Road to 1.5.1 (? complete)

gistfile1.txt

NEEDSDOCS no ticket associated
	https://github.com/jquery/jquery/commit/4e97543
		FEATURE (SMALL) SHOULD THIS BE REMOVED BEFORE 1.5.1??
		.invert() method on promises
			returns an inverted promise that is resolved/rejected when the underlying deferred is rejected/resolved)

	https://github.com/jquery/jquery/commit/e0b1bb8
		http://api.jquery.com/extending-ajax/ <-- the code example at the very bottom should be updated to match the new regex

	#8122
		jQuery.support.cors = true;
		For non supported environment that allow cross-domain while not supporting CORS is to set jQuery.support.cors to true right after jQuery has been loaded

These + 1.5 look like b/c breakers for context option handling
	https://github.com/jquery/jquery/commit/485fa0c and https://github.com/jquery/jquery/commit/5b421fe
	context option handling needsdocs in general

Fixed bugs:
	- #2551 Make sure .val() works after form.reset() in IE
	- #4537 Make sure .clone(true) correctly clones namespaced events
	- #4966 Don't add "px" to unit-less properties when animating them
	- #6774
	- #7531
		--> Fix again for IE9RC. Enhances ajaxSetup so that it can take an optional target option, in which case target will be updated instead of ajaxSettings. That way, fields that shouldn't be deep extended can be listed and dealt with in one place. jQuery.ajax now makes use of ajaxSetup with target to create its internal settings object
		--> NEEDSDOCS new parameter for jQuery.ajaxSetup
	- #7568 Follow-up fix for #5862. Objects with a length property weren't serialized properly by jQuery.param
	- #7653 Changes regexp to detect local protocol so that it will accept any protocol finishing by -extension
	- #7668 Sizzle and jQuery QUnit tests are out of sync
	- #7912 Make $.fx.prototype.cur() compatible with cssHooks
	- #7922 live('click') does not fire when when live('submit') is bound first in IE
	- #7945 Make jQuery.param() serialize plain objects with a property named jquery correctly
	- #8033 jQuery 1.4.4+ fails to load on pages with old Prototype (<= 1.5) or Current Prototype + Scriptaculous in IE
	- #8039
	- #8052 Update jQuery.support.noCloneEvent test to function properly in IE9
		--> Fixes bigges showbreaker for IE9
	- #8095 Properly handles the case where browser cache needs to be bypassed while server-side logic still delivers proper 304 responses. Unit test added
	- #8098 Use the fast document.head when available
	- #8099 Always restore to correct display value based on element's expected default display
		--> This solution should be reused for the pre- $.support checks too but didn't make it for 1.5.1
		--> REVERTED FOUC in Opera confirmed by multiple peopl
	- #8107 Fix argument handling for $.ajax for multiple method signatues and add test case
	- #8108 Remove event.js code that was doing this before specifically for events, which is now redundant. Fixes #8108. 1.5-stable
		--> FIX PARTIALLY REVERTED
		--> BLOCKER FOR 1.5.1
		--> snover, dave
		--> Fixed now by snover. The trick to fix the JSON stringify issue shouldn't be mentioned.
	- #8115 Renames all references to jXHR with jqXHR in the code (like was done in the doc)
		--> LOL
	- #8123 The default for .clone() is to not clone any events
		--> The nasty clone regression bug (big issue for jQuery UI)
	- #8125 Status is set to 200 for requests with status 0 when location.protocol if "file:". Added test/localfile.html to control it works
	- #8129 Fix cloning multiple selected options in IE8
	- #8135 Makes sure any exception thrown by Firefox when trying to access an XMLHttpRequest property when a network error occured is caught and notified as an error. Added test/networkerror.html to test the behavior
	- #8138 Access to document.location is made only once at load time and if it fails (throwing an exception in IE when document.domain is already set), we use the href of an A element instead
	- #8145 Add readyWait tests
	- #8146 introducing the xhrFields option with is a map of fieldName/fieldValue to set on the native xhr. Can be used to set withCredentials to true for cross-domain requests if needed
		--> NEEDSDOCS: new xhrFields option
	- #8152 applying the same special cases for protocol "chrome-extension:" as were for "file:" (needs tests). Re-organizes and fixes the handling of special cases for HTTP status code in the xhr transport
		--> NEEDSDOCS (?) @jaubourg new isLocal option
		--> NEEDSDOCS (?) @jaubourg this + some other commits special case *-extensions and widget urls
	- #8177 XHR transport now considers 304 Not Modified responses as 200 OK if no conditional request header was provided (as per the XMLHttpRequest specification)
	- #8193 Fixes abort in prefilter. No global event will be fired in that case even if the global option is set to true. Unit test added
		--> NEEDSDOCS
	- #8198 Remove unnecessary "script.type = text/javascript;"
	- #8200 Unexpose $.support._scriptEval as it's not needed. Use a private var instead
	- #8209 Make sure that mousing over Chrome "internal div" doesn't trigger a mouseleave
	- #8219 Introduces the mimeType option to override content-type header in conversion (and in native xhr when possible). Adds companion overrideMimeType method on jqXHR object (it simply sets the option)
		--> NEEDSDOCS new mimeType option and overrideMimeType function on jqXHR
	- #8220 Remove backslashes from tag name filter
	- #8245 Ajax now ensures header names are capitalized so that non-compliant xhr implementations don't override them
	- #8250 ajax does not work in opera 10 widgets
	- #8277 Default data, setup via ajaxSetup is not send in ajax utility functions (.get/.post/.load/.getJSON/.getScript)
	- #8279 Makes sure response headers with empty values are handled properly and do not prevent proper parsing of the entire response headers string

	- build-script fix: Remove sed from post-build code, due to portability issues between GNU and BSD versions

Bug fixes by (github names)
	antonkovalyov, csnover, danheberden, davidmurdoch, dmethvin, gnarf37, jaubourg, jeresig, jitter, jrburke, lrbabe, mathiasbynens, rwldrn, SlexAxton, voxwerk

and SlexAxton!