http://jsfiddle.net/fa98L/

Still an issue in 1.4.3.

Retargeting due to severity of regressions in 1.4.3.

Retargeting for 1.4.4 as per John’s request.

I'm a bit confused as to why this is a bug. I think IE is correct with this ( for once ). Unless you meant td and not tr.

tr's doesn't have the colspan attribute

td's always has a colspan value, even if you don't set it (defaults to 1)

Which might be why you were gettings those results.

I did some alternative code to getting a tr where the child td's don't have a colspan (or in my case, have a colspan > 1) here (tested in FF, Chrome, Safari and IE 6,7,8,9)

I made a misprint in a header:) But:

$('tr').not(':has([colspan])')

Such code means: "Get tr which don't have td with colspan attribute".

It's a bug because it works fine in not IE browsers and 'has' can accept a selector in jQuery 1.4.* and '[colspan]' is a CSS 2.1 selector.

Replying to [comment:7 pinal]:

I made a misprint in a header:) But: $('tr').not(':has([colspan])') Such code means: "Get tr which don't have td with colspan attribute". It's a bug because it works fine in not IE browsers and 'has' can accept a selector in jQuery 1.4.* and '[colspan]' is a CSS 2.1 selector.

Yeah that what I thought, I've tried:

$('tr').not('td:has([colspan])') //doesn't work at all

And

//doesn't work in IE 6, 7 and 8 ( Works In non IE browsers and IE9)
$('tr').not('td').has('[colspan]') 

Example here:

Maybe because colspan defaults to 1 if it is not set IE says it is anyway.

I want to get such table rows by selector. I think $('tr').not(':has([colspan])') is right in selectoin syntax. May be this bug is duplicate of another one, but such selectoion must work.

Replying to [comment:9 pinal]:

I want to get such table rows by selector. I think $('tr').not(':has([colspan])') is right in selectoin syntax. May be this bug is duplicate of another one, but such selectoion must work.

Yes I agree. Although to me

$('tr').not(':has([colspan])') 

should be

$('tr').not('td:has([colspan])') 

to be syntactally correct, I may be wrong.

Both the examples I've gave should really work, as they are the same technically, but only one does.

I just was struck by this problem in a project of mine, and created a small fiddle of it, and afterwards searched the bugtracker and found this. Bad choice =)

Anyway, I created the fiddle so I give it to you:

http://www.jsfiddle.net/wailqill/jW5CZ/

Works in ff, not in ie. No matter what version of jQuery.

Retargeting to next minor release.

IE appears to be behaving differently in jsFiddle vs. true operation.

Here it thinks all td elements have a colspan, it even picks them up in CSS:

http://jsfiddle.net/MrTyzik/DSVeg/4/

However outside jsFiddle things match Chrome for me in IE9 beta:

http://csh.rit.edu/~ryanw/chaos/ie9-colspan.html

I haven't tried true IE8 or below yet.

Please note the colspan_not_one test at the end though using $('td[colspan!=1]'). In this case, IE does not match FF or Chrome. That feels like a related but different issue.

This is getting moved to 1.6 for when the $.attr happens.

Nominating ticket for 1.7 discussion.

+1, Seems like a bug, should be fixed

+0,

+1, Sizzle attributes again

+1, Buggers

+1, Probably due to attribute camelCase, timmywil and I can fix this.

+1, also, .hasAttr() might be nice

+1

Duplicate of #5637.