createDocumentFragment is not a function.

[anonymous]

Two variants of the $() function are getting confused and the older one is getting used when the newer one is wanted.

Code to trigger:

$('<input type="text" />', {
    name: 'test'
});

Note that I use that format specifically because of this text in the documentation for that function: "Note that Internet Explorer will not allow you to create an input element and change its type; you must specify the type using '<input type="checkbox" />' for example."

Known work arounds:

$('<input/>', {
   name: 'test'
});

However that would still run afoul the IE issue. This is what I ended up using instead:

$('<input type="text" />').attr({
    name: 'test'
});

This was tested with the minified version of jQuery 1.4.2 and the development version of 1.4.3 both on Firefox 3.6.3 and Chromium 5.0.375.127 (0).

[jitter]

jQuery( html, props ) isn't meant to be used in the way you do in your "trigger code".

The api docs for  jQuery( html, props ) state

html A string defining a single, standalone, HTML element (e.g. <div/> or <div></div>).

Passing in as html anything other (complex html, not single standalone tag e.g as above with "inline" attributes) while the second parameter (as in this case) isn't a context but a map of attributes will trigger the exception you mentioned.

Doing

$('<input />', {
    type: 'text',
    name: 'test'
});

fails in IE as stated by the docs. So the only valid way to do this is for input / button elements in IE is exactly what you ended up doing.

Check also my comment on this  jQuery-UI commit which caused similar issues (the commit not my comment :D ).

[SlexAxton]

I think perhaps the docs could be cleared up on this a bit, but it's not something that needs to change in jQuery. The idea is that the docs would let you know that your use-case is invalid in IE. I think programatically it doesn't make sense to try and sniff out. The solution you came up with is the ideal use.

[anonymous]

No arguments about the wontfix codeside, but could the docs get an update soon please. I had the same misunderstanding (and as an aside, the  pre 1.4 behaviour didn't break the same way - it still didn't work properly, but depending on the args, it kind of looks like it does)

[addyosmani]

Updated:  http://api.jquery.com/jQuery/

[rwaldron]

#8950 is a duplicate