812107 - Graphical glitch in print preview when scroll

Status: VERIFIED FIXED

(Core :: Print Preview, defect)

Description

[Alice0775 White]

Attachment: screen shot

Build Identifier:
http://hg.mozilla.org/mozilla-central/rev/87928cd21b40
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121114102936

This is very similar Bug 794709, but this is recent regression.
This problem only happens in print preview.

Steps to reproduce:
 1. Open long page such as https://developer.mozilla.org/en-US/docs/Firefox_4_for_developers
 2. Print Preview (Alt > File > Print Preview)
 3. Scroll Up/Down

Actual results:
 Gray area appears at bottom of screen
 Text disappears from bottom of screen
 As page scrolls down, the range spreads out

Expected results:
 No such problem

Regression window(m-c)
Good:
http://hg.mozilla.org/mozilla-central/rev/fc1684f4d3a9
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121114044850
Bad:
http://hg.mozilla.org/mozilla-central/rev/87928cd21b40
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121114102936
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=fc1684f4d3a9&tochange=87928cd21b40

Regression window(m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/2b66b88a0b0f
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121113221949
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/24c0c55af0ed
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121113222650
Pushlog;
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=2b66b88a0b0f&tochange=24c0c55af0ed

Suspected: Bug 811570

Comment 1

[Alice0775 White]

Screen capture http://youtu.be/3230eG-jdns

Comment 2

[Alice0775 White]

This problem happesn in Linux too. However, the glitch begins from the upper part of screen.

http://hg.mozilla.org/mozilla-central/rev/87928cd21b40
Mozilla/5.0 (X11; Linux i686; rv:19.0) Gecko/19 Firefox/19.0a1 ID:20121114102936

Comment 3

[Alice0775 White]

Triggered by:
1cfa8c4279e5	Matt Woodrow — Bug 810186 - Don't unnecessarily nest inactive layers. r=roc

Comment 7

[Tony Mechelynck [:tonymec]]

I see this on Linux64, both on Firefox Nightly 20.0a1 and on SeaMonkey 2.17a1 but not on Firefox 17.0esrpre. See Build IDs and changeset IDs in bug 817214 comment #0

Comment 8

[Matt Woodrow (:mattwoodrow)]

Attachment: Don't attempt to do scroll compensation for nsDisplayTransform

In this case we had two nested transforms (nsSimplePageSequence and nsPageFrame). The outer one was setting a scale.

The scroll offset code wasn't adjusting for the scale on the container and content within it ends up in the wrong place. This doesn't happen for content, because transformed frames are normally the reference frame, so we don't need scroll compensation.

I think it's easiest to just not bother here. Maybe making the transforms created for print preview reference frames would be worth it, since we'll probably be invalidating more than we need to now.

Comment 9

[Matt Woodrow (:mattwoodrow)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/54f08a854e4e

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/54f08a854e4e

Comment 11

[Tony Mechelynck [:tonymec]]

Mozilla/5.0 (X11; Linux x86_64; rv:20.0) Gecko/20121204 Firefox/20.0 ID:20121204174201 CSet: 1942b4d64dc8

Mozilla/5.0 (X11; Linux x86_64; rv:20.0) Gecko/20100101 Firefox/20.0 SeaMonkey/2.17a1 ID:20121204121822 c-c:1c212e41559a m-c:e8f0504ccbb9

Neither of these hourlies exhibit the bug.
I cannot test on Mac or Windows (I only have this Linux machine).

Please test Windows builds of Fx20 and Sm2.17a1 and if OK, set VERIFIED and verified-Fx20. Then it will be time to evaluate the need (or not) for approval-mozilla-aurora.

Comment 12

[Alice0775 White]

I cannot reproduce the problem anymore in the following Firefox&SeaMonkey win32 build.

http://hg.mozilla.org/mozilla-central/rev/1942b4d64dc8
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:20.0) Gecko/20121204 Firefox/20.0 ID:20121204174201

http://hg.mozilla.org/comm-central/rev/1c212e41559a
http://hg.mozilla.org/mozilla-central/rev/1942b4d64dc8
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:20.0) Gecko/20100101 Firefox/20.0 SeaMonkey/2.17a1 ID:20121204174222

Comment 13

[Tony Mechelynck [:tonymec]]

Setting VERIFIED on the basis of comment #11 and 12.

If you (anyone) still see the bug in a nightly or hourly containing mozilla-central changeset 54f08a854e4e please post a comment with the details.

Comment 15

[Lukas Blakk [:lsblakk] use ?needinfo]

Since this is verified, please nominate for Aurora uplift.

Comment 16

[Matt Woodrow (:mattwoodrow)]

Comment on attachment 688052 [details] [diff] [review]
Don't attempt to do scroll compensation for nsDisplayTransform

[Approval Request Comment]
Bug caused by (feature/regressing bug #): Bug 811570
User impact if declined: Broken scrolling in print preview.
Testing completed (on m-c, etc.): Been on m-c for a while
Risk to taking this patch (and alternatives if risky): Low risk, reverts back to previous behaviour in some cases.
String or UUID changes made by this patch: None

Comment 17

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/31a8f85db8c4

Comment 18

[Ada [:adalucinet]]

Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/20100101 Firefox/19.0
Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:19.0) Gecko/20100101 Firefox/19.0
Mozilla/5.0 (X11; Linux i686; rv:19.0) Gecko/20100101 Firefox/19.0


Works for me on Firefox 19 beta 5 (Build ID: 20130206083616) following STR from comment 0.

Comment 19

[u123541]

Just discovered this bug on 23.0a1 (2013-04-13)
Should I re-open this bug or create a new one? 
(Didn't find page containing reporting rules; if any...)

Comment 20

[Olli Pettay [:smaug][[e-mail]]]

Open a new one, please.

Comment 21

[Tony Mechelynck [:tonymec]]

(In reply to Pierre Fortin from comment #19)
> Just discovered this bug on 23.0a1 (2013-04-13)
> Should I re-open this bug or create a new one? 
> (Didn't find page containing reporting rules; if any...)

I think you should report a new bug, with the "regression" keyword, and mention in its Description that this bug seems to have happened again. You could also add Matt Woodrow to its CC.

Bug reporting guidelines start at https://developer.mozilla.org/en-US/docs/Mozilla/QA/Bug_writing_guidelines and continue on pages linked (recursively) from there.

Comment 22

[Alice0775 White]

(In reply to Pierre Fortin from comment #19)
> Just discovered this bug on 23.0a1 (2013-04-13)
> Should I re-open this bug or create a new one? 
> (Didn't find page containing reporting rules; if any...)

maybe Bug 860253