360697 - Printing nested framesets spools infinite pages

Status: RESOLVED WORKSFORME

(Core :: Printing: Output, defect)

Description

[Jason Karcz]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0

This is a simplified example, but whet the user presses the "Go" link in my demonstrative page, the bottom frame is selected. However, this frame is then expanded into another frameset. When the user prints that page, the default selection is "The Selected Frame". This causes Firefox to endlessly spool pages to the printer (easily >1000pp if not caught).

Reproducible: Always

Steps to Reproduce:
1. Press "Go" on my demo page
2. Select File->Print
3. Print using default settings

Actual Results:  
Pages are spooled endlessly (until all FF windows are closed)

Expected Results:  
The page should be printed as displayed.

In a computer lab setting where students don't realize what will happen and a web application is being used that triggers this flaw, it can turn into a massive waste of physical resources.

Additionally, this was not an issue in FF 1.0.7, but is in FF 1.5.0.6 and above (I   haven't tested any versions between 1.0.7 and 1.5.0.6)

Comment 1

[Marius Strumyla]

After you click "Go", firefox gets confused which frame is focused and should be printed.
IE6 prints both frames Left and Right with "Only the selected frame" checked. This kind of implies that IE selects two frames.

Tested with Firefox 2.0 on Win2k and Adobe PDF as Printer.

Comment 2

[Marius Strumyla]

Attachment: Left frame

Comment 3

[Marius Strumyla]

Attachment: Right frame

Comment 4

[Marius Strumyla]

Attachment: Frameset 2

Comment 5

[Marius Strumyla]

Attachment: Crap frame

Comment 6

[Marius Strumyla]

Attachment: Go frame

Comment 7

[Marius Strumyla]

Attachment: Testcase

Comment 8

[Marius Strumyla]

The latest trunk 20061115 does not spool pages endlessly anymore. I'll post some history how trunk behaviour changed wrt the testcase.

Comment 9

[Marius Strumyla]

Attachment: Printout with 20060223 trunk

The endless spooling was fixed between builds 20060222 and 20060223. 
Checkins between 2006-02-22 08:00 and 2006-02-23 10:00 :
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2006-02-22+08%3A00%3A00&maxdate=2006-02-23+10%3A00%3A00&cvsroot=%2Fcvsroot
Not sure which bugfix changed the behavior. Maybe bug 323923?

Comment 10

[Marius Strumyla]

Attachment: Printout with 20060330 trunk

Then trunk behavior changed between builds 20060329 and 20060330. See print sample. 
Checkins between 2006-03-29 05:00 and 2006-03-30 07:00 :
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2006-03-29+05%3A00%3A00&maxdate=2006-03-30+07%3A00%3A00&cvsroot=%2Fcvsroot
Possible suspect bug 331415

Comment 11

[Marius Strumyla]

Attachment: Printout with 20060411 trunk

Then trunk behavior once again changed between builds 20060410 and 20060411.
Checkins between 2006-04-10 10:00 and 2006-04-11 07:00 :
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2006-04-10+10%3A00%3A00&maxdate=2006-04-11+07%3A00%3A00&cvsroot=%2Fcvsroot
Possible suspect bug 332927

Comment 12

[Marius Strumyla]

This latest printout has dark background. It must be a regression from bug 332927.

Comment 13

[Eli Friedman]

(In reply to comment #12)
> This latest printout has dark background. It must be a regression from bug
> 332927.

Hmm, yeah, my fault.  I guess nobody noticed because the default setting for printing is no background.  It was a while ago, but I think I pertly fixed an issue with the background.

The cause of it is pretty simple: http://lxr.mozilla.org/seamonkey/source/layout/style/ua.css#196.  It shouldn't be too hard to fix, but I'll file another bug so I don't clutter this one.

About the original report: I have no idea what's causing infinite spooling. I'm not likely to try and diagnose it either (I generally don't do patches for branch, so I don't have a branch debug build; also, it's hard to get fixes into an incremental release).

BTW, what excatly does pages being spooled endlessly mean?  Just an inifinite number of completely blank pages coming out of the printer?  Or pages with just a header?  Or the same page over and over?  This is the first bug I've seen of that sort.  I can imagine the print code behaving oddly because it doesn't really expect the currently focused frame to be a frameset, but I don't see how it could keep on printing forever unless it somehow kept on printing the same document over and over.

Comment 14

[Marius Strumyla]

Basically it prints the first two pages as they appear in "Printout with 20060223 trunk" over and over.

Comment 15

[Pablo]

Hello! I have tried to reproduce the issue with the latest versions of Firefox 93.0 on windows 10
Marking this as Resolved > Worksforme since the issue no longer is reproducible.