From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0rc3) Gecko/20020523 BuildID: 2002052309 Printing to seems to print only one side of content at this site. Reproducible: Always Steps to Reproduce: 1. Open http://www.stud.ifi.uio.no/~haavardw/hfag/?dest=notater 2. Click on some of the headings to expand its content 3. In the menu bar of mozilla, choose File->Print... 4. Choose "Print to: file" (as opposed to Print to: printer) 5. Select "Printer: Postscript/default". 6. View the result file in some postscript viewer and compare to the page shown by mozilla. Actual Results: When comparing the postscript file and the page shown by mozilla, you'll se that only one page of content have been printed. In addition, there is a second page; almost blank, just containing a small footer. Expected Results: The whole content on the page printed to a postscript file. Validates as "XHTML 1.1 plus MathML 2.0": http://validator.w3.org/check?uri=http://www.stud.ifi.uio.no/~haavardw/hfag/?dest=notater

Håvard Wall wrote: > Validates as "XHTML 1.1 plus MathML 2.0": MathML printing is not supported by the PostScript module (see http://mozilla.org/releases/mozilla1.0/#printing). Please use Xprint module (see http://xprint.mozdev.org/ or http://puck.informatik.med.uni-giessen.de/people/gisburn/work/xprint/) for such cases...

Reporter: Does this postscript file print OK for you ?

The attachment did not contain enough content to show whether xprint would produce several pages or not. I would be glad if you could make a new one based on this url instead: http://www.stud.ifi.uio.no/~haavardw/mozillatest/cutoff.xml This is a new version with all its content expanded by default. It does not contain any MathML, and validates as xhtml 1.0 strict. [Unfortunately I'm not privilegied to install Xprint on my system (I tried, but I did not have access to 'lpc' which returned with illegal option 'a'; lpc status returned "no permission to control server".)]

This bug also show using File->Print Preview

The problem is still present in Mozilla 1.0

.. but both Opera 6 and MS Internet Explorer print this page correctly...

I get a ton of asserts with floaters, I think core layout should take a look first. (PLus there is MathML in this)

Reassigning to mathml to deal with assertions first.

The included CSS file has the following rule: ".main > * { float: left; }" It causes every direct descendant (including therefore MathML) to try to behave as a floater. Why would people want |float| on <math>, rather than just wrapping the whole lot in a <div> and putting their float there? Maybe mathml.css needs bullet proofing rules such as |float: none ! important| and friends? Or alternatively, the similar gymnastics done for other floated frames in the frame construction code could be replicated for the <math> case?

First: I don't believe the problem is MathML. Please use the following link instead which shows the same problem without any MathML: http://www.stud.ifi.uio.no/~haavardw/mozillatest/cutoff.xml (I think the reassignment to mathml was wrong.) Secondly: The MathML is not a direct descendant of class="main". All math is wrapped inside <div class="content">. This should prevent the math to behave like a floater. (?)

When investigating, it is customary to bounce the bugs when trying to collect further understanding and isolate the problem. The assertions that you heard earlier were/are firing on one (or perhaps many) <math> because it is being treated as a floater. It looks like a float rule is reaching it. Anyway, it is quite probable that the fact that the initial bug as reported, "print (preview) cuts off after only one side of content", is due to that CSS rule since it hits other elements as well. Maybe the fix is just to get the CSS stylesheet right. Re-assigning to core layout based on your above reduced non-MathML example (and since <math> is not expected to honor the float rule, do nothing for MathML is a possibility, but I might consider doing a graceful handling to avoid the assertions).

haavardw@ifi.uio.no wrote: > The attachment did not contain enough content to show whether xprint would > produce several pages or not. Oups... sorry... ;-( > I would be glad if you could make a new one based on this url instead: > http://www.stud.ifi.uio.no/~haavardw/mozillatest/cutoff.xml > This is a new version with all its content expanded by default. It does not > contain any MathML, and validates as xhtml 1.0 strict. OK, I'll file a new output file when my currrent build is "done" ...

haavardw@ifi.uio.no wrote: > [Unfortunately I'm not privilegied to install Xprint on my system > (I tried, but > I did not have access to 'lpc' which returned with illegal > option 'a'; lpc status returned "no permission to control server".)] Ugh. Weired. Can you go to http://xprint.mozdev.org/bugs/enter_bug.cgi?product=xprint and file a bug for that, please ? Thanks!

*** This bug has been marked as a duplicate of 45788 ***