Preparing PDFs for use as Prawn templates

3
votes

We've got a system that takes in a large variety of PDFs from unknown sources, and then uses them as templates for new PDFs generated by Prawn.

Occasionally some PDFs don't work as templates for Prawn- they either trigger a generic Prawn error ("Prawn::Errors::TemplateError => Error reading template file. If you are sure it's a valid PDF, it may be a bug.") or the resulting PDF comes out malformed.

(It's a known issue that some PDFs don't work as templates in Prawn, so I'm not trying to address that here: [1] [2])

If I take any of the problematic PDFs, and manually re-save them on my Mac using Preview > Save As [new PDF], I can then always use them as Prawn templates without any problem.

My question is, is there some (open source) server-side utility I can use that might be able to do the same thing- i.e. process problematic PDFs into something Prawn can use?

1
rubypdfprawn
What are really problematic PDFs? and what basis is Prawn restricting them? - uday
Do you want to automate the resave PDF procedure to allow the praswn to use it? You can just use the ruby code to automate it, if you have already known the procedure. - Малъ Скрылевъ
@uDaY - Myriad formatting issues- see the referenced links for details. - Yarin

1 Answers

0
votes

Yarin, it at least partially depends on why the PDFs don't work in the first place. If you can use them after re-saving with Apple's (quite bad) preview PDF code, you should be able to get the same result using a number of different tactics:

-) Use an actual PDF library to open and save the PDF files (libraries from Adobe and Global Graphics come to mind). These are typically commercial products but (I know the Adobe library the best) they do allow you to open a file and save it, performing a number of optimisations in the process. The Adobe libraries are currently licensed through a company called DataLogics (http://www.datalogics.com)

-) Use a commercial product that embeds these libraries. callas pdfToolbox comes to mind (warning, I'm affiliated with this product). This basically gives you the same possibilities as the previous point, but in a somewhat easier to use package (command-line use for example).

-) Use an open source product. I'm not very well positioned to provide useful links for that.

There is another approach that may work depending on your workflow and files. In graphic arts bad files are sometimes "made better" by a process called re-distilling; you basically convert the PDF file to PostScript and re-distill the postscript into PDF again. Because this rewrites the whole file structure, it often fixes fundamental problems. However, it also comes with risks as you're going through a different file format. Libraries such as GhostScript (watch the licensing conditions) may allow you to do this.

Given that your files seem to be fixed simply by using preview, I would think a redistilling approach would be overly dangerous and overkill. I would look into finding a good PDF library that can automatically open and save your files.