Hi Maruan,

I have successfully converted pdf to images using pdfbox. However there are couple of attached pdf which does not shows up filled in details.

In my use case i need to merge different pdf's and than convert to series of images. Since these pdf are not converted to image properly i would like to avoid merge or fix image generation. Can you please point me to right direction.

Converting this pdf with imagemagick works fine but i would like to avoid os level tool.

Really appreciate your help

Thanks
-Ankur Tripathi


On Thu, Jun 6, 2013 at 4:01 PM, Maruan Sahyoun <sahyoun@fileaffairs.de> wrote:
Hi,

the question if a PDF can be rendered successfully is not dependent on the fact that it's a PDF/A file. In general PDFBox does a good job in converting a PDF to image and it supports PDF/A as well as not PDF/A compliant files. There are some limitations though which may or may not apply to you.

- there are limitations in PDFs render mode i.e. not all possible render modes are supported
- there are limitations in PDFs shading i.e. not all shadings are supported
- font rendering is dependent on awt i.e. PDFBox generates a font from embedded font which is passed to awt. This works in most but not all cases.
- always test if e.g. Adobe Reader can display the file
.

If while rendering PDFBox hit's an (yet) unsupported feature it will be reported. If you come across such a limitation please log an enhancement request in Jira (you should search first if someone else already had a similar issue and add to that) so we can look into removing the limitation. Of course if you are able to contribute that's even better.

For PDF/A (PDF/A-1b) PDFBox passes several test suites. So if you think you have a valid PDF/A file and PDFBox complains we are very interested in finding out why this is the case. But it's very likely that your file might not be PDF/A-1b compliant.

If you have specific questions/issues please feel free to ask.

Bottom line - I think PDFBox will help you doing the conversion. Your milage will vary dependent on the files content.


BR
Maruan Sahyoun

Am 06.06.2013 um 12:12 schrieb Ankur Tripathi <ankur@intelligrape.com>:

> Hi,
>
> I have a use case in my project where i want to convert every page of pdf
> to image. I have tried different opensource libraries like pdfrenderere,
> open source version of jpedal etc but with each of them we have problems
> with PDF/A pdf's. Before trying pdfbox i would like to know that if support
> for embedded font and pdf/a is available in pdfbox api. If not is there any
> way to identify if a particular pdf can not be converted to image, I
> already tried http://pdfbox.apache.org/cookbook/pdfavalidation.html to
> validate uploaded pdf's but it fails for all of our pdfs but we are able to
> convert them into image properly. There are only few formed filled pdf
> which have not been converted to image properly.
>
> Thanks for help.
>
>
> Thanks
> -Ankur Tripathi