159914 - Printing a div's content "eats" a few line down the first page

Status: RESOLVED FIXED

(Core :: Printing: Output, defect)

Description

[Claudio Weber Abramo]

[This was submitted previously but didn't show in my bug list, so I am
re-submitting it.]

A div's content do not completely print. Only the first page is printed.

Comment 1

[sujay]

correct me if I'm wrong anyone, but I feel this is a DUP of bug 122750

*** This bug has been marked as a duplicate of 122750 ***

Comment 2

[sujay]

verified.

Comment 3

[Jean-Marc Desperrier]

This has only *partially* been fixed by bug 154892.

The breaking down the first page doesn't happen properly, the last line is only partially printed, and in fact two lines have not been printed between the first and the second page (I have a modified test case with numbers on the line that makes that more visible).

Changing the bug title to reflect this, reopening the bug.

Comment 4

[Jean-Marc Desperrier]

I have put on line a modified version of the test case that shows the problem more precisely through line numbers :
http://jmdesp.free.fr/i18n/bug_159914/example4.html

Comment 5

[Jean-Marc Desperrier]

Attachment: Screen capture that line 28 and 29 are not properly displayed

Comment 6

[Boris Zbarsky [:bzbarsky]]

Please attach the testcase to the bug, so it won't go missing.

Comment 7

[fantasai]

Attachment: simplified testcase

Comment 8

[fantasai]

It seems like the y position is not being subtracted from the available height. The calculation is there on line 429, but is maybe using the wrong values.

Comment 9

[fantasai]

Attachment: patch

This fixes the test case. It's a very simple fix for printing dataloss, so I think it should get fixed in the FF3.0(.x) cycle.

It doesn't handle bottom-positioned cases, but those are pretty complicated with pagination. I can try to write a patch for that, too, but I'd like to do that separately.

Comment 10

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

What about the top:auto case? This won't be right in that case?

Comment 11

[fantasai]

Attachment: simplified testcase with auto top

The auto position seems to be incorporated into the offset data already. This testcase works, for instance. It's cases where we have top: auto; and bottom: something fixed; that would be problematic here.

Comment 12

[fantasai]

Attachment: testcase with margin

Comment 13

[fantasai]

Attachment: patch v2

Forgot to subtract out margins, too. Now copying the y position calculation below.

Comment 14

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

What about bottom margin and border? I guess we don't care much if the bottom margin goes off the page. But the border?

Comment 15

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

Hmm, I guess that should be OK.

Comment 16

[Mike Shaver (:shaver emeritus)]

Not going to block 1.9 for this, but 3.0.1 will like it a lot!

Comment 17

[fantasai]

Rationale: This is a very simple, low-risk fix for printing dataloss.

Comment 18

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

That's true, but this is also going to be fairly rare (only affecting top:auto abs-pos elements that span pages) and something that didn't work at all before in FF2, so not a regression.

Comment 19

[fantasai]

top: <length> || (top:auto && bottom: auto;) elements that span pages, actually.

Comment 20

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

Hmm, I guess so.

Comment 21

[fantasai]

Comment on attachment 322733 [details] [diff] [review]
patch v2

This has been sitting on wanted1.9.0.x? and approval1.9? since May. Trying approval1.9.0.2?...

Rationale: This is a very simple, low-risk fix for printing dataloss.

Comment 22

[fantasai]

Attachment: reftests

Comment 23

[Samuel Sidler (old account; do not CC)]

fantasai, can you get this landed on mozilla-central? We need some bake time before considering for 1.9.0.x.

Comment 24

[Samuel Sidler (old account; do not CC)]

Comment on attachment 322733 [details] [diff] [review]
patch v2

Please re-request approval after this lands and bakes on mozilla-central.

Comment 25

[Dão Gottwald [:dao]]

http://hg.mozilla.org/index.cgi/mozilla-central/rev/d6384ffd49f1
http://hg.mozilla.org/index.cgi/mozilla-central/rev/4b6b3b388fe2

hg patches, please...

Comment 26

[Samuel Sidler (old account; do not CC)]

Comment on attachment 322733 [details] [diff] [review]
patch v2

Approved for 1.9.0.2. Please land in CVS. a=ss

Be sure to land the tests as well.

Comment 27

[fantasai]

Checked into CVS for 1.9.0.2.

Comment 28

[juan becerra [:juanb]]

Verified on Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.4; en-US; rv:1.9.0.2) Gecko/2008090512 Firefox/3.0.2, and Fx3.0.2build4 on Vista. Print Preview displays the document correctly as does the paper printout, using the three test cases here.