286388 - cursor:url() overwrites child elements cursor value (cascade doesn't work)

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect, P1)

Description

[Peter van der Woude [:Peter6]]

Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050315
Firefox/1.0+

I know the summary is wrong but I don't know the proper wording.

repro:
1.Open testcase
2.Case 1 shows normal behaviour, the cursor in the red box is pointer, in the
green box is wait
3.Case 2 shows wrong behaviour, the cursor in the red box is url(),pointer , in
the green box should be wait but url(),pointer is displayed

expected:
the cursor url() to behave the same as default cursors

( bug 38447 cursor:url() implementation )

testcase follows

Comment 1

[Peter van der Woude [:Peter6]]

Attachment: cursor for testcase

Comment 2

[Peter van der Woude [:Peter6]]

Attachment: testcase

Comment 3

[Ben Basson]

This should probably be marked as a regression, otherwise I think (contrary to
your suggestion on the forum) that your summary is ok. 

Just in case anyone missed it though... cursor:url() properties are being
inherited by child elements, even if they have an overriding value set. This
should not be the case.

Comment 4

[Christian :Biesinger (don't email me, ping me on IRC)]

(In reply to comment #3)
> This should probably be marked as a regression

why? this never worked correctly.

Comment 5

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: patch (untested)

This will probably fix it, but the url() support doesn't work for me (GTK2), at
least based on the testcase in this bug, so I can't tell for sure.  Could
somebody test?

Comment 6

[Christian :Biesinger (don't email me, ping me on IRC)]

yep, the patch does seem to work.

it seems this line, a bit down, can be removed then:
       if (ui->mCursorArray.Count() == 0) {

Comment 7

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: patch

Comment 8

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

I was suspicious of that code when I reviewed it, and I should have thought
about it more carefully.  It's also completely broken in the non-null aStartData
with 'inherit' case.

Comment 9

[Christian :Biesinger (don't email me, ping me on IRC)]

sorry about this bug :( thanks for fixing it!

Comment 10

[Sergey Sokoloff]

I hereby request for blocking-aviary1.1 and blocking1.8b2 flags, the fix seems
important.

Comment 11

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 177684 [details] [diff] [review]
patch

r+sr=bzbarsky

Comment 12

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Fix checked in to trunk, 2005-03-16 20:38 -0800.

Comment 13

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

FWIW, it's not working for me because I hit this early return in
widget/src/gtk2/nsWindow.cpp's SetCursor(imgIContainer*):

    nsCOMPtr<nsIGdkPixbufImage> pixImg(do_GetInterface(frame));
    if (!pixImg)
        return NS_ERROR_NOT_AVAILABLE;

Comment 14

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Please respond to that comment in bug 38447 rather than here.

Comment 15

[Peter van der Woude [:Peter6]]

->verified

thanks for the quick fix