1110686 - GIF with no frame delay renders a white pixel

Status: NEW

(Core :: Graphics: ImageLib, defect)

Description

[Lars Dɪᴇᴄᴋᴏᴡ]

Attachment: GIF with no frame delay

To reproduce:

Display attached image with Firefox 34.0.5. Since the browser defaults to rescale-to-window, click once to set image scaling to 1:1. Observe the colour of the pixel at coordinates 1313,892 relative to the top-left corner of the image.

Actual result:

Colour is white.

Expected result:

Colour should be a deep blue.


Further information:

This bug does not occur with Opera 12.16, Chromium 39.0.2171.65, gifview (from gifsicle 1.87).

When the GIF frame delay is changed from none to 1 centisecond each, Firefox renders the pixel correctly.

Comment 1

[Lars Dɪᴇᴄᴋᴏᴡ]

Attachment: GIF with 1 centisecond frame delay each

File contains the same encoded image data as above, just the frame delay is different.

Comment 2

[Alice0775 White]

It seems to be same between Firefox34 and IE11 on windows7x64.

Comment 3

[Glenn Randers-Pehrson]

I'm seeing the reported behavior using a recent build from hg tip (Firefox37.0a1 changeset 219326), on Ubuntu.

Comment 4

[Milan Sreckovic [:milan] (needinfo for best results)]

A recent regression?

Comment 5

[Seth Fowler [:seth] [:s2h]]

Fixed in bug 1126330. We now play back the 0-frame-delay GIF just like Blink.

Comment 6

[Lars Dɪᴇᴄᴋᴏᴡ]

Still broken in Firefox 44. Please retest.

Comment 7

[u279076]

(In reply to Lars Dɪᴇᴄᴋᴏᴡ from comment #6)
> Still broken in Firefox 44. Please retest.

This looks fine on my end. Can you please provide more details?

Comment 8

[Lars Dɪᴇᴄᴋᴏᴡ]

Attachment: bug1110686.png

I don't know, what more details do you need?

Screenshot of the rendering defect highlighted with a pink circle attached.

Comment 9

[u279076]

(In reply to Lars Dɪᴇᴄᴋᴏᴡ from comment #8)
> Created attachment 8717731 [details]
> bug1110686.png
> Screenshot of the rendering defect highlighted with a pink circle attached.

Thanks, I can see the issue now on the GIF with no frame delay. I'm updating the summary to more accurately describe the issue. I'll see if I can track down if this was ever fixed and if so, what fixed it and what regressed it.

Comment 10

[u279076]

Okay, so I've tested and reproduced the issue Firefox 44.0, 43.0, 42.0 and 41.0. I think it's safe to say that this issue was never fixed.

Seth, do you have any ideas as to what could be happening here?

Comment 11

[BugBot [:suhaib / :marco/ :calixte]]

Clear a needinfo that is pending on an inactive user.

Inactive users most likely will not respond; if the missing information is essential and cannot be collected another way, the bug maybe should be closed as INCOMPLETE.

For more information, please visit auto_nag documentation.