2052162 - Regression since 150b1: pdf printing with blank pages / wrong page breaks

Status: RESOLVED FIXED

(Core :: Printing: Output, defect)

Description

[Maxim Britov]

Attachment: fff152.png

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:152.0) Gecko/20100101 Firefox/152.0

Steps to reproduce:

Opening electro document in edi provider.
Open preview and print:
FF 149 - 12 pages in preview
FF150b1..153b7 - 122 pages

Comment 1

[Maxim Britov]

Attachment: fff152 another.png

Comment 2

[Maxim Britov]

Attachment: fff149.png

Comment 3

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

The Bugbug bot thinks this bug should belong to the 'Core::Printing: Output' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 4

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

The severity field is not set for this bug.
:jwatt, could you have a look please?

For more information, please visit BugBot documentation.

Comment 5

[Daniel Holbert [:dholbert]]

Thanks for the bug report! Unfortunately it looks like the affected URL is not publicly viewable. Is there any chance you can generate a testcase that reproduces the bug? The "SingleFile" extension might be useful as a way of turning the page that you're viewing into a testcase, if that page is appropriate for sharing (not sure) -- you can get that here https://addons.mozilla.org/en-US/android/addon/single-file/

If there's private data included in the testcase that's not suitable for posting publicly as an attachment on this bug, you're also welcome to send a testcase directly via email to me at [e-mail] (if you're comfortable doing so) and I can try to construct a reduced testcase with all the content removed, and then I'll delete the email.

I also noticed you're using 16 pages-per-sheet -- I'm guessing that's just to make the bug easier to see in screenshots, right? (Or, does the bug specifically only happen when you're printing multiple pages per sheet?)

I have a couple of other ideas for threads to investigate (notably mozregression which might be a way to identify the change that caused the issue), but let's start with this set of questions for now. :) Thanks!

Comment 6

[Maxim Britov]

(In reply to Daniel Holbert [:dholbert] from comment #5)

If there's private data included in the testcase that's not suitable for posting publicly as an attachment on this bug, you're also welcome to send a testcase directly via email to me at [e-mail] (if you're comfortable doing so) and I can try to construct a reduced testcase with all the content removed, and then I'll delete the email.

Sent video as private email

Comment 7

[Daniel Holbert [:dholbert]]

Thanks, Maxim!

Maxim managed to construct a standalone testcase, shared privately -- and with that, I was able to bisect with mozregression and determine that this was a regression from this commit:
https://hg-edge.mozilla.org/integration/autoland/rev/1e396370701364d07e54fb4877df04835526d947 ("Bug 343048 - Get rid of ColGroup list")

I'll try to construct a shareable reduced testcase from Maxim's testcase later on today.

Comment 8

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

Set release status flags based on info from the regressing bug 343048

Comment 9

[Daniel Holbert [:dholbert]]

Attachment: testcase 1 (reduced)

Here's a reduced testcase.

In Chrome and pre-regression Firefox, this gives me 6 pages in print-preview.
In post-regression Firefox, this gives me 78 pages (most of the pages being blank).

One other notable thing - this seems to be dependent on the fragmentation fallback codepath -- if I turn that off by toggling about:config pref layout.display-list.improve-fragmentation to false, then the issue goes away.

Comment 10

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

:emilio, since you are the author of the regressor, bug 343048, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 11

[Emilio Cobos Álvarez [:emilio]]

table { overflow: hidden } works around the issue, indeed.

Comment 12

[Emilio Cobos Álvarez [:emilio]]

Attachment: Bug 2052162 - Fix nsTableFrame overflow handling. r=dholbert,#layout

Only rowgroups should be considered. Also, there's a separate issue
where the Reflow() loop might've pushed some of the frames in mRowGroups
to the next continuation, but we'd still consider overflow for them.

Comment 13

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/4952382c53f9
https://hg.mozilla.org/integration/autoland/rev/867e285bfe89
Fix nsTableFrame overflow handling. r=layout-reviewers,dshin

Comment 14

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/61380 for changes under testing/web-platform/tests

Comment 15

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/9f6e48768c15
https://hg.mozilla.org/integration/autoland/rev/d66ec39a5507
Update wpt metadata in table-in-flex-overflow-print.html

Comment 16

[Daniel Holbert [:dholbert]]

Test timeout (annotated in comment 15) is odd, but it might "just" be an instance of bug 1886588.

Comment 17

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/00085474a6bc
https://hg.mozilla.org/integration/autoland/rev/a3db9a3700d1
Typo error.

Comment 18

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/34a8e3e23ece
https://hg.mozilla.org/integration/autoland/rev/0fcf4927761d
Revert "Bug 2052162 - Typo error." for causing wpt failures in table-in-flex-overflow-print.html.

Comment 19

[Serban Stanca [:SerbanS]]

Reverted this because it was causing wpt failures in table-in-flex-overflow-print.html.

• Revert link
• Push with failures
• Failure Log
• Failure line: TEST-UNEXPECTED-TIMEOUT | /css/printing/table-in-flex-overflow-print.html | expected PASS

Comment 20

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR was closed without merging

Comment 21

[Emilio Cobos Álvarez [:emilio]]

(In reply to Daniel Holbert [:dholbert] from comment #16)

Test timeout (annotated in comment 15) is odd, but it might "just" be an instance of bug 1886588.

Yes it is, but it got annotated in the wrong folder (testing/web-platform/tests rather than ../meta

Comment 22

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/75256767ec00
https://hg.mozilla.org/integration/autoland/rev/09e41de24fb3
Fix nsTableFrame overflow handling. r=layout-reviewers,dshin

Comment 23

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/7540cc7881cd
https://hg.mozilla.org/integration/autoland/rev/1d57075f3858
metadata update

Comment 24

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/5592282568e9
https://hg.mozilla.org/integration/autoland/rev/604ea731b733
metadata update

Comment 25

[Cristian Tuns]

https://hg.mozilla.org/mozilla-central/rev/09e41de24fb3
https://hg.mozilla.org/mozilla-central/rev/1d57075f3858
https://hg.mozilla.org/mozilla-central/rev/604ea731b733

Comment 26

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot