563436 - Content is progressively compressed to left in print/print-preview, with margin-right & float:left

Status: RESOLVED FIXED

(Core :: Printing: Output, defect)

Description

[Daniel Holbert [:dholbert]]

STEPS TO REPRODUCE:
 1. Print-preview attached testcase.

ACTUAL RESULTS:
 * 1st page looks fine.
 * 2nd page has much skinnier column of content than 1st page.
 * 3rd page & beyond have content compressed to only allow 1 char per line.

EXPECTED RESULTS:
2nd page and beyond should have same column-width as 1st page. (or at least, no skinnier of a column)

Comment 1

[Daniel Holbert [:dholbert]]

Attachment: testcase 1

Comment 2

[Daniel Holbert [:dholbert]]

Attachment: PDF printout of testcase 1

Comment 3

[Daniel Holbert [:dholbert]]

This bug is visible in mozilla-central nightlies, as well as Firefox 3.6:
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.3a5pre) Gecko/20100503 Minefield/3.7a5pre
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.5pre) Gecko/20100503 Namoroka/3.6.5pre

It doesn't affect Firefox 3.5, however:
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.8) Gecko/20100202 Firefox/3.5.8

Comment 5

[Daniel Holbert [:dholbert]]

Adding the URL that I reduced the testcase from.

There also might already be an open bug on this, with a testcase... I had some pangs of deja vu when reducing this.

Comment 6

[Matt]

I can reproduce on windows so marking OS to "All".  However, the number pattern I get on page 2 is different than what you show in the PDF.


Regression range:

works:
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2a1pre) Gecko/20090713 Minefield/3.6a1pre
http://hg.mozilla.org/mozilla-central/rev/b9012a18d2c6

broken:
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2a1pre) Gecko/20090714 Minefield/3.6a1pre
http://hg.mozilla.org/mozilla-central/rev/9b4dd06e1c9d

http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=b9012a18d2c6&tochange=9b4dd06e1c9d

Comment 7

[Daniel Holbert [:dholbert]]

Thanks Matt!  Based on that regression range, this looks like a regression from bug 503183.

Comment 8

[Matt]

Attachment: Windows image crop

An image to show the different number pattern I saw compared to the original PDF.

Comment 9

[Daniel Holbert [:dholbert]]

(In reply to comment #6)
> However, the number pattern
> I get on page 2 is different than what you show in the PDF.

That Windows vs Linux difference is expected -- it could be caused by width differences in the (platform-specific) fonts, and/or differences in default page-margin widths.

Comment 10

[Adrian Florinescu [:aflorinescu]]

Issues still reproducible on the latest Nightly 49.0a1 Build ID 20160505030327;

Comment 11

[Ale]

Bug still present on the latest dev edition 59.0b10 build 20180215111455.

Comment 12

[Ale]

This page (https://biorob2.epfl.ch/pages/cabelguen/) shows the same problem, but is it the same bug or something else? There is no "float" style on the text.

Comment 13

[Ting-Yu Lin [:TYLin] (PDT, UTC-7)]

Attachment: Bug 563436 Part 1 - Minor clean up in ReflowOverflowContainerChildren(). r?dholbert,#layout

Remove prevRect and comments that don't seem useful; move variable
declarations closer to their usage sites. This is a preparation for Part 2.

This doesn't change behavior.

Comment 14

[Ting-Yu Lin [:TYLin] (PDT, UTC-7)]

Attachment: Bug 563436 Part 2 - Always override inline-size for overflow container frames. r?dholbert,#layout

prevInFlow.ISize(wm) has subtracted the margin in the inline-axis. If we use
it as the current overflow container continuation's available inline-size, we
might subtract the margin again when computing its auto inline-size [1], which
leads to progressively compressed inline-size as we see in various testcases.
However, fixing availSpace is not correct because not all types of frames
computes their inline-sizes using the available inline-size.

This patch generalizes the solution for flex item in bug 1811548, i.e. we always
override inline-size for the current overflow container continuation.

[1] https://searchfox.org/mozilla-central/rev/ff6e63804e6f7b548a26338708663093ce16be11/layout/generic/nsIFrame.cpp#6736-6737

Comment 15

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/4630d52e22d6
Part 1 - Minor clean up in ReflowOverflowContainerChildren(). r=dholbert
https://hg.mozilla.org/integration/autoland/rev/cec72c3f4e5b
Part 2 - Always override inline-size for overflow container frames. r=dholbert,layout-reviewers

Comment 16

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

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/45118 for changes under testing/web-platform/tests

Comment 17

[pstanciu]

https://hg.mozilla.org/mozilla-central/rev/4630d52e22d6
https://hg.mozilla.org/mozilla-central/rev/cec72c3f4e5b

Comment 18

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

Upstream PR merged by moz-wptsync-bot