[Magick-bugs] Problems creating PDFs from Indexed PNGs

[Edward Tandi]

Glenn Randers-Pehrson wrote:
> At 07:17 PM 9/18/2006 +0100, Edward Tandi wrote:
>   
>> I am trying to create a PDF document from 4 scanned PNG images.
>>
>> When the PNGs are in indexed mode (gimp:image->mode->indexed), the
>> "convert *.png out.pdf" command succeeds, but when trying to view it,
>> only the first two pages show. I've tried viewing this with evince, xpdf
>> and acroread -same result.
>>
>> When I convert these scans to normal grayscale images, it is possible to
>> see all the pages in the PDF document.
>>
>> I think this is a bug. Does anyone know of a workaround that allows me
>> to use indexed PNGs?
>>
>>     
>
> With ImageMagick-6.2.9-6 I got the expected behavior, namely 4 pages
> with an image.  I used "convert logo: logo.png" and then
> "convert logo.png logo.png logo.png logo.png 4logos.pdf".  I checked
> the logo.png with "pngcrush" to be sure it is an indexed PNG.
> I checked the 4logos.pdf with "convert 4logos.pdf new.png" and got
> four output files, and I looked at the file with acroread on a windows
> machine.
>
> I imagine it is possible that your input files have offsets that are
> throwing them off the PDF page.  Try the "+repage" option to get rid
> of any offsets.
>   
Thanks, I tried the +repage option, but same result -only the first few
pages show.

It looks like this bug only appears with certain input files.
Imagemagick.org now have a reproducible test case and are working on a fix.

It's good to see such a fast response.