0.9.0

• 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

• Fixed minor package regression (closes #60)

0.8.2

• Now fails gracefully in cases that involve special chars, which is inline with jQuery (closes #59)
• text() now decode special entities (closes #52)
• updated travis.yml to test node 4.x

0.8.1

• fixed regression where if you created an element, it would update the root
• compatible with node 4.x (again)

0.8.0

• Updated CSS parser to use FB55/CSSselect. Cheerio now supports most CSS3 psuedo selectors thanks to @fb55.
• ignoreWhitespace now on by default again. See #55 for context.
• Changed $(':root') to $.root(), cleaned up $.clone()
• Support for .eq(i) thanks to @alexbardas
• Removed support for node 0.4.x
• Fixed memory leak where package.json was continually loaded
• Tons more tests

0.7.0

• Now testing with node v0.7.7
• Added travis-ci integration
• Replaced should.js with expect.js. Browser testing to come
• Fixed spacing between attributes and their values
• Added HTML tidy/pretty print
• Exposed node-htmlparser2 parsing options
• Revert .replaceWith(...) to be consistent with jQuery

0.6.2

• Fixed .replaceWith(...) regression

0.6.1

• Added .first(), .last(), and .clone() commands.
• Option to parse using whitespace added to .load.
• Many bug fixes to make cheerio more aligned with jQuery.
• Added $(':root') to select the highest level element.

Many thanks to the contributors that made this release happen: @ironchefpython and @siddMahen

0.6.0

• Important: $(...).html() now returns inner HTML, which is in line with the jQuery spec
• $.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.
• Halted htmlparser2 at v2.2.2 until single attributes bug gets fixed.

0.5.1

• Fixed minor regression: $(...).text(fn) would fail
• Fixed regression: HTML pages with comments would fail