Side navigation
#12856 closed bug (fixed)
Opened November 06, 2012 02:14PM UTC
Closed December 04, 2012 04:44AM UTC
Last modified December 04, 2012 07:33PM UTC
Syntax error, unrecognized expression in jquery 1.8+
| Reported by: | pedro.rainho@gmail.com | Owned by: | timmywil |
|---|---|---|---|
| Priority: | low | Milestone: | 1.9 |
| Component: | selector | Version: | 1.8.0 |
| Keywords: | Cc: | ||
| Blocked by: | Blocking: |
Description
This is an error that happen in ipad and when I run this selector
$headers.filter(':not([colspan])[data-column="' + i + '"]:last,[colspan="1"][data-column="' + i + '"]:last')
I have the error
Syntax error, unrecognized expression :not([colspan])[data-column="0"]:last,[colspan="1"][data-column="0"]:last
I've tested with Jquery 1.8.2: http://jsfiddle.net/W8Mm3/2/
I've tested with Jquery 1.8.1: http://jsfiddle.net/W8Mm3/
I've tested with Jquery 1.8.0: http://jsfiddle.net/W8Mm3/3
Both have the same error:
I've tested with Jquery 1.7: http://jsfiddle.net/W8Mm3/1/
I've tested with Jquery 1.7.1: http://jsfiddle.net/W8Mm3/4/
I've tested with Jquery 1.7.2: http://jsfiddle.net/W8Mm3/5/
And I don't have that error
So the error was introduced in jquery 1.8.0
Attachments (0)
Change History (12)
Changed November 06, 2012 03:06PM UTC by comment:1
| component: | unfiled → selector |
|---|---|
| resolution: | → worksforme |
| status: | new → closed |
Changed November 06, 2012 03:26PM UTC by comment:2
Looks like it's this bug, not sure if it's been fixed or not: https://github.com/Mottie/tablesorter/issues/132
Also it's not clear to me why it would only affect Safari?? I don't have an iPad handy but I can't repro on desktop Safari 5.1.
Changed November 06, 2012 04:48PM UTC by comment:3
I've wrote "This is an error that happen in ipad"
It doesn't happen in pc browsers.
Also I don't know why someone close the bug and didn't bother to test it well.
Mr. timmywil using you demo: http://jsfiddle.net/timmywil/W8Mm3/21/
The error HAPPEN in IPAD 1 - version IOS 5
Did you even test it?
Changed November 06, 2012 04:50PM UTC by comment:4
Replying to [comment:1 timmywil]:
There's too much going on in the tests. When reduced, I am unable to reproduce: http://jsfiddle.net/timmywil/W8Mm3/21/
Did you even test you demo in ipad? I've test it and the error happen. Please re-open the bug.
Changed November 06, 2012 09:41PM UTC by comment:5
| milestone: | None → 1.8.3 |
|---|---|
| priority: | undecided → high |
| resolution: | worksforme |
| status: | closed → reopened |
Sorry, I didn't see the "in ipad". It does seem ridiculous that this would only be a problem in Safari.
Changed November 06, 2012 09:47PM UTC by comment:6
| priority: | high → low |
|---|---|
| status: | reopened → open |
_mobile_ Safari even. However, I don't see the issue in iOS 6.0 (http://cl.ly/image/0Z2e3t0i3b3z). My guess is that there was a bug in mobile Safari.
Changed November 07, 2012 09:06AM UTC by comment:7
This problem doesn't happen on pc safari and I can't see if it happen in ios 6 since I don't have an ipad2+. I do know that happen in ios 5 (ipad1).
Yesterday me and the author of tablesorter we manage to split the selector in parts
from this:
h = $headers.filter(':not([colspan])[data-column="' + i + '"]:last,[colspan="1"][data-column="' + i + '"]:last')
to this:
h = $headers.filter(':not([colspan])');
h = h
.add( $headers.filter('[colspan="1"]') ) // ie8 fix
.filter('[data-column="' + i + '"]:last');
And for now it's working. I start doing this because I also had other errors with long selectors and the solution was divide the selector in parts.
Changed November 07, 2012 11:55AM UTC by comment:8
I've tested with Ipad 2 and this doesn't happen. So it's a bug in ipad 1 IOS 5
Changed November 11, 2012 08:34PM UTC by comment:9
| milestone: | 1.8.3 → 1.9 |
|---|
Changed November 19, 2012 05:29PM UTC by comment:10
| owner: | → timmywil |
|---|---|
| status: | open → assigned |
Changed December 04, 2012 04:44AM UTC by comment:11
| resolution: | → fixed |
|---|---|
| status: | assigned → closed |
Fix #12856: keep PSEUDO regex non-greedy
Changeset: 3ab26340dcdf001da4e93df2f187a8343615673c
Changed December 04, 2012 07:33PM UTC by comment:12
Can we fix it so that the bot gives us a real link to the changeset?
There's too much going on in the tests. When reduced, I am unable to reproduce:
http://jsfiddle.net/timmywil/W8Mm3/21/