Returns merged.pdf, which contains the third page of multi-page-1.pdf and the third page of multi-page–2.pdf.
I would like to merged.pdf to contain all of the pages of each multi-page pdf. I have so far not found a way of telling the convert command to use a range of pages, although I have tried adding [0-1] and [0,1] at the end of the filenames.
Interestingly, this ghostscript command (which I found via StackOverflow but cannot re-find) does work as I would like it to:
Error: /undefinedfilename in (dNOPAUSE) Operand stack:
Execution stack: %interp_exit .runexec2 --nostringval–
–nostringval-- --nostringval-- 2 %stopped_push --nostringval-- --nostringval-- --nostringval-- false 1 %stopped_push Dictionary stack: --dict:1159/1684(ro)(G)–
–dict:0/20(G)-- --dict:77/200(L)-- Current allocation mode is local Last OS error: No such file or directory GPL Ghostscript 9.07:
Unrecoverable error, exit code 1 [~/Downloads/toolkit_docs] % Total
% Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:–:-- --:–:–
–:–:-- 0 % Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed 100 7461 100 7461 0 0 8590 0 --:–:-- --:–:–
–:–:-- 16543 100 7628 100 7628 0 0 8360 0 --:–:-- --:–:-- --:–:-- 15225
So not working for me quite yet, but I can feel that we’re getting closer!
Portable Document Format is a file format which contains forward
and backward links. It is not a stream format like PostScript.
You cannot pipe PDF files to the stdin of Ghostscript. Instead
you must either give the PDF filename on the command line, or
use it as the argument of the Ghostscript run command.
Ghostscript 8.00 and later can read piped PDF by copying
to a temporary file.
I could easily write some bash that would download the pdfs to tempfiles then feed them to gs, but I’m guessing that won’t work since you can’t save files on heroku, right?
Well, you can save files on heroku but they are only guaranteed to be around for the duration of the request and may or may not be available on subsequent requests. I believe the determining factor is going to be whether or not the dyno has been spun down.
Either way, you should be able to use locally saved files reliably for short-lived processes. docs here
We were hoping to avoid saving to Heroku’s tmp directory, but that is probably what we’re going to have to do. If that is the case, we can just loop through each PDF and still use ImageMagick’s convert command.
I still wonder, though: If ghostscript can merge multi-page PDFs and ImageMagick uses ghostscript, then why can’t we use ImageMagick to convert PDFs the way that ghostscript does? (eg: no index numbers needed to select page numbers).
Still needs to be refactored a bit, but it is functional. It finds the files I would live to merge via the attachment instance (the object paperclip is created the attachment for). Those files are themselves paperclip attachments. I download each of those documents into the ‘tmp’ directory, then use the ‘pdf-reader’ gem to find the page count and use that page count to iterate over each page of each PDF in order to add it to the ‘convert’ command input.
I could go into more detail on the rest, but I think I will save that for a blog post.