[cups.development] pdf with transparent image and cups rasterfilter

Michael Sweet msweet at apple.com
Fri Jun 11 19:21:31 PDT 2010


OK, you'll be using Ubuntu's PDF filters, so please follow up with the Ubuntu folks.

On Jun 11, 2010, at 5:29 PM, Paul Newall wrote:

> Ubuntu 9.10
> evince 2.28.1
> poppler 0.12.0
> cups 1.4.1
> 
> The file was made in MS word 2007, evince says it is pdf v 1.5. It's a tga image with some transparent areas positioned over the top of some text, so should not be very complex (though who knows what MS word actually does?).
> It displays correctly in evince.
> The print preview looks OK in evince.
> But printing from evince results in a blank page, and a quick look at what happens in my filter suggests the raster is all zero bytes.
> 
> However, I can print to file in evince, making a postscript file.
> If I open that ps file in evince, it looks OK and prints OK.
> 
>> From the cups error log it looks as though evince normally sends the pdf file to cups, without first converting to ps.
> So I guess cups is not managing to convert this pdf to a raster, but evince can convert it to ps OK when it prints to file?
> 
> If you want to play with the file, it is here:
> https://sourceforge.net/projects/cupsdriverkodak/files/SampleTransparent.pdf/download
> 
>> What OS/Linux distribution?
>> 
>> What version of CUPS?
>> 
>> On Jun 10, 2010, at 1:36 PM, Paul Newall wrote:
>> 
>>> I have a cups raster filter, using colorspace CUPS_CSPACE_K or =
>> CUPS_CSPACE_CMYK
>>> Some pdf files fail to print.  The common factor seems to be that they =
>> contain an image with some transparency.
>>> The filter is behaving as though the raster was blank (every byte of =
>> the data =3D 0). It's possible that there's some problem with the =
>> filter, but before I spend a lot of time looking at that...
>>> Are there any known problems with cups converting pdfs with =
>> transparency to raster format?
>>> Any ideas for how I should investigate this?
>>> Maybe the pdf files are not to specification in some way? Is there a =
>> pdf file validity check I can run? or are there sample "correct" pdf =
>> files somewhere I can download? Or sample postscript files with =
>> transparency that I can use to see if the problem applies only to pdfs?
>>> _______________________________________________
>>> cups-dev mailing list
>>> cups-dev at easysw.com
>>> http://lists.easysw.com/mailman/listinfo/cups-dev
>> 
>> ________________________________________________________________________
>> Michael Sweet, Senior Printing System Engineer, PWG Chair
>> 
>> 
>> 
>> 
> 
> _______________________________________________
> cups-dev mailing list
> cups-dev at easysw.com
> http://lists.easysw.com/mailman/listinfo/cups-dev

________________________________________________________________________
Michael Sweet, Senior Printing System Engineer, PWG Chair








More information about the cups mailing list