749118 - CSS opacity regression

Status: RESOLVED FIXED

(Core :: Graphics, defect)

Description

[Camaleon]

Attachment: The animated GIF is rendered as it should

There seems to be a regression with CSS opacity in Firefox 12 (or at least it was not present in earlier versions). I am attaching 3 files:

- sample_regression.html (with an opacity level set that exposes the problem)
- sample.html (opacity level set to "1" and working as it should)
- sample.jpg (background image)
- dots.gif (animated gif)

What happens is that when setting the transparency to a layer, the second frame of the animated GIF inside the layer is lost.

Comment 1

[Camaleon]

Attachment: The animated GIF losts the second frame

Comment 2

[Camaleon]

Attachment: A sample background image

Comment 3

[Camaleon]

Attachment: A sample GIF animated image

Comment 4

[Alice0775 White]

Regression window(m-c)
Cannot reproduce:
http://hg.mozilla.org/mozilla-central/rev/e1ae143cc842
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0a1) Gecko/20111026 Firefox/10.0a1 ID:20111026120007
Can reproduce:
http://hg.mozilla.org/mozilla-central/rev/16a8d2ab5240
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0a1) Gecko/20111026 Firefox/10.0a1 ID:20111026163203
pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=e1ae143cc842&tochange=16a8d2ab5240

Regression window(m-i)
Cannot reproduce:
http://hg.mozilla.org/integration/mozilla-inbound/rev/98013fe19dcb
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0a1) Gecko/20111025 Firefox/10.0a1 ID:20111025195747
Can reproduce:
http://hg.mozilla.org/integration/mozilla-inbound/rev/d7f3bfc7cd46
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0a1) Gecko/20111025 Firefox/10.0a1 ID:20111025203245
pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=98013fe19dcb&tochange=d7f3bfc7cd46

Last good; 68d6ef3d84a3
First bad; 8b89d7037306
Triggered by;
8b89d7037306	Matt Woodrow — Bug 695275 - Fix conversion of ThebesLayers to ImageLayers. r=roc

Comment 5

[Matt Woodrow (:mattwoodrow)]

Attachment: Translate the source correctly

Comment 6

[Alex Keybl [:akeybl]]

This is not a recent regression. Please nominate for Aurora/Beta approval if you'd like to get this in earlier than FF15.

Comment 7

[Joe Drew (not getting mail)]

Comment on attachment 618856 [details] [diff] [review]
Translate the source correctly

Review of attachment 618856 [details] [diff] [review]:
-----------------------------------------------------------------

::: image/src/RasterImage.cpp
@@ +852,5 @@
>                                                               gfxASurface::ImageFormatARGB32);
>    gfxContext ctx(imgsurface);
>    ctx.SetOperator(gfxContext::OPERATOR_SOURCE);
> +  ctx.Rectangle(framerect);
> +  ctx.Translate(framerect.TopLeft());

A better fix to this would be ctx.SetPattern(pattern, framerect.TopLeft());

Comment 8

[Matt Woodrow (:mattwoodrow)]

That API doesn't exist :)

We could adjust the pattern matrix instead? But that doesn't sound easier than my fix, and we'd need to account for a possible existing pattern matrix.

Comment 9

[Joe Drew (not getting mail)]

Comment on attachment 618856 [details] [diff] [review]
Translate the source correctly

Review of attachment 618856 [details] [diff] [review]:
-----------------------------------------------------------------

Oh, I was thinking of SetSource. Sorry!

Comment 10

[Matt Woodrow (:mattwoodrow)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/15d1960c419e

Comment 11

[Matt Brubeck (:mbrubeck)]

https://hg.mozilla.org/mozilla-central/rev/15d1960c419e

Comment 12

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Comment on attachment 618856 [details] [diff] [review]
Translate the source correctly

Review of attachment 618856 [details] [diff] [review]:
-----------------------------------------------------------------

Let's fix this Web-facing regression ASAP

Comment 13

[Alex Keybl [:akeybl]]

(In reply to Robert O'Callahan (:roc) (Mozilla Corporation) from comment #12)
> Let's fix this Web-facing regression ASAP

What's the risk associated with this patch for a FF10 regression?

Comment 14

[Matt Woodrow (:mattwoodrow)]

(In reply to Alex Keybl [:akeybl] from comment #13)
> What's the risk associated with this patch for a FF10 regression?

Should be very low risk, it's just changing the positioning of the image.

Comment 15

[Lukas Blakk [:lsblakk] use ?needinfo]

Comment on attachment 618856 [details] [diff] [review]
Translate the source correctly

sounds low risk enough, approving

Comment 16

[Robert O'Callahan (:roc) (email my personal email if necessary)]

I will land this.

Comment 17

[Robert O'Callahan (:roc) (email my personal email if necessary)]

https://hg.mozilla.org/releases/mozilla-aurora/rev/03c10fdb4fd0
https://hg.mozilla.org/releases/mozilla-beta/rev/4ed154f44f8d

Comment 18

[Virgil Dicu [:virgil] [QA]]

Mozilla/5.0 (X11; Linux i686; rv:13.0) Gecko/20100101 Firefox/13.0

Verified in Ubuntu 12.04, Mac OS 10.6 and Windows 7 on Firefox 13 beta 6. Used attachment in comment 0.

Comment 19

[Virgil Dicu [:virgil] [QA]]

Mozilla/5.0 (X11; Linux i686; rv:14.0) Gecko/20100101 Firefox/14.0

Verified in Ubuntu 12.04, Mac OS 10.8 and Windows 7 on Firefox 14 beta 7. Used attachment in comment 0.