></p>
parent = parents[0];
if (parent.isEmpty(nonEmptyElements) || parent.firstChild === parent.lastChild && parent.firstChild.name === 'br') {
parent.empty().remove();
}
} else if (node.parent) {
// If it's an LI try to find a UL/OL for it or wrap it
if (node.name === 'li') {
sibling = node.prev;
if (sibling && (sibling.name === 'ul' || sibling.name === 'ul')) {
sibling.append(node);
continue;
}
sibling = node.next;
if (sibling && (sibling.name === 'ul' || sibling.name === 'ul')) {
sibling.insert(node, sibling.firstChild, true);
continue;
}
node.wrap(self.filterNode(new Node('ul', 1)));
continue;
}
// Try wrapping the element in a DIV
if (schema.isValidChild(node.parent.name, 'div') && schema.isValidChild('div', node.name)) {
node.wrap(self.filterNode(new Node('div', 1)));
} else {
// We failed wrapping it, then remove or unwrap it
if (specialElements[node.name]) {
node.empty().remove();
} else {
node.unwrap();
}
}
}
}
}
/**
* Runs the specified node though the element and attributes filters.
*
* @method filterNode
* @param {tinymce.html.Node} Node the node to run filters on.
* @return {tinymce.html.Node} The passed in node.
*/
self.filterNode = function(node) {
var i, name, list;
// Run element filters
if (name in nodeFilters) {
list = matchedNodes[name];
if (list) {
list.push(node);
} else {
matchedNodes[name] = [node];
}
}
// Run attribute filters
i = attributeFilters.length;
while (i--) {
name = attributeFilters[i].name;
if (name in node.attributes.map) {
list = matchedAttributes[name];
if (list) {
list.push(node);
} else {
matchedAttributes[name] = [node];
}
}
}
return node;
};
/**
* Adds a node filter function to the parser, the parser will collect the specified nodes by name
* and then execute the callback ones it has finished parsing the document.
*
* @example
* parser.addNodeFilter('p,h1', function(nodes, name) {
* for (var i = 0; i < nodes.length; i++) {
* console.log(nodes[i].name);
* }
* });
* @method addNodeFilter
* @method {String} name Comma separated list of nodes to collect.
* @param {function} callback Callback function to execute once it has collected nodes.
*/
self.addNodeFilter = function(name, callback) {
each(explode(name), function(name) {
var list = nodeFilters[name];
if (!list) {
nodeFilters[name] = list = [];
}
list.push(callback);
});
};
/**
* Adds a attribute filter function to the parser, the parser will collect nodes that has the specified attributes
* and then execute the callback ones it has finished parsing the document.
*
* @example
* parser.addAttributeFilter('src,href', function(nodes, name) {
* for (var i = 0; i < nodes.length; i++) {
* console.log(nodes[i].name);
* }
* });
* @method addAttributeFilter
* @method {String} name Comma separated list of nodes to collect.
* @param {function} callback Callback function to execute once it has collected nodes.
*/
self.addAttributeFilter = function(name, callback) {
each(explode(name), function(name) {
var i;
for (i = 0; i < attributeFilters.length; i++) {
if (attributeFilters[i].name === name) {
attributeFilters[i].callbacks.push(callback);
return;
}
}
attributeFilters.push({name: name, callbacks: [callback]});
});
};
/**
* Parses the specified HTML string into a DOM like node tree and returns the result.
*
* @example
* var rootNode = new DomParser({...}).parse('<b>text</b>');
* @method parse
* @param {String} html Html string to sax parse.
* @param {Object} args Optional args object that gets passed to all filter functions.
* @return {tinymce.html.Node} Root node containing the tree.
*/
self.parse = function(html, args) {
var parser, rootNode, node, nodes, i, l, fi, fl, list, name, validate;
var blockElements, startWhiteSpaceRegExp, invalidChildren = [], isInWhiteSpacePreservedElement;
var endWhiteSpaceRegExp, allWhiteSpaceRegExp, isAllWhiteSpaceRegExp, whiteSpaceElements;
var children, nonEmptyElements, rootBlockName;
args = args || {};
matchedNodes = {};
matchedAttributes = {};
blockElements = extend(makeMap('script,style,head,html,body,title,meta,param'), schema.getBlockElements());
nonEmptyElements = schema.getNonEmptyElements();
children = schema.children;
validate = settings.validate;
rootBlockName = "forced_root_block" in args ? args.forced_root_block : settings.forced_root_block;
whiteSpaceElements = schema.getWhiteSpaceElements();
startWhiteSpaceRegExp = /^[ \t\r\n]+/;
endWhiteSpaceRegExp = /[ \t\r\n]+$/;
allWhiteSpaceRegExp = /[ \t\r\n]+/g;
isAllWhiteSpaceRegExp = /^[ \t\r\n]+$/;
function addRootBlocks() {
var node = rootNode.firstChild, next, rootBlockNode;
// Removes whitespace at beginning and end of block so:
// <p> x </p> -> <p>x</p>
function trim(rootBlockNode) {
if (rootBlockNode) {
node = rootBlockNode.firstChild;
if (node && node.type == 3) {
node.value = node.value.replace(startWhiteSpaceRegExp, '');
}
node = rootBlockNode.lastChild;
if (node && node.type == 3) {
node.value = node.value.replace(endWhiteSpaceRegExp, '');
}
}
}
// Check if rootBlock is valid within rootNode for example if P is valid in H1 if H1 is the contentEditabe root
if (!schema.isValidChild(rootNode.name, rootBlockName.toLowerCase())) {
return;
}
while (node) {
next = node.next;
if (node.type == 3 || (node.type == 1 && node.name !== 'p' &&
!blockElements[node.name] && !node.attr('data-mce-type'))) {
if (!rootBlockNode) {
// Create a new root block element
rootBlockNode = createNode(rootBlockName, 1);
rootBlockNode.attr(settings.forced_root_block_attrs);
rootNode.insert(rootBlockNode, node);
rootBlockNode.append(node);
} else {
rootBlockNode.append(node);
}
} else {
trim(rootBlockNode);
rootBlockNode = null;
}
node = next;
}
trim(rootBlockNode);
}
function createNode(name, type) {
var node = new Node(name, type), list;
if (name in nodeFilters) {
list = matchedNodes[name];
if (list) {
list.push(node);
} else {
matchedNodes[name] = [node];
}
}
return node;
}
function removeWhitespaceBefore(node) {
var textNode, textNodeNext, textVal, sibling, blockElements = schema.getBlockElements();
for (textNode = node.prev; textNode && textNode.type === 3;) {
textVal = textNode.value.replace(endWhiteSpaceRegExp, '');
// Found a text node with non whitespace then trim that and break
if (textVal.length > 0) {
textNode.value = textVal;
return;
}
textNodeNext = textNode.next;
// Fix for bug #7543 where bogus nodes would produce empty
// text nodes and these would be removed if a nested list was before it
if (textNodeNext) {
if (textNodeNext.type == 3 && textNodeNext.value.length) {
textNode = textNode.prev;
continue;
}
if (!blockElements[textNodeNext.name] && textNodeNext.name != 'script' && textNodeNext.name != 'style') {
textNode = textNode.prev;
continue;
}
}
sibling = textNode.prev;
textNode.remove();
textNode = sibling;
}
}
function cloneAndExcludeBlocks(input) {
var name, output = {};
for (name in input) {
if (name !== 'li' && name != 'p') {
output[name] = input[name];
}
}
return output;
}
parser = new SaxParser({
validate: validate,
allow_script_urls: settings.allow_script_urls,
allow_conditional_comments: settings.allow_conditional_comments,
// Exclude P and LI from DOM parsing since it's treated better by the DOM parser
self_closing_elements: cloneAndExcludeBlocks(schema.getSelfClosingElements()),
cdata: function(text) {
node.append(createNode('#cdata', 4)).value = text;
},
text: function(text, raw) {
var textNode;
// Trim all redundant whitespace on non white space elements
if (!isInWhiteSpacePreservedElement) {
text = text.replace(allWhiteSpaceRegExp, ' ');
if (node.lastChild && blockElements[node.lastChild.name]) {
text = text.replace(startWhiteSpaceRegExp, '');
}
}
// Do we need to create the node
if (text.length !== 0) {
textNode = createNode('#text', 3);
textNode.raw = !!raw;
node.append(textNode).value = text;
}
},
comment: function(text) {
node.append(createNode('#comment', 8)).value = text;
},
pi: function(name, text) {
node.append(createNode(name, 7)).value = text;
removeWhitespaceBefore(node);
},
doctype: function(text) {
var newNode;
newNode = node.append(createNode('#doctype', 10));
newNode.value = text;
removeWhitespaceBefore(node);
},
start: function(name, attrs, empty) {
var newNode, attrFiltersLen, elementRule, attrName, parent;
elementRule = validate ? schema.getElementRule(name) : {};
if (elementRule) {
newNode = createNode(elementRule.outputName || name, 1);
newNode.attributes = attrs;
newNode.shortEnded = empty;
node.append(newNode);
// Check if node is valid child of the parent node is the child is
// unknown we don't collect it since it's probably a custom element
parent = children[node.name];
if (parent && children[newNode.name] && !parent[newNode.name]) {
invalidChildren.push(newNode);
}
attrFiltersLen = attributeFilters.length;
while (attrFiltersLen--) {
attrName = attributeFilters[attrFiltersLen].name;
if (attrName in attrs.map) {
list = matchedAttributes[attrName];
if (list) {
list.push(newNode);
} else {
matchedAttributes[attrName] = [newNode];
}
}
}
// Trim whitespace before block
if (blockElements[name]) {
removeWhitespaceBefore(newNode);
}
// Change current node if the element wasn't empty i.e not <br /> or <img />
if (!empty) {
node = newNode;
}
// Check if we are inside a whitespace preserved element
if (!isInWhiteSpacePreservedElement && whiteSpaceElements[name]) {
isInWhiteSpacePreservedElement = true;
}
}
},
end: function(name) {
var textNode, elementRule, text, sibling, tempNode;
elementRule = validate ? schema.getElementRule(name) : {};
if (elementRule) {
if (blockElements[name]) {
if (!isInWhiteSpacePreservedElement) {
// Trim whitespace of the first node in a block
textNode = node.firstChild;
if (textNode && textNode.type === 3) {
text = textNode.value.replace(startWhiteSpaceRegExp, '');
// Any characters left after trim or should we remove it
if (text.length > 0) {
textNode.value = text;
textNode = textNode.next;
} else {
sibling = textNode.next;
textNode.remove();
textNode = sibling;
// Remove any pure whitespace siblings
while (textNode && textNode.type === 3) {
text = textNode.value;
sibling = textNode.next;
if (text.length === 0 || isAllWhiteSpaceRegExp.test(text)) {
textNode.remove();
textNode = sibling;
}
textNode = sibling;
}
}
}
// Trim whitespace of the last node in a block
textNode = node.lastChild;
if (textNode && textNode.type === 3) {
text = textNode.value.replace(endWhiteSpaceRegExp, '');
// Any characters left after trim or should we remove it
if (text.length > 0) {
textNode.value = text;
textNode = textNode.prev;
} else {
sibling = textNode.prev;
textNode.remove();
textNode = sibling;
// Remove any pure whitespace siblings
while (textNode && textNode.type === 3) {
text = textNode.value;
sibling = textNode.prev;
if (text.length === 0 || isAllWhiteSpaceRegExp.test(text)) {
textNode.remove();
textNode = sibling;
}
textNode = sibling;
}
}
}
}
// Trim start white space
// Removed due to: #5424
/*textNode = node.prev;
if (textNode && textNode.type === 3) {
text = textNode.value.replace(startWhiteSpaceRegExp, '');
if (text.length > 0)
textNode.value = text;
else
textNode.remove();
}*/
}
// Check if we exited a whitespace preserved element
if (isInWhiteSpacePreservedElement && whiteSpaceElements[name]) {
isInWhiteSpacePreservedElement = false;
}
// Handle empty nodes
if (elementRule.removeEmpty || elementRule.paddEmpty) {
if (node.isEmpty(nonEmptyElements)) {
if (elementRule.paddEmpty) {
node.empty().append(new Node('#text', '3')).value = '\u00a0';
} else {
// Leave nodes that have a name like <a name="name">
if (!node.attributes.map.name && !node.attributes.map.id) {
tempNode = node.parent;
if (blockElements[node.name]) {
node.empty().remove();
} else {
node.unwrap();
}
node = tempNode;
return;
}
}
}
}
node = node.parent;
}
}
}, schema);
rootNode = node = new Node(args.context || settings.root_name, 11);
parser.parse(html);
// Fix invalid children or report invalid children in a contextual parsing
if (validate && invalidChildren.length) {
if (!args.context) {
fixInvalidChildren(invalidChildren);
} else {
args.invalid = true;
}
}
// Wrap nodes in the root into block elements if the root is body
if (rootBlockName && (rootNode.name == 'body' || args.isRootContent)) {
addRootBlocks();
}
// Run filters only when the contents is valid
if (!args.invalid) {
// Run node filters
for (name in matchedNodes) {
list = nodeFilters[name];
nodes = matchedNodes[name];
// Remove already removed children
fi = nodes.length;
while (fi--) {
if (!nodes[fi].parent) {
nodes.splice(fi, 1);
}
}
for (i = 0, l = list.length; i < l; i++) {
list[i](nodes, name, args);
}
}
// Run attribute filters
for (i = 0, l = attributeFilters.length; i < l; i++) {
list = attributeFilters[i];
if (list.name in matchedAttributes) {
nodes = matchedAttributes[list.name];
// Remove already removed children
fi = nodes.length;
while (fi--) {
if (!nodes[fi].parent) {
nodes.splice(fi, 1);
}
}
for (fi = 0, fl = list.callbacks.length; fi < fl; fi++) {
list.callbacks[fi](nodes, list.name, args);
}
}
}
}
return rootNode;
};
// Remove <br> at end of block elements Gecko and WebKit injects BR elements to
// make it possible to place the caret inside empty blocks. This logic tries to remove
// these elements and keep br elements that where intended to be there intact
if (settings.remove_trailing_brs) {
self.addNodeFilter('br', function(nodes) {
var i, l = nodes.length, node, blockElements = extend({}, schema.getBlockElements());
var nonEmptyElements = schema.getNonEmptyElements(), parent, lastParent, prev, prevName;
var elementRule, textNode;
// Remove brs from body element as well
blockElements.body = 1;
// Must loop forwards since it will otherwise remove all brs in <p>a<br><br><br></p>
for (i = 0; i < l; i++) {
node = nodes[i];
parent = node.parent;
if (blockElements[node.parent.name] && node === parent.lastChild) {
// Loop all nodes to the left of the current node and check for other BR elements
// excluding bookmarks since they are invisible
prev = node.prev;
while (prev) {
prevName = prev.name;
// Ignore bookmarks
if (prevName !== "span" || prev.attr('data-mce-type') !== 'bookmark') {
// Found a non BR element
if (prevName !== "br") {
break;
}
// Found another br it's a <br><br> structure then don't remove anything
if (prevName === 'br') {
node = null;
break;
}
}
prev = prev.prev;
}
if (node) {
node.remove();
// Is the parent to be considered empty after we removed the BR
if (parent.isEmpty(nonEmptyElements)) {
elementRule = schema.getElementRule(parent.name);
// Remove or padd the element depending on schema rule
if (elementRule) {
if (elementRule.removeEmpty) {
parent.remove();
} else if (elementRule.paddEmpty) {
parent.empty().append(new Node('#text', 3)).value = '\u00a0';
}
}
}
}
} else {
// Replaces BR elements inside inline elements like <p><b><i><br></i></b></p>
// so they become <p><b><i>&nbsp;</i></b></p>
lastParent = node;
while (parent && parent.firstChild === lastParent && parent.lastChild === lastParent) {
lastParent = parent;
if (blockElements[parent.name]) {
break;
}
parent = parent.parent;
}
if (lastParent === parent) {
textNode = new Node('#text', 3);
textNode.value = '\u00a0';
node.replace(textNode);
}
}
}
});
}
// Force anchor names closed, unless the setting "allow_html_in_named_anchor" is explicitly included.
if (!settings.allow_html_in_named_anchor) {
self.addAttributeFilter('id,name', function(nodes) {
var i = nodes.length, sibling, prevSibling, parent, node;
while (i--) {
node = nodes[i];
if (node.name === 'a' && node.firstChild && !node.attr('href')) {
parent = node.parent;
// Move children after current node
sibling = node.lastChild;
do {
prevSibling = sibling.prev;
parent.insert(sibling, node);
sibling = prevSibling;
} while (sibling);
}
}
});
}
if (settings.validate && schema.getValidClasses()) {
self.addAttributeFilter('class', function(nodes) {
var i = nodes.length, node, classList, ci, className, classValue;
var validClasses = schema.getValidClasses(), validClassesMap, valid;
while (i--) {
node = nodes[i];
classList = node.attr('class').split(' ');
classValue = '';
for (ci = 0; ci < classList.length; ci++) {
className = classList[ci];
valid = false;
validClassesMap = validClasses['*'];
if (validClassesMap && validClassesMap[className]) {
valid = true;
}
validClassesMap = validClasses[node.name];
if (!valid && validClassesMap && validClassesMap[className]) {
valid = true;
}
if (valid) {
if (classValue) {
classValue += ' ';
}
classValue += className;
}
}
if (!classValue.length) {
classValue = null;
}
node.attr('class', classValue);
}
});
}
};
});

View File

@@ -0,0 +1,268 @@
/**
* Entities.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/*jshint bitwise:false */
/*eslint no-bitwise:0 */
/**
* Entity encoder class.
*
* @class tinymce.html.Entities
* @static
* @version 3.4
*/
define("tinymce/html/Entities", [
"tinymce/util/Tools"
], function(Tools) {
var makeMap = Tools.makeMap;
var namedEntities, baseEntities, reverseEntities,
attrsCharsRegExp = /[&<>\"\u0060\u007E-\uD7FF\uE000-\uFFEF]|[\uD800-\uDBFF][\uDC00-\uDFFF]/g,
textCharsRegExp = /[<>&\u007E-\uD7FF\uE000-\uFFEF]|[\uD800-\uDBFF][\uDC00-\uDFFF]/g,
rawCharsRegExp = /[<>&\"\']/g,
entityRegExp = /&#([a-z0-9]+);?|&([a-z0-9]+);/gi,
asciiMap = {
128: "\u20AC", 130: "\u201A", 131: "\u0192", 132: "\u201E", 133: "\u2026", 134: "\u2020",
135: "\u2021", 136: "\u02C6", 137: "\u2030", 138: "\u0160", 139: "\u2039", 140: "\u0152",
142: "\u017D", 145: "\u2018", 146: "\u2019", 147: "\u201C", 148: "\u201D", 149: "\u2022",
150: "\u2013", 151: "\u2014", 152: "\u02DC", 153: "\u2122", 154: "\u0161", 155: "\u203A",
156: "\u0153", 158: "\u017E", 159: "\u0178"
};
// Raw entities
baseEntities = {
'\"': '&quot;', // Needs to be escaped since the YUI compressor would otherwise break the code
"'": '&#39;',
'<': '&lt;',
'>': '&gt;',
'&': '&amp;',
'\u0060': '&#96;'
};
// Reverse lookup table for raw entities
reverseEntities = {
'&lt;': '<',
'&gt;': '>',
'&amp;': '&',
'&quot;': '"',
'&apos;': "'"
};
// Decodes text by using the browser
function nativeDecode(text) {
var elm;
elm = document.createElement("div");
elm.innerHTML = text;
return elm.textContent || elm.innerText || text;
}
// Build a two way lookup table for the entities
function buildEntitiesLookup(items, radix) {
var i, chr, entity, lookup = {};
if (items) {
items = items.split(',');
radix = radix || 10;
// Build entities lookup table
for (i = 0; i < items.length; i += 2) {
chr = String.fromCharCode(parseInt(items[i], radix));
// Only add non base entities
if (!baseEntities[chr]) {
entity = '&' + items[i + 1] + ';';
lookup[chr] = entity;
lookup[entity] = chr;
}
}
return lookup;
}
}
// Unpack entities lookup where the numbers are in radix 32 to reduce the size
namedEntities = buildEntitiesLookup(
'50,nbsp,51,iexcl,52,cent,53,pound,54,curren,55,yen,56,brvbar,57,sect,58,uml,59,copy,' +
'5a,ordf,5b,laquo,5c,not,5d,shy,5e,reg,5f,macr,5g,deg,5h,plusmn,5i,sup2,5j,sup3,5k,acute,' +
'5l,micro,5m,para,5n,middot,5o,cedil,5p,sup1,5q,ordm,5r,raquo,5s,frac14,5t,frac12,5u,frac34,' +
'5v,iquest,60,Agrave,61,Aacute,62,Acirc,63,Atilde,64,Auml,65,Aring,66,AElig,67,Ccedil,' +
'68,Egrave,69,Eacute,6a,Ecirc,6b,Euml,6c,Igrave,6d,Iacute,6e,Icirc,6f,Iuml,6g,ETH,6h,Ntilde,' +
'6i,Ograve,6j,Oacute,6k,Ocirc,6l,Otilde,6m,Ouml,6n,times,6o,Oslash,6p,Ugrave,6q,Uacute,' +
'6r,Ucirc,6s,Uuml,6t,Yacute,6u,THORN,6v,szlig,70,agrave,71,aacute,72,acirc,73,atilde,74,auml,' +
'75,aring,76,aelig,77,ccedil,78,egrave,79,eacute,7a,ecirc,7b,euml,7c,igrave,7d,iacute,7e,icirc,' +
'7f,iuml,7g,eth,7h,ntilde,7i,ograve,7j,oacute,7k,ocirc,7l,otilde,7m,ouml,7n,divide,7o,oslash,' +
'7p,ugrave,7q,uacute,7r,ucirc,7s,uuml,7t,yacute,7u,thorn,7v,yuml,ci,fnof,sh,Alpha,si,Beta,' +
'sj,Gamma,sk,Delta,sl,Epsilon,sm,Zeta,sn,Eta,so,Theta,sp,Iota,sq,Kappa,sr,Lambda,ss,Mu,' +
'st,Nu,su,Xi,sv,Omicron,t0,Pi,t1,Rho,t3,Sigma,t4,Tau,t5,Upsilon,t6,Phi,t7,Chi,t8,Psi,' +
't9,Omega,th,alpha,ti,beta,tj,gamma,tk,delta,tl,epsilon,tm,zeta,tn,eta,to,theta,tp,iota,' +
'tq,kappa,tr,lambda,ts,mu,tt,nu,tu,xi,tv,omicron,u0,pi,u1,rho,u2,sigmaf,u3,sigma,u4,tau,' +
'u5,upsilon,u6,phi,u7,chi,u8,psi,u9,omega,uh,thetasym,ui,upsih,um,piv,812,bull,816,hellip,' +
'81i,prime,81j,Prime,81u,oline,824,frasl,88o,weierp,88h,image,88s,real,892,trade,89l,alefsym,' +
'8cg,larr,8ch,uarr,8ci,rarr,8cj,darr,8ck,harr,8dl,crarr,8eg,lArr,8eh,uArr,8ei,rArr,8ej,dArr,' +
'8ek,hArr,8g0,forall,8g2,part,8g3,exist,8g5,empty,8g7,nabla,8g8,isin,8g9,notin,8gb,ni,8gf,prod,' +
'8gh,sum,8gi,minus,8gn,lowast,8gq,radic,8gt,prop,8gu,infin,8h0,ang,8h7,and,8h8,or,8h9,cap,8ha,cup,' +
'8hb,int,8hk,there4,8hs,sim,8i5,cong,8i8,asymp,8j0,ne,8j1,equiv,8j4,le,8j5,ge,8k2,sub,8k3,sup,8k4,' +
'nsub,8k6,sube,8k7,supe,8kl,oplus,8kn,otimes,8l5,perp,8m5,sdot,8o8,lceil,8o9,rceil,8oa,lfloor,8ob,' +
'rfloor,8p9,lang,8pa,rang,9ea,loz,9j0,spades,9j3,clubs,9j5,hearts,9j6,diams,ai,OElig,aj,oelig,b0,' +
'Scaron,b1,scaron,bo,Yuml,m6,circ,ms,tilde,802,ensp,803,emsp,809,thinsp,80c,zwnj,80d,zwj,80e,lrm,' +
'80f,rlm,80j,ndash,80k,mdash,80o,lsquo,80p,rsquo,80q,sbquo,80s,ldquo,80t,rdquo,80u,bdquo,810,dagger,' +
'811,Dagger,81g,permil,81p,lsaquo,81q,rsaquo,85c,euro', 32);
var Entities = {
/**
* Encodes the specified string using raw entities. This means only the required XML base entities will be encoded.
*
* @method encodeRaw
* @param {String} text Text to encode.
* @param {Boolean} attr Optional flag to specify if the text is attribute contents.
* @return {String} Entity encoded text.
*/
encodeRaw: function(text, attr) {
return text.replace(attr ? attrsCharsRegExp : textCharsRegExp, function(chr) {
return baseEntities[chr] || chr;
});
},
/**
* Encoded the specified text with both the attributes and text entities. This function will produce larger text contents
* since it doesn't know if the context is within a attribute or text node. This was added for compatibility
* and is exposed as the DOMUtils.encode function.
*
* @method encodeAllRaw
* @param {String} text Text to encode.
* @return {String} Entity encoded text.
*/
encodeAllRaw: function(text) {
return ('' + text).replace(rawCharsRegExp, function(chr) {
return baseEntities[chr] || chr;
});
},
/**
* Encodes the specified string using numeric entities. The core entities will be
* encoded as named ones but all non lower ascii characters will be encoded into numeric entities.
*
* @method encodeNumeric
* @param {String} text Text to encode.
* @param {Boolean} attr Optional flag to specify if the text is attribute contents.
* @return {String} Entity encoded text.
*/
encodeNumeric: function(text, attr) {
return text.replace(attr ? attrsCharsRegExp : textCharsRegExp, function(chr) {
// Multi byte sequence convert it to a single entity
if (chr.length > 1) {
return '&#' + (((chr.charCodeAt(0) - 0xD800) * 0x400) + (chr.charCodeAt(1) - 0xDC00) + 0x10000) + ';';
}
return baseEntities[chr] || '&#' + chr.charCodeAt(0) + ';';
});
},
/**
* Encodes the specified string using named entities. The core entities will be encoded
* as named ones but all non lower ascii characters will be encoded into named entities.
*
* @method encodeNamed
* @param {String} text Text to encode.
* @param {Boolean} attr Optional flag to specify if the text is attribute contents.
* @param {Object} entities Optional parameter with entities to use.
* @return {String} Entity encoded text.
*/
encodeNamed: function(text, attr, entities) {
entities = entities || namedEntities;
return text.replace(attr ? attrsCharsRegExp : textCharsRegExp, function(chr) {
return baseEntities[chr] || entities[chr] || chr;
});
},
/**
* Returns an encode function based on the name(s) and it's optional entities.
*
* @method getEncodeFunc
* @param {String} name Comma separated list of encoders for example named,numeric.
* @param {String} entities Optional parameter with entities to use instead of the built in set.
* @return {function} Encode function to be used.
*/
getEncodeFunc: function(name, entities) {
entities = buildEntitiesLookup(entities) || namedEntities;
function encodeNamedAndNumeric(text, attr) {
return text.replace(attr ? attrsCharsRegExp : textCharsRegExp, function(chr) {
return baseEntities[chr] || entities[chr] || '&#' + chr.charCodeAt(0) + ';' || chr;
});
}
function encodeCustomNamed(text, attr) {
return Entities.encodeNamed(text, attr, entities);
}
// Replace + with , to be compatible with previous TinyMCE versions
name = makeMap(name.replace(/\+/g, ','));
// Named and numeric encoder
if (name.named && name.numeric) {
return encodeNamedAndNumeric;
}
// Named encoder
if (name.named) {
// Custom names
if (entities) {
return encodeCustomNamed;
}
return Entities.encodeNamed;
}
// Numeric
if (name.numeric) {
return Entities.encodeNumeric;
}
// Raw encoder
return Entities.encodeRaw;
},
/**
* Decodes the specified string, this will replace entities with raw UTF characters.
*
* @method decode
* @param {String} text Text to entity decode.
* @return {String} Entity decoded string.
*/
decode: function(text) {
return text.replace(entityRegExp, function(all, numeric) {
if (numeric) {
if (numeric.charAt(0).toLowerCase() === 'x') {
numeric = parseInt(numeric.substr(1), 16);
} else {
numeric = parseInt(numeric, 10);
}
// Support upper UTF
if (numeric > 0xFFFF) {
numeric -= 0x10000;
return String.fromCharCode(0xD800 + (numeric >> 10), 0xDC00 + (numeric & 0x3FF));
}
return asciiMap[numeric] || String.fromCharCode(numeric);
}
return reverseEntities[all] || namedEntities[all] || nativeDecode(all);
});
}
};
return Entities;
});

View File

@@ -0,0 +1,496 @@
/**
* Node.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/**
* This class is a minimalistic implementation of a DOM like node used by the DomParser class.
*
* @example
* var node = new tinymce.html.Node('strong', 1);
* someRoot.append(node);
*
* @class tinymce.html.Node
* @version 3.4
*/
define("tinymce/html/Node", [], function() {
var whiteSpaceRegExp = /^[ \t\r\n]*$/, typeLookup = {
'#text': 3,
'#comment': 8,
'#cdata': 4,
'#pi': 7,
'#doctype': 10,
'#document-fragment': 11
};
// Walks the tree left/right
function walk(node, root_node, prev) {
var sibling, parent, startName = prev ? 'lastChild' : 'firstChild', siblingName = prev ? 'prev' : 'next';
// Walk into nodes if it has a start
if (node[startName]) {
return node[startName];
}
// Return the sibling if it has one
if (node !== root_node) {
sibling = node[siblingName];
if (sibling) {
return sibling;
}
// Walk up the parents to look for siblings
for (parent = node.parent; parent && parent !== root_node; parent = parent.parent) {
sibling = parent[siblingName];
if (sibling) {
return sibling;
}
}
}
}
/**
* Constructs a new Node instance.
*
* @constructor
* @method Node
* @param {String} name Name of the node type.
* @param {Number} type Numeric type representing the node.
*/
function Node(name, type) {
this.name = name;
this.type = type;
if (type === 1) {
this.attributes = [];
this.attributes.map = {};
}
}
Node.prototype = {
/**
* Replaces the current node with the specified one.
*
* @example
* someNode.replace(someNewNode);
*
* @method replace
* @param {tinymce.html.Node} node Node to replace the current node with.
* @return {tinymce.html.Node} The old node that got replaced.
*/
replace: function(node) {
var self = this;
if (node.parent) {
node.remove();
}
self.insert(node, self);
self.remove();
return self;
},
/**
* Gets/sets or removes an attribute by name.
*
* @example
* someNode.attr("name", "value"); // Sets an attribute
* console.log(someNode.attr("name")); // Gets an attribute
* someNode.attr("name", null); // Removes an attribute
*
* @method attr
* @param {String} name Attribute name to set or get.
* @param {String} value Optional value to set.
* @return {String/tinymce.html.Node} String or undefined on a get operation or the current node on a set operation.
*/
attr: function(name, value) {
var self = this, attrs, i, undef;
if (typeof name !== "string") {
for (i in name) {
self.attr(i, name[i]);
}
return self;
}
if ((attrs = self.attributes)) {
if (value !== undef) {
// Remove attribute
if (value === null) {
if (name in attrs.map) {
delete attrs.map[name];
i = attrs.length;
while (i--) {
if (attrs[i].name === name) {
attrs = attrs.splice(i, 1);
return self;
}
}
}
return self;
}
// Set attribute
if (name in attrs.map) {
// Set attribute
i = attrs.length;
while (i--) {
if (attrs[i].name === name) {
attrs[i].value = value;
break;
}
}
} else {
attrs.push({name: name, value: value});
}
attrs.map[name] = value;
return self;
}
return attrs.map[name];
}
},
/**
* Does a shallow clones the node into a new node. It will also exclude id attributes since
* there should only be one id per document.
*
* @example
* var clonedNode = node.clone();
*
* @method clone
* @return {tinymce.html.Node} New copy of the original node.
*/
clone: function() {
var self = this, clone = new Node(self.name, self.type), i, l, selfAttrs, selfAttr, cloneAttrs;
// Clone element attributes
if ((selfAttrs = self.attributes)) {
cloneAttrs = [];
cloneAttrs.map = {};
for (i = 0, l = selfAttrs.length; i < l; i++) {
selfAttr = selfAttrs[i];
// Clone everything except id
if (selfAttr.name !== 'id') {
cloneAttrs[cloneAttrs.length] = {name: selfAttr.name, value: selfAttr.value};
cloneAttrs.map[selfAttr.name] = selfAttr.value;
}
}
clone.attributes = cloneAttrs;
}
clone.value = self.value;
clone.shortEnded = self.shortEnded;
return clone;
},
/**
* Wraps the node in in another node.
*
* @example
* node.wrap(wrapperNode);
*
* @method wrap
*/
wrap: function(wrapper) {
var self = this;
self.parent.insert(wrapper, self);
wrapper.append(self);
return self;
},
/**
* Unwraps the node in other words it removes the node but keeps the children.
*
* @example
* node.unwrap();
*
* @method unwrap
*/
unwrap: function() {
var self = this, node, next;
for (node = self.firstChild; node;) {
next = node.next;
self.insert(node, self, true);
node = next;
}
self.remove();
},
/**
* Removes the node from it's parent.
*
* @example
* node.remove();
*
* @method remove
* @return {tinymce.html.Node} Current node that got removed.
*/
remove: function() {
var self = this, parent = self.parent, next = self.next, prev = self.prev;
if (parent) {
if (parent.firstChild === self) {
parent.firstChild = next;
if (next) {
next.prev = null;
}
} else {
prev.next = next;
}
if (parent.lastChild === self) {
parent.lastChild = prev;
if (prev) {
prev.next = null;
}
} else {
next.prev = prev;
}
self.parent = self.next = self.prev = null;
}
return self;
},
/**
* Appends a new node as a child of the current node.
*
* @example
* node.append(someNode);
*
* @method append
* @param {tinymce.html.Node} node Node to append as a child of the current one.
* @return {tinymce.html.Node} The node that got appended.
*/
append: function(node) {
var self = this, last;
if (node.parent) {
node.remove();
}
last = self.lastChild;
if (last) {
last.next = node;
node.prev = last;
self.lastChild = node;
} else {
self.lastChild = self.firstChild = node;
}
node.parent = self;
return node;
},
/**
* Inserts a node at a specific position as a child of the current node.
*
* @example
* parentNode.insert(newChildNode, oldChildNode);
*
* @method insert
* @param {tinymce.html.Node} node Node to insert as a child of the current node.
* @param {tinymce.html.Node} ref_node Reference node to set node before/after.
* @param {Boolean} before Optional state to insert the node before the reference node.
* @return {tinymce.html.Node} The node that got inserted.
*/
insert: function(node, ref_node, before) {
var parent;
if (node.parent) {
node.remove();
}
parent = ref_node.parent || this;
if (before) {
if (ref_node === parent.firstChild) {
parent.firstChild = node;
} else {
ref_node.prev.next = node;
}
node.prev = ref_node.prev;
node.next = ref_node;
ref_node.prev = node;
} else {
if (ref_node === parent.lastChild) {
parent.lastChild = node;
} else {
ref_node.next.prev = node;
}
node.next = ref_node.next;
node.prev = ref_node;
ref_node.next = node;
}
node.parent = parent;
return node;
},
/**
* Get all children by name.
*
* @method getAll
* @param {String} name Name of the child nodes to collect.
* @return {Array} Array with child nodes matchin the specified name.
*/
getAll: function(name) {
var self = this, node, collection = [];
for (node = self.firstChild; node; node = walk(node, self)) {
if (node.name === name) {
collection.push(node);
}
}
return collection;
},
/**
* Removes all children of the current node.
*
* @method empty
* @return {tinymce.html.Node} The current node that got cleared.
*/
empty: function() {
var self = this, nodes, i, node;
// Remove all children
if (self.firstChild) {
nodes = [];
// Collect the children
for (node = self.firstChild; node; node = walk(node, self)) {
nodes.push(node);
}
// Remove the children
i = nodes.length;
while (i--) {
node = nodes[i];
node.parent = node.firstChild = node.lastChild = node.next = node.prev = null;
}
}
self.firstChild = self.lastChild = null;
return self;
},
/**
* Returns true/false if the node is to be considered empty or not.
*
* @example
* node.isEmpty({img: true});
* @method isEmpty
* @param {Object} elements Name/value object with elements that are automatically treated as non empty elements.
* @return {Boolean} true/false if the node is empty or not.
*/
isEmpty: function(elements) {
var self = this, node = self.firstChild, i, name;
if (node) {
do {
if (node.type === 1) {
// Ignore bogus elements
if (node.attributes.map['data-mce-bogus']) {
continue;
}
// Keep empty elements like <img />
if (elements[node.name]) {
return false;
}
// Keep bookmark nodes and name attribute like <a name="1"></a>
i = node.attributes.length;
while (i--) {
name = node.attributes[i].name;
if (name === "name" || name.indexOf('data-mce-bookmark') === 0) {
return false;
}
}
}
// Keep comments
if (node.type === 8) {
return false;
}
// Keep non whitespace text nodes
if ((node.type === 3 && !whiteSpaceRegExp.test(node.value))) {
return false;
}
} while ((node = walk(node, self)));
}
return true;
},
/**
* Walks to the next or previous node and returns that node or null if it wasn't found.
*
* @method walk
* @param {Boolean} prev Optional previous node state defaults to false.
* @return {tinymce.html.Node} Node that is next to or previous of the current node.
*/
walk: function(prev) {
return walk(this, null, prev);
}
};
/**
* Creates a node of a specific type.
*
* @static
* @method create
* @param {String} name Name of the node type to create for example "b" or "#text".
* @param {Object} attrs Name/value collection of attributes that will be applied to elements.
*/
Node.create = function(name, attrs) {
var node, attrName;
// Create node
node = new Node(name, typeLookup[name] || 1);
// Add attributes if needed
if (attrs) {
for (attrName in attrs) {
node.attr(attrName, attrs[attrName]);
}
}
return node;
};
return Node;
});

View File

@@ -0,0 +1,474 @@
/**
* SaxParser.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/*eslint max-depth:[2, 9] */
/**
* This class parses HTML code using pure JavaScript and executes various events for each item it finds. It will
* always execute the events in the right order for tag soup code like <b><p></b></p>. It will also remove elements
* and attributes that doesn't fit the schema if the validate setting is enabled.
*
* @example
* var parser = new tinymce.html.SaxParser({
* validate: true,
*
* comment: function(text) {
* console.log('Comment:', text);
* },
*
* cdata: function(text) {
* console.log('CDATA:', text);
* },
*
* text: function(text, raw) {
* console.log('Text:', text, 'Raw:', raw);
* },
*
* start: function(name, attrs, empty) {
* console.log('Start:', name, attrs, empty);
* },
*
* end: function(name) {
* console.log('End:', name);
* },
*
* pi: function(name, text) {
* console.log('PI:', name, text);
* },
*
* doctype: function(text) {
* console.log('DocType:', text);
* }
* }, schema);
* @class tinymce.html.SaxParser
* @version 3.4
*/
define("tinymce/html/SaxParser", [
"tinymce/html/Schema",
"tinymce/html/Entities",
"tinymce/util/Tools"
], function(Schema, Entities, Tools) {
var each = Tools.each;
/**
* Returns the index of the end tag for a specific start tag. This can be
* used to skip all children of a parent element from being processed.
*
* @private
* @method findEndTag
* @param {tinymce.html.Schema} schema Schema instance to use to match short ended elements.
* @param {String} html HTML string to find the end tag in.
* @param {Number} startIndex Indext to start searching at should be after the start tag.
* @return {Number} Index of the end tag.
*/
function findEndTag(schema, html, startIndex) {
var count = 1, index, matches, tokenRegExp, shortEndedElements;
shortEndedElements = schema.getShortEndedElements();
tokenRegExp = /<([!?\/])?([A-Za-z0-9\-_\:\.]+)((?:\s+[^"\'>]+(?:(?:"[^"]*")|(?:\'[^\']*\')|[^>]*))*|\/|\s+)>/g;
tokenRegExp.lastIndex = index = startIndex;
while ((matches = tokenRegExp.exec(html))) {
index = tokenRegExp.lastIndex;
if (matches[1] === '/') { // End element
count--;
} else if (!matches[1]) { // Start element
if (matches[2] in shortEndedElements) {
continue;
}
count++;
}
if (count === 0) {
break;
}
}
return index;
}
/**
* Constructs a new SaxParser instance.
*
* @constructor
* @method SaxParser
* @param {Object} settings Name/value collection of settings. comment, cdata, text, start and end are callbacks.
* @param {tinymce.html.Schema} schema HTML Schema class to use when parsing.
*/
function SaxParser(settings, schema) {
var self = this;
function noop() {}
settings = settings || {};
self.schema = schema = schema || new Schema();
if (settings.fix_self_closing !== false) {
settings.fix_self_closing = true;
}
// Add handler functions from settings and setup default handlers
each('comment cdata text start end pi doctype'.split(' '), function(name) {
if (name) {
self[name] = settings[name] || noop;
}
});
/**
* Parses the specified HTML string and executes the callbacks for each item it finds.
*
* @example
* new SaxParser({...}).parse('<b>text</b>');
* @method parse
* @param {String} html Html string to sax parse.
*/
self.parse = function(html) {
var self = this, matches, index = 0, value, endRegExp, stack = [], attrList, i, text, name;
var isInternalElement, removeInternalElements, shortEndedElements, fillAttrsMap, isShortEnded;
var validate, elementRule, isValidElement, attr, attribsValue, validAttributesMap, validAttributePatterns;
var attributesRequired, attributesDefault, attributesForced;
var anyAttributesRequired, selfClosing, tokenRegExp, attrRegExp, specialElements, attrValue, idCount = 0;
var decode = Entities.decode, fixSelfClosing, filteredUrlAttrs = Tools.makeMap('src,href,data,background,formaction,poster');
var scriptUriRegExp = /((java|vb)script|mhtml):/i, dataUriRegExp = /^data:/i;
function processEndTag(name) {
var pos, i;
// Find position of parent of the same type
pos = stack.length;
while (pos--) {
if (stack[pos].name === name) {
break;
}
}
// Found parent
if (pos >= 0) {
// Close all the open elements
for (i = stack.length - 1; i >= pos; i--) {
name = stack[i];
if (name.valid) {
self.end(name.name);
}
}
// Remove the open elements from the stack
stack.length = pos;
}
}
function parseAttribute(match, name, value, val2, val3) {
var attrRule, i, trimRegExp = /[\s\u0000-\u001F]+/g;
name = name.toLowerCase();
value = name in fillAttrsMap ? name : decode(value || val2 || val3 || ''); // Handle boolean attribute than value attribute
// Validate name and value pass through all data- attributes
if (validate && !isInternalElement && name.indexOf('data-') !== 0) {
attrRule = validAttributesMap[name];
// Find rule by pattern matching
if (!attrRule && validAttributePatterns) {
i = validAttributePatterns.length;
while (i--) {
attrRule = validAttributePatterns[i];
if (attrRule.pattern.test(name)) {
break;
}
}
// No rule matched
if (i === -1) {
attrRule = null;
}
}
// No attribute rule found
if (!attrRule) {
return;
}
// Validate value
if (attrRule.validValues && !(value in attrRule.validValues)) {
return;
}
}
// Block any javascript: urls or non image data uris
if (filteredUrlAttrs[name] && !settings.allow_script_urls) {
var uri = value.replace(trimRegExp, '');
try {
// Might throw malformed URI sequence
uri = decodeURIComponent(uri);
} catch (ex) {
// Fallback to non UTF-8 decoder
uri = unescape(uri);
}
if (scriptUriRegExp.test(uri)) {
return;
}
if (!settings.allow_html_data_urls && dataUriRegExp.test(uri) && !/^data:image\//i.test(uri)) {
return;
}
}
// Add attribute to list and map
attrList.map[name] = value;
attrList.push({
name: name,
value: value
});
}
// Precompile RegExps and map objects
tokenRegExp = new RegExp('<(?:' +
'(?:!--([\\w\\W]*?)-->)|' + // Comment
'(?:!\\[CDATA\\[([\\w\\W]*?)\\]\\]>)|' + // CDATA
'(?:!DOCTYPE([\\w\\W]*?)>)|' + // DOCTYPE
'(?:\\?([^\\s\\/<>]+) ?([\\w\\W]*?)[?/]>)|' + // PI
'(?:\\/([^>]+)>)|' + // End element
'(?:([A-Za-z0-9\\-_\\:\\.]+)((?:\\s+[^"\'>]+(?:(?:"[^"]*")|(?:\'[^\']*\')|[^>]*))*|\\/|\\s+)>)' + // Start element
')', 'g');
attrRegExp = /([\w:\-]+)(?:\s*=\s*(?:(?:\"((?:[^\"])*)\")|(?:\'((?:[^\'])*)\')|([^>\s]+)))?/g;
// Setup lookup tables for empty elements and boolean attributes
shortEndedElements = schema.getShortEndedElements();
selfClosing = settings.self_closing_elements || schema.getSelfClosingElements();
fillAttrsMap = schema.getBoolAttrs();
validate = settings.validate;
removeInternalElements = settings.remove_internals;
fixSelfClosing = settings.fix_self_closing;
specialElements = schema.getSpecialElements();
while ((matches = tokenRegExp.exec(html))) {
// Text
if (index < matches.index) {
self.text(decode(html.substr(index, matches.index - index)));
}
if ((value = matches[6])) { // End element
value = value.toLowerCase();
// IE will add a ":" in front of elements it doesn't understand like custom elements or HTML5 elements
if (value.charAt(0) === ':') {
value = value.substr(1);
}
processEndTag(value);
} else if ((value = matches[7])) { // Start element
value = value.toLowerCase();
// IE will add a ":" in front of elements it doesn't understand like custom elements or HTML5 elements
if (value.charAt(0) === ':') {
value = value.substr(1);
}
isShortEnded = value in shortEndedElements;
// Is self closing tag for example an <li> after an open <li>
if (fixSelfClosing && selfClosing[value] && stack.length > 0 && stack[stack.length - 1].name === value) {
processEndTag(value);
}
// Validate element
if (!validate || (elementRule = schema.getElementRule(value))) {
isValidElement = true;
// Grab attributes map and patters when validation is enabled
if (validate) {
validAttributesMap = elementRule.attributes;
validAttributePatterns = elementRule.attributePatterns;
}
// Parse attributes
if ((attribsValue = matches[8])) {
isInternalElement = attribsValue.indexOf('data-mce-type') !== -1; // Check if the element is an internal element
// If the element has internal attributes then remove it if we are told to do so
if (isInternalElement && removeInternalElements) {
isValidElement = false;
}
attrList = [];
attrList.map = {};
attribsValue.replace(attrRegExp, parseAttribute);
} else {
attrList = [];
attrList.map = {};
}
// Process attributes if validation is enabled
if (validate && !isInternalElement) {
attributesRequired = elementRule.attributesRequired;
attributesDefault = elementRule.attributesDefault;
attributesForced = elementRule.attributesForced;
anyAttributesRequired = elementRule.removeEmptyAttrs;
// Check if any attribute exists
if (anyAttributesRequired && !attrList.length) {
isValidElement = false;
}
// Handle forced attributes
if (attributesForced) {
i = attributesForced.length;
while (i--) {
attr = attributesForced[i];
name = attr.name;
attrValue = attr.value;
if (attrValue === '{$uid}') {
attrValue = 'mce_' + idCount++;
}
attrList.map[name] = attrValue;
attrList.push({name: name, value: attrValue});
}
}
// Handle default attributes
if (attributesDefault) {
i = attributesDefault.length;
while (i--) {
attr = attributesDefault[i];
name = attr.name;
if (!(name in attrList.map)) {
attrValue = attr.value;
if (attrValue === '{$uid}') {
attrValue = 'mce_' + idCount++;
}
attrList.map[name] = attrValue;
attrList.push({name: name, value: attrValue});
}
}
}
// Handle required attributes
if (attributesRequired) {
i = attributesRequired.length;
while (i--) {
if (attributesRequired[i] in attrList.map) {
break;
}
}
// None of the required attributes where found
if (i === -1) {
isValidElement = false;
}
}
// Invalidate element if it's marked as bogus
if ((attr = attrList.map['data-mce-bogus'])) {
if (attr === 'all') {
index = findEndTag(schema, html, tokenRegExp.lastIndex);
tokenRegExp.lastIndex = index;
continue;
}
isValidElement = false;
}
}
if (isValidElement) {
self.start(value, attrList, isShortEnded);
}
} else {
isValidElement = false;
}
// Treat script, noscript and style a bit different since they may include code that looks like elements
if ((endRegExp = specialElements[value])) {
endRegExp.lastIndex = index = matches.index + matches[0].length;
if ((matches = endRegExp.exec(html))) {
if (isValidElement) {
text = html.substr(index, matches.index - index);
}
index = matches.index + matches[0].length;
} else {
text = html.substr(index);
index = html.length;
}
if (isValidElement) {
if (text.length > 0) {
self.text(text, true);
}
self.end(value);
}
tokenRegExp.lastIndex = index;
continue;
}
// Push value on to stack
if (!isShortEnded) {
if (!attribsValue || attribsValue.indexOf('/') != attribsValue.length - 1) {
stack.push({name: value, valid: isValidElement});
} else if (isValidElement) {
self.end(value);
}
}
} else if ((value = matches[1])) { // Comment
// Padd comment value to avoid browsers from parsing invalid comments as HTML
if (value.charAt(0) === '>') {
value = ' ' + value;
}
if (!settings.allow_conditional_comments && value.substr(0, 3) === '[if') {
value = ' ' + value;
}
self.comment(value);
} else if ((value = matches[2])) { // CDATA
self.cdata(value);
} else if ((value = matches[3])) { // DOCTYPE
self.doctype(value);
} else if ((value = matches[4])) { // PI
self.pi(value, matches[5]);
}
index = matches.index + matches[0].length;
}
// Text
if (index < html.length) {
self.text(decode(html.substr(index)));
}
// Close any open elements
for (i = stack.length - 1; i >= 0; i--) {
value = stack[i];
if (value.valid) {
self.end(value.name);
}
}
};
}
SaxParser.findEndTag = findEndTag;
return SaxParser;
});

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,158 @@
/**
* Serializer.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/**
* This class is used to serialize down the DOM tree into a string using a Writer instance.
*
*
* @example
* new tinymce.html.Serializer().serialize(new tinymce.html.DomParser().parse('<p>text</p>'));
* @class tinymce.html.Serializer
* @version 3.4
*/
define("tinymce/html/Serializer", [
"tinymce/html/Writer",
"tinymce/html/Schema"
], function(Writer, Schema) {
/**
* Constructs a new Serializer instance.
*
* @constructor
* @method Serializer
* @param {Object} settings Name/value settings object.
* @param {tinymce.html.Schema} schema Schema instance to use.
*/
return function(settings, schema) {
var self = this, writer = new Writer(settings);
settings = settings || {};
settings.validate = "validate" in settings ? settings.validate : true;
self.schema = schema = schema || new Schema();
self.writer = writer;
/**
* Serializes the specified node into a string.
*
* @example
* new tinymce.html.Serializer().serialize(new tinymce.html.DomParser().parse('<p>text</p>'));
* @method serialize
* @param {tinymce.html.Node} node Node instance to serialize.
* @return {String} String with HTML based on DOM tree.
*/
self.serialize = function(node) {
var handlers, validate;
validate = settings.validate;
handlers = {
// #text
3: function(node) {
writer.text(node.value, node.raw);
},
// #comment
8: function(node) {
writer.comment(node.value);
},
// Processing instruction
7: function(node) {
writer.pi(node.name, node.value);
},
// Doctype
10: function(node) {
writer.doctype(node.value);
},
// CDATA
4: function(node) {
writer.cdata(node.value);
},
// Document fragment
11: function(node) {
if ((node = node.firstChild)) {
do {
walk(node);
} while ((node = node.next));
}
}
};
writer.reset();
function walk(node) {
var handler = handlers[node.type], name, isEmpty, attrs, attrName, attrValue, sortedAttrs, i, l, elementRule;
if (!handler) {
name = node.name;
isEmpty = node.shortEnded;
attrs = node.attributes;
// Sort attributes
if (validate && attrs && attrs.length > 1) {
sortedAttrs = [];
sortedAttrs.map = {};
elementRule = schema.getElementRule(node.name);
if (elementRule) {
for (i = 0, l = elementRule.attributesOrder.length; i < l; i++) {
attrName = elementRule.attributesOrder[i];
if (attrName in attrs.map) {
attrValue = attrs.map[attrName];
sortedAttrs.map[attrName] = attrValue;
sortedAttrs.push({name: attrName, value: attrValue});
}
}
for (i = 0, l = attrs.length; i < l; i++) {
attrName = attrs[i].name;
if (!(attrName in sortedAttrs.map)) {
attrValue = attrs.map[attrName];
sortedAttrs.map[attrName] = attrValue;
sortedAttrs.push({name: attrName, value: attrValue});
}
}
attrs = sortedAttrs;
}
}
writer.start(node.name, attrs, isEmpty);
if (!isEmpty) {
if ((node = node.firstChild)) {
do {
walk(node);
} while ((node = node.next));
}
writer.end(name);
}
} else {
handler(node);
}
}
// Serialize element and treat all non elements as fragments
if (node.type == 1 && !settings.inner) {
walk(node);
} else {
handlers[11](node);
}
return writer.getContent();
};
};
});

View File

@@ -0,0 +1,363 @@
/**
* Styles.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/**
* This class is used to parse CSS styles it also compresses styles to reduce the output size.
*
* @example
* var Styles = new tinymce.html.Styles({
* url_converter: function(url) {
* return url;
* }
* });
*
* styles = Styles.parse('border: 1px solid red');
* styles.color = 'red';
*
* console.log(new tinymce.html.StyleSerializer().serialize(styles));
*
* @class tinymce.html.Styles
* @version 3.4
*/
define("tinymce/html/Styles", [], function() {
return function(settings, schema) {
/*jshint maxlen:255 */
/*eslint max-len:0 */
var rgbRegExp = /rgb\s*\(\s*([0-9]+)\s*,\s*([0-9]+)\s*,\s*([0-9]+)\s*\)/gi,
urlOrStrRegExp = /(?:url(?:(?:\(\s*\"([^\"]+)\"\s*\))|(?:\(\s*\'([^\']+)\'\s*\))|(?:\(\s*([^)\s]+)\s*\))))|(?:\'([^\']+)\')|(?:\"([^\"]+)\")/gi,
styleRegExp = /\s*([^:]+):\s*([^;]+);?/g,
trimRightRegExp = /\s+$/,
undef, i, encodingLookup = {}, encodingItems, validStyles, invalidStyles, invisibleChar = '\uFEFF';
settings = settings || {};
if (schema) {
validStyles = schema.getValidStyles();
invalidStyles = schema.getInvalidStyles();
}
encodingItems = ('\\" \\\' \\; \\: ; : ' + invisibleChar).split(' ');
for (i = 0; i < encodingItems.length; i++) {
encodingLookup[encodingItems[i]] = invisibleChar + i;
encodingLookup[invisibleChar + i] = encodingItems[i];
}
function toHex(match, r, g, b) {
function hex(val) {
val = parseInt(val, 10).toString(16);
return val.length > 1 ? val : '0' + val; // 0 -> 00
}
return '#' + hex(r) + hex(g) + hex(b);
}
return {
/**
* Parses the specified RGB color value and returns a hex version of that color.
*
* @method toHex
* @param {String} color RGB string value like rgb(1,2,3)
* @return {String} Hex version of that RGB value like #FF00FF.
*/
toHex: function(color) {
return color.replace(rgbRegExp, toHex);
},
/**
* Parses the specified style value into an object collection. This parser will also
* merge and remove any redundant items that browsers might have added. It will also convert non hex
* colors to hex values. Urls inside the styles will also be converted to absolute/relative based on settings.
*
* @method parse
* @param {String} css Style value to parse for example: border:1px solid red;.
* @return {Object} Object representation of that style like {border: '1px solid red'}
*/
parse: function(css) {
var styles = {}, matches, name, value, isEncoded, urlConverter = settings.url_converter;
var urlConverterScope = settings.url_converter_scope || this;
function compress(prefix, suffix, noJoin) {
var top, right, bottom, left;
top = styles[prefix + '-top' + suffix];
if (!top) {
return;
}
right = styles[prefix + '-right' + suffix];
if (!right) {
return;
}
bottom = styles[prefix + '-bottom' + suffix];
if (!bottom) {
return;
}
left = styles[prefix + '-left' + suffix];
if (!left) {
return;
}
var box = [top, right, bottom, left];
i = box.length - 1;
while (i--) {
if (box[i] !== box[i + 1]) {
break;
}
}
if (i > -1 && noJoin) {
return;
}
styles[prefix + suffix] = i == -1 ? box[0] : box.join(' ');
delete styles[prefix + '-top' + suffix];
delete styles[prefix + '-right' + suffix];
delete styles[prefix + '-bottom' + suffix];
delete styles[prefix + '-left' + suffix];
}
/**
* Checks if the specific style can be compressed in other words if all border-width are equal.
*/
function canCompress(key) {
var value = styles[key], i;
if (!value) {
return;
}
value = value.split(' ');
i = value.length;
while (i--) {
if (value[i] !== value[0]) {
return false;
}
}
styles[key] = value[0];
return true;
}
/**
* Compresses multiple styles into one style.
*/
function compress2(target, a, b, c) {
if (!canCompress(a)) {
return;
}
if (!canCompress(b)) {
return;
}
if (!canCompress(c)) {
return;
}
// Compress
styles[target] = styles[a] + ' ' + styles[b] + ' ' + styles[c];
delete styles[a];
delete styles[b];
delete styles[c];
}
// Encodes the specified string by replacing all \" \' ; : with _<num>
function encode(str) {
isEncoded = true;
return encodingLookup[str];
}
// Decodes the specified string by replacing all _<num> with it's original value \" \' etc
// It will also decode the \" \' if keep_slashes is set to fale or omitted
function decode(str, keep_slashes) {
if (isEncoded) {
str = str.replace(/\uFEFF[0-9]/g, function(str) {
return encodingLookup[str];
});
}
if (!keep_slashes) {
str = str.replace(/\\([\'\";:])/g, "$1");
}
return str;
}
function processUrl(match, url, url2, url3, str, str2) {
str = str || str2;
if (str) {
str = decode(str);
// Force strings into single quote format
return "'" + str.replace(/\'/g, "\\'") + "'";
}
url = decode(url || url2 || url3);
if (!settings.allow_script_urls) {
var scriptUrl = url.replace(/[\s\r\n]+/, '');
if (/(java|vb)script:/i.test(scriptUrl)) {
return "";
}
if (!settings.allow_svg_data_urls && /^data:image\/svg/i.test(scriptUrl)) {
return "";
}
}
// Convert the URL to relative/absolute depending on config
if (urlConverter) {
url = urlConverter.call(urlConverterScope, url, 'style');
}
// Output new URL format
return "url('" + url.replace(/\'/g, "\\'") + "')";
}
if (css) {
css = css.replace(/[\u0000-\u001F]/g, '');
// Encode \" \' % and ; and : inside strings so they don't interfere with the style parsing
css = css.replace(/\\[\"\';:\uFEFF]/g, encode).replace(/\"[^\"]+\"|\'[^\']+\'/g, function(str) {
return str.replace(/[;:]/g, encode);
});
// Parse styles
while ((matches = styleRegExp.exec(css))) {
name = matches[1].replace(trimRightRegExp, '').toLowerCase();
value = matches[2].replace(trimRightRegExp, '');
// Decode escaped sequences like \65 -> e
/*jshint loopfunc:true*/
/*eslint no-loop-func:0 */
value = value.replace(/\\[0-9a-f]+/g, function(e) {
return String.fromCharCode(parseInt(e.substr(1), 16));
});
if (name && value.length > 0) {
// Don't allow behavior name or expression/comments within the values
if (!settings.allow_script_urls && (name == "behavior" || /expression\s*\(|\/\*|\*\//.test(value))) {
continue;
}
// Opera will produce 700 instead of bold in their style values
if (name === 'font-weight' && value === '700') {
value = 'bold';
} else if (name === 'color' || name === 'background-color') { // Lowercase colors like RED
value = value.toLowerCase();
}
// Convert RGB colors to HEX
value = value.replace(rgbRegExp, toHex);
// Convert URLs and force them into url('value') format
value = value.replace(urlOrStrRegExp, processUrl);
styles[name] = isEncoded ? decode(value, true) : value;
}
styleRegExp.lastIndex = matches.index + matches[0].length;
}
// Compress the styles to reduce it's size for example IE will expand styles
compress("border", "", true);
compress("border", "-width");
compress("border", "-color");
compress("border", "-style");
compress("padding", "");
compress("margin", "");
compress2('border', 'border-width', 'border-style', 'border-color');
// Remove pointless border, IE produces these
if (styles.border === 'medium none') {
delete styles.border;
}
// IE 11 will produce a border-image: none when getting the style attribute from <p style="border: 1px solid red"></p>
// So let us assume it shouldn't be there
if (styles['border-image'] === 'none') {
delete styles['border-image'];
}
}
return styles;
},
/**
* Serializes the specified style object into a string.
*
* @method serialize
* @param {Object} styles Object to serialize as string for example: {border: '1px solid red'}
* @param {String} elementName Optional element name, if specified only the styles that matches the schema will be serialized.
* @return {String} String representation of the style object for example: border: 1px solid red.
*/
serialize: function(styles, elementName) {
var css = '', name, value;
function serializeStyles(name) {
var styleList, i, l, value;
styleList = validStyles[name];
if (styleList) {
for (i = 0, l = styleList.length; i < l; i++) {
name = styleList[i];
value = styles[name];
if (value !== undef && value.length > 0) {
css += (css.length > 0 ? ' ' : '') + name + ': ' + value + ';';
}
}
}
}
function isValid(name, elementName) {
var styleMap;
styleMap = invalidStyles['*'];
if (styleMap && styleMap[name]) {
return false;
}
styleMap = invalidStyles[elementName];
if (styleMap && styleMap[name]) {
return false;
}
return true;
}
// Serialize styles according to schema
if (elementName && validStyles) {
// Serialize global styles and element specific styles
serializeStyles('*');
serializeStyles(elementName);
} else {
// Output the styles in the order they are inside the object
for (name in styles) {
value = styles[name];
if (value !== undef && value.length > 0) {
if (!invalidStyles || isValid(name, elementName)) {
css += (css.length > 0 ? ' ' : '') + name + ': ' + value + ';';
}
}
}
}
return css;
}
};
};
});

View File

@@ -0,0 +1,199 @@
/**
* Writer.js
*
* Released under LGPL License.
* Copyright (c) 1999-2015 Ephox Corp. All rights reserved
*
* License: http://www.tinymce.com/license
* Contributing: http://www.tinymce.com/contributing
*/
/**
* This class is used to write HTML tags out it can be used with the Serializer or the SaxParser.
*
* @class tinymce.html.Writer
* @example
* var writer = new tinymce.html.Writer({indent: true});
* var parser = new tinymce.html.SaxParser(writer).parse('<p><br></p>');
* console.log(writer.getContent());
*
* @class tinymce.html.Writer
* @version 3.4
*/
define("tinymce/html/Writer", [
"tinymce/html/Entities",
"tinymce/util/Tools"
], function(Entities, Tools) {
var makeMap = Tools.makeMap;
/**
* Constructs a new Writer instance.
*
* @constructor
* @method Writer
* @param {Object} settings Name/value settings object.
*/
return function(settings) {
var html = [], indent, indentBefore, indentAfter, encode, htmlOutput;
settings = settings || {};
indent = settings.indent;
indentBefore = makeMap(settings.indent_before || '');
indentAfter = makeMap(settings.indent_after || '');
encode = Entities.getEncodeFunc(settings.entity_encoding || 'raw', settings.entities);
htmlOutput = settings.element_format == "html";
return {
/**
* Writes the a start element such as <p id="a">.
*
* @method start
* @param {String} name Name of the element.
* @param {Array} attrs Optional attribute array or undefined if it hasn't any.
* @param {Boolean} empty Optional empty state if the tag should end like <br />.
*/
start: function(name, attrs, empty) {
var i, l, attr, value;
if (indent && indentBefore[name] && html.length > 0) {
value = html[html.length - 1];
if (value.length > 0 && value !== '\n') {
html.push('\n');
}
}
html.push('<', name);
if (attrs) {
for (i = 0, l = attrs.length; i < l; i++) {
attr = attrs[i];
html.push(' ', attr.name, '="', encode(attr.value, true), '"');
}
}
if (!empty || htmlOutput) {
html[html.length] = '>';
} else {
html[html.length] = ' />';
}
if (empty && indent && indentAfter[name] && html.length > 0) {
value = html[html.length - 1];
if (value.length > 0 && value !== '\n') {
html.push('\n');
}
}
},
/**
* Writes the a end element such as </p>.
*
* @method end
* @param {String} name Name of the element.
*/
end: function(name) {
var value;
/*if (indent && indentBefore[name] && html.length > 0) {
value = html[html.length - 1];
if (value.length > 0 && value !== '\n')
html.push('\n');
}*/
html.push('</', name, '>');
if (indent && indentAfter[name] && html.length > 0) {
value = html[html.length - 1];
if (value.length > 0 && value !== '\n') {
html.push('\n');
}
}
},
/**
* Writes a text node.
*
* @method text
* @param {String} text String to write out.
* @param {Boolean} raw Optional raw state if true the contents wont get encoded.
*/
text: function(text, raw) {
if (text.length > 0) {
html[html.length] = raw ? text : encode(text);
}
},
/**
* Writes a cdata node such as <![CDATA[data]]>.
*
* @method cdata
* @param {String} text String to write out inside the cdata.
*/
cdata: function(text) {
html.push('<![CDATA[', text, ']]>');
},
/**
* Writes a comment node such as <!-- Comment -->.
*
* @method cdata
* @param {String} text String to write out inside the comment.
*/
comment: function(text) {
html.push('<!--', text, '-->');
},
/**
* Writes a PI node such as <?xml attr="value" ?>.
*
* @method pi
* @param {String} name Name of the pi.
* @param {String} text String to write out inside the pi.
*/
pi: function(name, text) {
if (text) {
html.push('<?', name, ' ', encode(text), '?>');
} else {
html.push('<?', name, '?>');
}
if (indent) {
html.push('\n');
}
},
/**
* Writes a doctype node such as <!DOCTYPE data>.
*
* @method doctype
* @param {String} text String to write out inside the doctype.
*/
doctype: function(text) {
html.push('<!DOCTYPE', text, '>', indent ? '\n' : '');
},
/**
* Resets the internal buffer if one wants to reuse the writer.
*
* @method reset
*/
reset: function() {
html.length = 0;
},
/**
* Returns the contents that got serialized.
*
* @method getContent
* @return {String} HTML contents that got written down.
*/
getContent: function() {
return html.join('').replace(/\n$/, '');
}
};
};
});