Legacy ImageMagick Discussions Archive

I've read some of the other posts with this error but can not figure out a solution.

I'm trying to convert a specific PDF to jpg using the following:

/usr/local/ImageMagick/bin/convert -density 400 -resize 25% -quality 92 ift.pdf ift.jpg

The result is:
convert: Postscript delegate failed `ift.pdf': No such file or directory @ pdf.c/ReadPDFImage/634.
convert: missing an image filename `ift.jpg' @ convert.c/ConvertImageCommand/2822.

I use this same command for all other pdf's and it works fine. If I drop "-density 400" with this file then it will create jpg's. I have gone through each page with ghostscript and each page renders ok:

Page 1
%%BoundingBox: 0 0 612 792
%%HiResBoundingBox: 0.000000 0.000000 611.999981 791.999976
>>showpage, press <return> to continue<<

Page 2
%%BoundingBox: 36 21 577 636
%%HiResBoundingBox: 36.575999 21.383999 576.377982 635.327981
>>showpage, press <return> to continue<<

etc... (all pages are successful)

My software:
Red Hat Enterprise Linux ES release 4 (Nahant Update 8)
ImageMagick 6.0.7 07/27/08 Q16 http://www.imagemagick.org
GPL Ghostscript 8.70 (2009-07-31)

Again this happens with this PDF only (others are fine). I'd post a link to the PDF, but it has confidential information in it....

Any ideas to try?

Best guess is that your IM is too old. Yours is over 500 versions old. IM is now at 6.6.0-4 and you are at 6.0.7-x. It is really time to upgrade!

Perhaps your pdf file is corrupt. Also I read that someone said there might be a problem with GS 8.70 and backed down to 8.69. You might try that. Or post a link to your pdf file so others can check it.

Also if the pdf contains an image in cmyk format, you could try

convert -colorspace rgb -density 400 ....

Thanks for your reply! Sorry, I gave you the system installed version of IM. Here is the version I'm using:

Version: ImageMagick 6.5.6-10 2009-10-12 Q16 http://www.imagemagick.org

I tried the colorspace flag with no luck.

I've seen the suggestion in a few of the posts that a pdf file may be corrupt - how is that identified? If you can read it in Acrobat, read with gs, and decode to jpg without -density option what other conditions exist that would classify the pdf as corrupt?

Going to try to downgrade gs as per your suggestion.

Thanks again.

I don't see a gs 8.69 on sf or anywhere else.

I downgraded to 8.64 but got the same error...

I'm unable to post the file because it is a confidential business proposal.

Don't know how to tell if corrupt. Can you do

identify -verbose image.pdf

What info does it give?


Might you have a conflict between two versions of IM and perhaps it is using the older one?


Can you upgrade to the current IM version in case that particular version has a problem?

Try

convert -density 400 ift.pdf -resize 25% -quality 92 ift.jpg

fmw42 wrote:identify -verbose image.pdf

Lots of output - http://72.167.255.120/identify-output.html

I'm pretty sure this PDF was created on a Mac if it makes any difference.

I noticed in the beginning of the output it says Filesize: 91.53MiB, but the file is only 11mb on disk -
-rw-r--r-- 1 root root 11353144 Mar 8 16:00 ift.pdf

fmw42 wrote:Might you have a conflict between two versions of IM and perhaps it is using the older one?

Maybe. I have gs installed from source into system dirs, and IM prefixed to /usr/local/ImageMagick/. Each time I call convert I use the full path.

fmw42 wrote:Try

convert -density 400 ift.pdf -resize 25% -quality 92 ift.jpg

That's actually the order I use in my translation program. Gives same error.

convert: Postscript delegate failed `ift.pdf': No such file or directory @ pdf.c/ReadPDFImage/634.
convert: missing an image filename `ift.jpg' @ convert.c/ConvertImageCommand/2822.

Going to try to upgrade IM a bit later - I'll let you know how it goes