Socket
Socket
Sign inDemoInstall

cheerio

Package Overview
Dependencies
4
Maintainers
1
Versions
69
Alerts
File Explorer

Advanced tools

npm Scripts
License

Install Socket

Detect and block malicious and high-risk dependencies

Install

Comparing version 0.8.3 to 0.9.0

benchmarks/htmlparser.js
test/api.attributes.js
test/api.manipulation.js
test/api.traversing.js
test/cheerio.js
test/fixtures.js
test/parse.js
test/render.js

18

History.md

@@ -0,1 +1,11 @@

0.9.0 / 2012-07-24
==================
* Added node 8.x support
* Removed node 4.x support
* Add html(dom) support (@wvl)
* fixed xss vulnerabilities on .attr(), .text(), & .html() (@benatkin, @FB55)
* Rewrote tests into javascript, removing coffeescript dependency (@davidchambers)
* Tons of cleanup (@davidchambers)
0.8.3 / 2012-06-12

@@ -6,3 +16,3 @@ ==================

0.8.2 / 2012-06-11
0.8.2 / 2012-06-11
==================

@@ -25,3 +35,3 @@

* ignoreWhitespace now on by default again. See #55 for context.
* Changed $(":root") to $.root(), cleaned up $.clone()
* Changed $(':root') to $.root(), cleaned up $.clone()
* Support for .eq(i) thanks to @alexbardas

@@ -63,3 +73,3 @@ * Removed support for node 0.4.x

* `$.html()` returns the full HTML string. `$.html([cheerioObject])` will return the outer(selected element's tag) and inner HTML of that object
* Fixed bug that prevented HTML strings with depth (eg. `append("<ul><li><li></ul>")`) from getting `parent`, `next`, `prev` attributes.
* Fixed bug that prevented HTML strings with depth (eg. `append('<ul><li><li></ul>')`) from getting `parent`, `next`, `prev` attributes.
* Halted [htmlparser2](https://github.com/FB55/node-htmlparser) at v2.2.2 until single attributes bug gets fixed.

@@ -124,3 +134,3 @@

* Added benchmark directory for future speed tests
* $("...").dom() was funky, so it was removed in favor of $("...").get(). $.dom() still works the same.
* $('...').dom() was funky, so it was removed in favor of $('...').get(). $.dom() still works the same.
* $.root now correctly static across all instances of $

@@ -127,0 +137,0 @@ * Added a screencast

5

index.js

@@ -8,5 +8,2 @@ var fs = require('fs');

*/
exports.version = (function() {
var pkg = fs.readFileSync(__dirname + '/package.json', 'utf8');
return JSON.parse(pkg).version;
})();
exports.version = require('./package').version;

53

lib/api/attributes.js

@@ -1,3 +0,3 @@

var _ = require("underscore"),
$ = require("../cheerio"),
var _ = require('underscore'),
$ = require('../cheerio'),
rclass = /[\n\t\r]/g,

@@ -14,20 +14,11 @@ rspace = /\s+/;

});
return this;
};
var hasClass = exports.hasClass = function(selector) {
var className = " " + selector + " ",
classes,
elem;
for(var i = 0; i < this.length; i++) {
elem = this[i];
// Add spaces to support multiple classes
classes = (" " + elem.attribs["class"] + " ").replace(rclass, " ");
if ($.isTag(elem) && elem.attribs && classes.indexOf(className) > -1) {
return true;
}
}
return false;
var hasClass = exports.hasClass = function(className) {
return _.any(this, function(elem) {
var attrs = elem.attribs;
return attrs && _.contains((attrs['class'] || '').split(/\s+/), className);
});
};

@@ -40,3 +31,3 @@

var $this = $(this),
className = $this.attr('class') || "";
className = $this.attr('class') || '';

@@ -57,3 +48,3 @@ $this.addClass(value.call(this, i, className));

for(var i = 0; i < numElements; i++) {
for (var i = 0; i < numElements; i++) {
$elem = $(this[i]);

@@ -64,6 +55,6 @@ // If selected element isnt a tag, move on

// If we don't already have classes
if (!$elem.attr("class")) {
if (!$elem.attr('class')) {
$elem.attr('class', classNames.join(' ').trim());
} else {
setClass = " " + $elem.attr("class") + " ";
setClass = ' ' + $elem.attr('class') + ' ';
numClasses = classNames.length;

@@ -73,4 +64,4 @@

for (var j = 0; j < numClasses; j++) {
if (!~setClass.indexOf(" " + classNames[j] + " "))
setClass += classNames[j] + " ";
if (!~setClass.indexOf(' ' + classNames[j] + ' '))
setClass += classNames[j] + ' ';
}

@@ -91,3 +82,3 @@

var $this = $(this),
className = $this.attr('class') || "";
className = $this.attr('class') || '';
$this.removeClass(value.call(this, j, className));

@@ -98,5 +89,5 @@ });

// If value isnt undefined and also not a string
if(value !== undefined && !_.isString(value)) return this;
if (value !== undefined && !_.isString(value)) return this;
var classNames = (value || "").split(rspace),
var classNames = (value || '').split(rspace),
numClasses = classNames.length,

@@ -111,4 +102,4 @@ className,

if(!$.isTag(this[i]) || !className) continue;
else if(!value) {
if (!$.isTag(this[i]) || !className) continue;
else if (!value) {
this[i].attribs['class'] = '';

@@ -119,7 +110,7 @@ continue;

// Separate out the classes
ret = (" " + className + " ").replace(rclass, " ");
ret = (' ' + className + ' ').replace(rclass, ' ');
for (var j = 0; j < numClasses; j++) {
className = classNames[j];
ret = ret.replace(" " + className + " ", " ");
ret = ret.replace(' ' + className + ' ', ' ');
}

@@ -133,2 +124,2 @@

module.exports = $.fn.extend(exports);
module.exports = $.fn.extend(exports);

28

lib/api/core.js
var underscore = require('underscore'),
$ = require('../cheerio'),
isArray = Array.isArray;
var size = exports.size = function() {

@@ -16,3 +16,3 @@ return this.length;

if(~num) {
if (~num) {
return this[num];

@@ -27,3 +27,3 @@ } else {

var ret = this.constructor();
if (isArray(elems)) {

@@ -34,10 +34,10 @@ push.apply(ret, elems);

}
ret.prevObject = this;
ret.context = this.context;
if (name === "find") {
ret.selector = this.selector + (this.selector ? " " : "") + selector;
} else if(name) {
ret.selector = this.selector + "." + name + "(" + selector + ")";
if (name === 'find') {
ret.selector = this.selector + (this.selector ? ' ' : '') + selector;
} else if (name) {
ret.selector = this.selector + '.' + name + '(' + selector + ')';
}

@@ -51,9 +51,9 @@ return ret;

element = raw;
while (element.prev) {
element = element.prev;
}
siblings.push(element);
while (element.next) {

@@ -63,6 +63,6 @@ element = element.next;

}
return siblings;
};
module.exports = $.fn.extend(exports);
module.exports = $.fn.extend(exports);

47

lib/api/manipulation.js

@@ -22,6 +22,6 @@ var _ = require('underscore'),

for(var i = 0; i < len; i++) {
for (var i = 0; i < len; i++) {
elem = elems[i];
// If a cheerio object
if(elem.cheerio) {
if (elem.cheerio) {
dom = dom.concat(elem.toArray());

@@ -41,7 +41,7 @@ } else {

this.each(function() {
if(_.isFunction(elems[0])) {
if (_.isFunction(elems[0])) {
// No yet supported
return this;
} else {
if(!this.children) this.children = [];
if (!this.children) this.children = [];
this.children = this.children.concat(dom);

@@ -65,7 +65,7 @@ updateDOM(this.children, this);

this.each(function() {
if(_.isFunction(elems[0])) {
if (_.isFunction(elems[0])) {
// No yet supported
return this;
} else {
if(!this.children) this.children = [];
if (!this.children) this.children = [];
this.children = dom.concat(this.children);

@@ -88,3 +88,3 @@ updateDOM(this.children, this);

// If not found, move on
if(!~index) return;
if (!~index) return;

@@ -112,3 +112,3 @@ // Add element after `this` element

// If not found, move on
if(!~index) return;
if (!~index) return;

@@ -134,3 +134,3 @@ // Add element before `this` element

// Filter if we have selector
if(selector)
if (selector)
elems = elems.find(selector);

@@ -142,3 +142,3 @@

if(!~index) return;
if (!~index) return;

@@ -156,10 +156,10 @@ siblings.splice(index, 1);

var replaceWith = exports.replaceWith = function(content) {
var content = (content.cheerio) ? content.toArray() : $.parse.eval(content);
content = content.cheerio ? content.toArray() : $.parse.eval(content);
this.each(function() {
var siblings = this.parent.children,
index = siblings.indexOf(this);
if(!~index) return;
if (!~index) return;
siblings.splice.apply(siblings, [index, 1].concat(content));

@@ -170,3 +170,3 @@

});
return this;

@@ -182,12 +182,13 @@ };

var html = exports.html = function(str) {
if(str === undefined || typeof(str) === 'object') {
if(!this[0] || !this[0].children) return null;
var html = exports.html = function(content) {
if (content === undefined) {
if (!this[0] || !this[0].children) return null;
return $.html(this[0].children);
}
str = $.parse.eval(str);
content = content.cheerio ? content.toArray() : $.parse.eval(content);
this.each(function() {
this.children = str;
this.children = content;
updateDOM(this.children, this);
});

@@ -204,3 +205,3 @@

// If `str` blank or an object
if(!str || typeof(str) === 'object') {
if (!str || typeof str === 'object') {
return $.text(this);

@@ -216,3 +217,3 @@ } else if (_.isFunction(str)) {

var elem = {
data : str,
data : $.encode(str),
type : 'text',

@@ -219,0 +220,0 @@ parent : null,

60

lib/api/parse.js

@@ -5,2 +5,3 @@ /*

var htmlparser = require('htmlparser2'),
_ = require('underscore'),
$ = require('../cheerio'),

@@ -29,3 +30,3 @@ isArray = Array.isArray;

update(dom, root);
return root;

@@ -48,33 +49,30 @@ };

parent = parent || null;
var prevIndex = -1,
lastElem = null,
len = dom.length;
for(var i = 0; i < len; i++) {
var prevElem = null;
_.each(dom, function(elem) {
// If tag and no attributes, add empty object
if($.isTag(dom[i].type) && dom[i].attribs === undefined)
dom[i].attribs = {};
if ($.isTag(elem.type) && elem.attribs === undefined)
elem.attribs = {};
// Set parent
dom[i].parent = parent;
elem.parent = parent;
// Previous Sibling
dom[i].prev = dom[prevIndex] || null;
elem.prev = prevElem;
// Next sibling
dom[i].next = null;
if(lastElem) lastElem.next = dom[i];
elem.next = null;
if (prevElem) prevElem.next = elem;
// Run through the children
if(dom[i].children)
connect(dom[i].children, dom[i]);
else if($.isTag(dom[i].type))
dom[i].children = [];
if (elem.children)
connect(elem.children, elem);
else if ($.isTag(elem.type))
elem.children = [];
// Get ready for next element
prevIndex = i;
lastElem = dom[i];
}
prevElem = elem;
});
return dom;

@@ -85,3 +83,3 @@ };

Update the dom structure, for one changed layer
* Much faster than reconnecting

@@ -92,5 +90,5 @@ */

arr = isArray(arr) ? arr : [arr];
// Update neighbors
for(var i = 0; i < arr.length; i++) {
for (var i = 0; i < arr.length; i++) {
arr[i].prev = arr[i-1] || null;

@@ -100,6 +98,6 @@ arr[i].next = arr[i+1] || null;

}
// Update parent
parent.children = arr;
return parent;

@@ -106,0 +104,0 @@ };

63

lib/api/render.js

@@ -21,7 +21,7 @@ /*

var formatAttrs = function(attributes) {
if(!attributes) return '';
if (!attributes) return '';
var output = [],
value;
// Loop through the attributes

@@ -36,3 +36,3 @@ for (var key in attributes) {

}
return output.join(' ');

@@ -60,3 +60,3 @@ };

include: 1,
yield: 1
'yield': 1
};

@@ -82,29 +82,24 @@

var len = dom.length,
output = [],
tidy = (opts.tidy ? true : false),
ending = (tidy ? "\n" : ""),
elem;
var output = [],
tidy = !!opts.tidy;
xmlMode = $.fn.options.xmlMode;
ignoreWhitespace = $.fn.options.ignoreWhitespace;
for(var i = 0; i < len; i++) {
elem = dom[i];
var pushVal = "";
_.each(dom, function(elem) {
var pushVal;
if(tagType[elem.type])
if (tagType[elem.type])
pushVal = renderTag(elem);
else if(elem.type === 'directive')
else if (elem.type === 'directive')
pushVal = renderDirective(elem);
else if(elem.type === 'comment')
else if (elem.type === 'comment')
pushVal = renderComment(elem);
else
pushVal = renderText(elem);
var spacing = '';
if (tidy) {
for(var k = 0; k < depth; k++){ spacing += ' ' };
output.push(spacing + pushVal + "\n");
spacing = Array(depth + 1).join(' ');
output.push(spacing + pushVal + '\n');
} else {

@@ -115,11 +110,11 @@ output.push(pushVal);

depth++;
if(elem.children)
if (elem.children)
output.push(render(elem.children, { tidy: tidy }));
depth--;
if((!singleTag[elem.name] || xmlMode) && tagType[elem.type])
output.push(spacing + "</" + elem.name + ">" + ending);
}
if ((!singleTag[elem.name] || xmlMode) && tagType[elem.type])
output.push(spacing + '</' + elem.name + '>');
});
return output.join('');
return output.join(tidy ? '\n' : '');
};

@@ -129,19 +124,11 @@

var tag = '<' + elem.name;
if(elem.attribs && _.size(elem.attribs)) {
if (elem.attribs && _.size(elem.attribs)) {
tag += ' ' + formatAttrs(elem.attribs);
}
if(xmlMode || !singleTag[elem.name]) {
tag += '>';
} else {
tag = tag.trim().replace(/\/$/, '');
tag += '/>';
}
return tag;
return tag + (!xmlMode && singleTag[elem.name] ? '/>' : '>');
};
var renderDirective = exports.render.renderDirective = function(elem) {
return "<" + elem.data + ">";
return '<' + elem.data + '>';
};

@@ -148,0 +135,0 @@

28

lib/api/traversing.js

@@ -1,7 +0,7 @@

var _ = require("underscore"),
var _ = require('underscore'),
select = require('cheerio-select'),
$ = require("../cheerio");
$ = require('../cheerio');
var find = exports.find = function(selector) {
if(!selector) return this;
if (!selector) return this;
try {

@@ -16,3 +16,3 @@ var elem = select(selector, this.toArray());

var parent = exports.parent = function(elem) {
if(this[0] && this[0].parent)
if (this[0] && this[0].parent)
return $(this[0].parent);

@@ -24,7 +24,7 @@ else

var next = exports.next = function(elem) {
if(!this[0]) return $();
if (!this[0]) return $();
var nextSibling = this[0].next;
while(nextSibling) {
if($.isTag(nextSibling)) return $(nextSibling);
while (nextSibling) {
if ($.isTag(nextSibling)) return $(nextSibling);
nextSibling = nextSibling.next;

@@ -36,7 +36,7 @@ }

var prev = exports.prev = function(elem) {
if(!this[0]) return $();
if (!this[0]) return $();
var prevSibling = this[0].prev;
while(prevSibling) {
if($.isTag(prevSibling)) return $(prevSibling);
while (prevSibling) {
if ($.isTag(prevSibling)) return $(prevSibling);
prevSibling = prevSibling.prev;

@@ -48,3 +48,3 @@ }

var siblings = exports.siblings = function(elem) {
if(!this[0]) return $();
if (!this[0]) return $();

@@ -63,3 +63,3 @@ var self = this,

var children = exports.children = function(selector) {
if(!this[0] || !this[0].children) return $();
if (!this[0] || !this[0].children) return $();

@@ -70,4 +70,4 @@ var children = _.filter(this[0].children, function(elem) {

if(selector === undefined) return $(children);
else if(_.isNumber(selector)) return $(children[selector]);
if (selector === undefined) return $(children);
else if (_.isNumber(selector)) return $(children[selector]);

@@ -74,0 +74,0 @@ return $(children).find(selector);

247

lib/api/utils.js
/*
Module Dependencies
*/
var _ = require("underscore"),
$ = require("../cheerio"),
var _ = require('underscore'),
$ = require('../cheerio'),
entities = require('entities');

@@ -37,2 +37,7 @@

// Expose encode and decode methods from FB55's node-entities library
// 0 = XML, 1 = HTML4 and 2 = HTML5
var encode = exports.encode = function(str) { return entities.encode(str, 0); };
var decode = exports.decode = function(str) { return entities.decode(str, 2); };
/*

@@ -42,3 +47,3 @@ isTag(type) includes <script> and <style> tags

var isTag = exports.isTag = function(type) {
if(type.type) type = type.type;
if (type.type) type = type.type;
return tags[type] || false;

@@ -49,3 +54,3 @@ };

var type = exports.type = function(obj) {
if(obj === null)
if (obj === null)
return String(obj);

@@ -56,98 +61,98 @@ else

var merge = exports.merge = function( first, second ) {
var i = first.length,
j = 0;
var merge = exports.merge = function(first, second) {
var i = first.length,
j = 0;
if ( typeof second.length === "number" ) {
for ( var l = second.length; j < l; j++ ) {
first[ i++ ] = second[ j ];
}
if (typeof second.length === 'number') {
for (var l = second.length; j < l; j++) {
first[i++] = second[j];
}
} else {
while ( second[j] !== undefined ) {
first[ i++ ] = second[ j++ ];
}
}
} else {
while (second[j] !== undefined) {
first[i++] = second[j++];
}
}
first.length = i;
first.length = i;
return first;
return first;
};
var makeArray = exports.makeArray = function( array, results ) {
var makeArray = exports.makeArray = function(array, results) {
var ret = results || [],
type = $.type(array);
if(!array) return ret;
if ( array.length == null || type === "string" || type === "function" || type === "regexp") {
push.call( ret, array );
} else {
merge( ret, array );
}
if (!array) return ret;
if (array.length == null || type === 'string' || type === 'function' || type === 'regexp') {
push.call(ret, array);
} else {
merge(ret, array);
}
return ret;
};
var inArray = exports.inArray = function( elem, array, i ) {
var len;
var inArray = exports.inArray = function(elem, array, i) {
var len;
if ( array ) {
if ( indexOf ) {
return indexOf.call( array, elem, i );
}
if (array) {
if (indexOf) {
return indexOf.call(array, elem, i);
}
len = array.length;
i = i ? i < 0 ? Math.max( 0, len + i ) : i : 0;
len = array.length;
i = i ? i < 0 ? Math.max(0, len + i) : i : 0;
for ( ; i < len; i++ ) {
// Skip accessing in sparse arrays
if ( i in array && array[ i ] === elem ) {
return i;
}
}
}
for (; i < len; i++) {
// Skip accessing in sparse arrays
if (i in array && array[i] === elem) {
return i;
}
}
}
return -1;
return -1;
};
// args is for internal usage only
var each = exports.each = function( object, callback, args ) {
var name, i = 0,
length = object.length,
isObj = length === undefined || _.isFunction( object );
var each = exports.each = function(object, callback, args) {
var name, i = 0,
length = object.length,
isObj = length === undefined || _.isFunction(object);
if ( args ) {
if ( isObj ) {
for ( name in object ) {
if ( callback.apply( object[ name ], args ) === false ) {
break;
}
}
} else {
for ( ; i < length; ) {
if ( callback.apply( object[ i++ ], args ) === false ) {
break;
}
}
}
if (args) {
if (isObj) {
for (name in object) {
if (callback.apply(object[name], args) === false) {
break;
}
}
} else {
for (; i < length;) {
if (callback.apply(object[i++], args) === false) {
break;
}
}
}
// A special, fast, case for the most common use of each
} else {
if ( isObj ) {
for ( name in object ) {
if ( callback.call( object[ name ], name, object[ name ] ) === false ) {
break;
}
}
} else {
for ( ; i < length; ) {
if ( callback.call( object[ i ], i, object[ i++ ] ) === false ) {
break;
}
}
}
}
// A special, fast, case for the most common use of each
} else {
if (isObj) {
for (name in object) {
if (callback.call(object[name], name, object[name]) === false) {
break;
}
}
} else {
for (; i < length;) {
if (callback.call(object[i], i, object[i++]) === false) {
break;
}
}
}
}
return object;
return object;
};

@@ -157,31 +162,31 @@

// The value/s can optionally be executed if it's a function
var access = exports.access = function( elems, key, value, exec, fn, pass ) {
var length = elems.length;
var access = exports.access = function(elems, key, value, exec, fn, pass) {
var length = elems.length;
// Setting many attributes
if ( typeof key === "object" ) {
for ( var k in key ) {
access( elems, k, key[k], exec, fn, value );
}
return elems;
}
// Setting many attributes
if (typeof key === 'object') {
for (var k in key) {
access(elems, k, key[k], exec, fn, value);
}
return elems;
}
// Setting one attribute
if ( value !== undefined ) {
// Optionally, function values get executed if exec is true
exec = !pass && exec && _.isFunction(value);
// Setting one attribute
if (value !== undefined) {
// Optionally, function values get executed if exec is true
exec = !pass && exec && _.isFunction(value);
for ( var i = 0; i < length; i++ ) {
fn( elems[i], key, exec ? value.call( elems[i], i, fn( elems[i], key ) ) : value, pass );
}
for (var i = 0; i < length; i++) {
fn(elems[i], key, exec ? value.call(elems[i], i, fn(elems[i], key)) : value, pass);
}
return elems;
}
return elems;
}
// Getting an attribute
return length ? fn( elems[0], key ) : undefined;
// Getting an attribute
return length ? fn(elems[0], key) : undefined;
};
var attr = exports.attr = function( elem, name, value, pass ) {
var type = elem.type;
var attr = exports.attr = function(elem, name, value, pass) {
var type = elem.type;

@@ -194,5 +199,8 @@ if (!elem || !isTag(elem))

}
// Return the entire attribs object if no attribute specified
if (!name) {
for(var a in elem.attribs) {
elem.attribs[a] = decode(elem.attribs[a]);
}
return elem.attribs;

@@ -207,7 +215,7 @@ }

// Set the attribute
elem.attribs[name] = "" + value;
elem.attribs[name] = '' + encode(value);
}
} else {
// Get the attributes
return elem.attribs[name];
} else if(elem.attribs[name]) {
// Get the (decoded) attribute
return decode(elem.attribs[name]);
}

@@ -217,6 +225,6 @@ };

var removeAttr = exports.removeAttr = function(elem, name) {
if(!isTag(elem.type) || !elem.attribs || !elem.attribs[name])
if (!isTag(elem.type) || !elem.attribs || !elem.attribs[name])
return;
if(rboolean.test(elem.attribs[name]))
if (rboolean.test(elem.attribs[name]))
elem.attribs[name] = false;

@@ -230,14 +238,14 @@ else

var ret = "",
var ret = '',
len = elems.length,
elem;
for(var i = 0; i < len; i ++) {
for (var i = 0; i < len; i ++) {
elem = elems[i];
if(elem.type === 'text') ret += entities.decode(elem.data, 2);
else if(elem.children && elem.type !== 'comment') {
if (elem.type === 'text') ret += decode(elem.data);
else if (elem.children && elem.type !== 'comment') {
ret += text(elem.children);
}
}
return ret;

@@ -252,3 +260,3 @@ };

var root = $.parse(html);
function fn(selector, context, r) {

@@ -259,6 +267,6 @@ // Overwrite our original root if we explicitly pass in a root

}
// Add the root to cheerio
$.extend({ '_root': root });
return _(fn).extend($);

@@ -273,3 +281,3 @@ };

} else {
return "";
return '';
}

@@ -285,3 +293,3 @@ };

} else {
return "";
return '';
}

@@ -296,3 +304,3 @@ };

} else {
return "";
return '';
}

@@ -305,2 +313,3 @@ };

module.exports = $.extend(exports);

47

lib/cheerio.js

@@ -8,3 +8,3 @@ /*

_ = require('underscore');
var cheerio = (function() {

@@ -14,7 +14,7 @@ var cheerio = function(selector, context, root) {

};
// A simple way to check for HTML strings or ID strings
// Prioritize #id over <tag>
var quickExpr = /^(?:[^#<]*(<[\w\W]+>)[^>]*$|#([\w\-]*)$)/;
cheerio.fn = cheerio.prototype = {

@@ -25,18 +25,18 @@ cheerio : '[cheerio object]',

// Handle $(''), $(null), and $(undefined)
if(!selector) return this;
if (!selector) return this;
// Handle the root
if(root) {
if (root) {
cheerio.extend({ '_root' : root });
if(typeof context === 'string')
if (typeof context === 'string')
selector = context + ' ' + selector;
context = root;
}
// Handle strings
if(typeof selector === 'string') {
if (typeof selector === 'string') {
var match = null;
// Handle HTML strings
if (selector.charAt(0) === "<" && selector.charAt(selector.length - 1) === ">" && selector.length >= 3) {
if (/^<.+>$/.test(selector)) {
match = [null, selector, null];

@@ -63,4 +63,4 @@ } else {

}
if(!context || context.cheerio) {
if (!context || context.cheerio) {
// HANDLE : $(expr, $(...))

@@ -71,3 +71,3 @@ return this.constructor(context || root).find(selector);

// HANDLE : $(expr, context)
if(typeof context === 'string') {
if (typeof context === 'string') {
context = cheerio.parse(context);

@@ -79,9 +79,9 @@ }

}
return cheerio.makeArray(selector, this);
},
options : {
ignoreWhitespace: false,
xmlMode: false,
lowerCaseTags: false
ignoreWhitespace : false,
xmlMode : false,
lowerCaseTags : false
},

@@ -92,5 +92,5 @@ selector : '',

};
cheerio.fn.init.prototype = cheerio.fn;
// Use underscores extend

@@ -102,3 +102,3 @@ cheerio.extend = cheerio.fn.extend = function (obj) {

return cheerio;
})();

@@ -111,5 +111,4 @@

*/
var api = ['core', 'utils', 'parse', 'render', 'attributes', 'traversing', 'manipulation'];
for (var t = 0; t < api.length; t++) {
require('./api/' + api[t] + '.js');
}
_.each(['core', 'utils', 'parse', 'render', 'attributes', 'traversing', 'manipulation'], function(name) {
require('./api/' + name + '.js');
});

5

package.json

@@ -6,3 +6,3 @@ {

"keywords": ["htmlparser", "jquery", "selector", "scraper"],
"version": "0.8.3",
"version": "0.9.0",
"repository": {

@@ -24,4 +24,3 @@ "type": "git",

"mocha" : "*",
"expect.js" : "*",
"coffee-script": "*"
"expect.js" : "*"
},

@@ -28,0 +27,0 @@ "scripts": {

53

Readme.md

@@ -10,4 +10,4 @@ # cheerio [![Build Status](https://secure.travis-ci.org/MatthewMueller/cheerio.png?branch=master)](http://travis-ci.org/MatthewMueller/cheerio)

var cheerio = require("cheerio"),
$ = cheerio.load("<h2 class = 'title'>Hello world</h2>");
var cheerio = require('cheerio'),
$ = cheerio.load('<h2 class = "title">Hello world</h2>');

@@ -35,3 +35,3 @@ $('h2.title').text('Hello there!');

__&#10049; Insanely flexible:__
Cheerio wraps around @tautologistics forgiving htmlparser. Cheerio can parse nearly any HTML or XML document.
Cheerio wraps around @FB55's forgiving htmlparser. Cheerio can parse nearly any HTML or XML document.

@@ -104,5 +104,5 @@ ## What about JSDOM?

This selector method is the starting point for traversing and manipulating the document. Like jQuery, it's the primary method for selecting elements in the document, but unlike jQuery it's built on top of the soup-select library, not the Sizzle engine.
This selector method is the starting point for traversing and manipulating the document. Like jQuery, it's the primary method for selecting elements in the document, but unlike jQuery it's built on top of the CSSSelect library, which implements most of the Sizzle selectors.
$(".apple", '#fruits').text()
$('.apple', '#fruits').text()
=> Apple

@@ -116,5 +116,2 @@

See [cheerio-soupselect](https://github.com/MatthewMueller/cheerio-soupselect) for a list of all available selectors,
as well as more detailed documentation regarding what is supported and what isn't.
### Attributes

@@ -190,3 +187,3 @@ Methods for getting and modifying attributes.

#### .next()
Gets the next sibling thats an element of the first selected element.
Gets the next sibling of the first selected element.

@@ -197,3 +194,3 @@ $('.apple').next().hasClass('orange')

#### .prev()
Gets the previous sibling thats an element of the first selected element.
Gets the previous sibling of the first selected element.

@@ -243,3 +240,3 @@ $('.orange').prev().hasClass('apple')

#### .eq( i )
Reduce the set of matched elements to the one at the specified index. Use `.eq(-i)` to count backwards from the last selected element.
Reduce the set of matched elements to the one at the specified index. Use `.eq(-i)` to count backwards from the last selected element.

@@ -410,3 +407,3 @@ $('li').eq(0).text()

$.dom()
=> [{
=> [{
type: 'tag',

@@ -425,3 +422,3 @@ name: 'ul',

prev: null,
next: null
next: null
}]

@@ -461,2 +458,22 @@

This will download the development packages and run the test suite.
## Contributors
These are some of the contributors that have made cheerio possible:
project: cheerio
commits: 292
active : 78 days
files : 26
authors:
223 Matt Mueller 76.4%
33 Matthew Mueller 11.3%
15 Siddharth Mahendraker 5.1%
10 David Chambers 3.4%
4 ironchefpython 1.4%
3 Jos Shepherd 1.0%
2 alexbardas 0.7%
1 mattym 0.3%
1 Chris O'Hara 0.3%
## Special Thanks

@@ -466,9 +483,6 @@

__&#8226; @tautologistics' node-htmlparser:__
This HTML parser can parse anything and produces really consistent results, even when the HTML string has errors. This man is a genius.
__&#8226; @FB55 for node-htmlparser2 & CSSSelect:__
Felix has a knack for writing speedy parsing engines. He completely re-wrote both @tautologistic's `node-htmlparser` and @harry's `node-soupselect` from the ground up, making both of them much faster and more flexible. Cheerio would not be possible without his foundational work
__&#8226; @harryf's node-soupselect:__
What an incredibly fast and precise CSS selector engine. I never really liked the feature-rich selector engines &#8212; I think this engine strikes a great balance.
__&#8226; @jQuery team:__
__&#8226; @jQuery team for jQuery:__
The core API is the best of it's class and despite dealing with all the browser inconsistencies the code base is extremely clean and easy to follow. Much of cheerio's implementation and documentation is from jQuery. Thanks guys.

@@ -478,2 +492,3 @@

The style, the structure, the open-source"-ness" of this library comes from studying TJ's style and using many of his libraries. This dude consistently pumps out high-quality libraries and has always been more than willing to help or answer questions. You rock TJ.
## License

@@ -480,0 +495,0 @@

74

test/api.utils.js

@@ -1,17 +0,12 @@

var $ = require('../'),
expect = require('expect.js');
var expect = require('expect.js'),
$ = require('../');
/*
Examples
*/
describe('$', function() {
describe('.html', function() {
it('() : should return innerHTML; $.html(obj) should return outerHTML', function() {
var div = $('div', '<div><span>foo</span><span>bar</span></div>'),
span = div.children().get(1);
var $div = $('div', '<div><span>foo</span><span>bar</span></div>');
var span = $div.children().get(1);
expect($(span).html()).to.equal('bar');

@@ -22,36 +17,32 @@ expect($.html(span)).to.equal('<span>bar</span>');

it('(<obj>) : should accept an object, an array, or a cheerio object', function() {
var span = $("<span>foo</span>");
expect($.html(span.get(0))).to.equal('<span>foo</span>');
expect($.html(span.get())).to.equal('<span>foo</span>');
expect($.html(span)).to.equal('<span>foo</span>');
var $span = $('<span>foo</span>');
expect($.html($span.get(0))).to.equal('<span>foo</span>');
expect($.html($span.get())).to.equal('<span>foo</span>');
expect($.html($span)).to.equal('<span>foo</span>');
});
it('(<value>) : should be able to set to an empty string', function() {
var elem = $("<span>foo</span>");
elem.html('');
expect($.html(elem)).to.equal('<span></span>');
var $elem = $('<span>foo</span>').html('');
expect($.html($elem)).to.equal('<span></span>');
});
it('() : of empty cheerio object should return null', function() {
expect($().html()).to.be(null);
});
});
describe('.clone', function() {
it('() : should return a copy', function() {
var src = $("<div><span>foo</span><span>bar</span><span>baz</span></div>").children(),
elem = src.clone();
var $src = $('<div><span>foo</span><span>bar</span><span>baz</span></div>').children();
var $elem = $src.clone();
expect($elem.length).to.equal(3);
expect($elem.parent().length).to.equal(1);
expect($elem.parent()[0].type).to.equal('root');
expect($elem.text()).to.equal($src.text());
$src.text('rofl');
expect($elem.text()).to.not.equal($src.text());
});
expect(elem.length).to.equal(3);
expect(elem.parent().length).to.equal(1);
expect(elem.parent()[0].type).to.equal('root');
expect(elem.text()).to.equal(src.text());
src.text('rofl');
expect(elem.text()).to.not.equal(src.text());
});
});

@@ -62,4 +53,4 @@

it('() : should return a cheerio-wrapped root object', function() {
var $html = $.load("<div><span>foo</span><span>bar</span></div>");
$html.root().append("<div id = 'test'></div>");
var $html = $.load('<div><span>foo</span><span>bar</span></div>');
$html.root().append('<div id="test"></div>');
expect($html.html()).to.equal('<div><span>foo</span><span>bar</span></div><div id="test"></div>');

@@ -71,9 +62,12 @@ });

describe('.load', function() {
it('(html) : should retain original root after creating a new node', function() {
var $html = $.load('<body><ul id = "fruits"></ul></body>');
var $html = $.load('<body><ul id="fruits"></ul></body>');
expect($html('body')).to.have.length(1);
var script = $html('<script>');
$html('<script>');
expect($html('body')).to.have.length(1);
});
});
});
});
benchmarks/htmlparser.coffee
test/api.attributes.coffee
test/api.manipulation.coffee
test/api.traversing.coffee
test/cheerio.coffee
test/parse.coffee
test/render.coffee
.travis.yml

Sorry, the diff of this file is not supported yet

Makefile

Sorry, the diff of this file is not supported yet

test/mocha.opts

Sorry, the diff of this file is not supported yet

SocketSocket SOC 2 Logo

Product

  • Package Alerts
  • Integrations
  • Docs
  • Pricing
  • FAQ
  • Roadmap

About

Packages

Stay in touch

Get open source security insights delivered straight into your inbox.

  • Terms
  • Privacy
  • Security

Made with ⚡️ by Socket Inc