1.2.2b2
click here to add a description
click here to add a homepage
jQuery JavaScript Library — Read more
Tagging the 1.2.2b2 release.
Fixed #2027 - make sure that cloned elements (within appendTo, etc.) have their events cloned by default.
Fixed a problem in the unit tests for IE where an optimization made the test case not work. The optimization was fine. It was the unit test that was taking a short cut which caused it to fail after the optimization went in.
Changed the $(document).ready() code to try and solve some problems in Safari, Opera, and IE.
added version to all builds; changed current version to 1.2.2-pre - to be updated after each release so its clear that a build is from which milestone; build.xml cleanup (removing uselss lite, docs, test and _with_plguins targets); deleted useless docs build files (replaced by wiki and wiki xml exporter and api browsers)
offset: make sure there is a parent node to work on while calculating scroll offsets (#2073)
offset: make sure border width always gets calculated in pixels
Fixed #2077 by adding a var to make it a local variable instead of initializing a global one.
Short-circuit jQuery.event.fix if it has already fixed the current event object.
Insure width/height do not return a negative number.
Second part of the fix for #2071. An empty string "" was being sent to .bind() and when the events were being cleared it went in to an infinite recursive loop till memory was out. The test was !types in the function and changing it to types == undefined fixed the error.
Changed CRLF end of lines to just LF. Other than that there were no changes in this commit. Try svn diff -x --ignore-eol-style -r 4224:4225 to see that.
Fixed #2076 where .val() could return an array instead of undefined when the jquery object was empty. The bug was created during the .val() refactoring for the javascript strict FF ticket.
width and height methods are now working properly
Fixed #1887 - trigger fails if extra is not a function
Fixed #2070 by adding a test for !nodeType to isArrayLike so DOM elements like SELECT are not considered array-like (even though they really are).
Fixed #2069. The ready helper and shortcuts act the same. You can also still bind, unbind and trigger the ready event on the document element but doing so follows the events API unlike the ready helper method.
test for #2070
Test for #2069
Fixed #2062 by adding a check to see if the selector is array-like in .not() before testing it as an array. Otherwise it does a straight comparison during the filter test.
Fixed #2064
Tagging the 1.2.2b release.
Fixed #2037 where Opera would mis-state the value of 'display' after an innerHTML was done in some cases.
Updating changlog
Added code to remove the script tag on successful jsonp calls.
The extra & was getting gobbled, oops.
Added a fix for bug #2020 - if you want to do data: {callback: "?"}, do jsonp: "callback" instead.
Cleaned up mouseenter and mouseleave special events
Changed window.frameElement to window == top because of some access denied errors when the iframe wasn't same origin.
Added code so an iframe in IE won't count on doScroll to determine if the dom is ready. We will probably have to do more investigation on a better (and consistant) method for inside an iframe in IE.
