The library doesn't work for pdfs with large number of pages about pdf2image HOT 15 CLOSED

If you just want a specific page there are the first_page and last_page parameters. As for the output_folder, you really want to use a tempfile.NamedDirectory to ensure that the files get cleaned up afterwards.

from pdf2image.

From the top of my head, at that size, you want to make sure that your are using an output_folder and that your format is not ppm.

I'll make a test case on my side as I have never tested more than 500 pages.

from pdf2image.

Just tested with a 2368 pages PDF. If you allow for enough open file handle it does not seem to cause any issue on Ubuntu Server 18.04.

from pdf2image.

please may i know, what change i should make in my linux system to allow enough file handle?

from pdf2image.

On my system, I was able to simply ulimit -n 70000 to make it work.

You might want to make sure that this will persist between reboots.

from pdf2image.

I made it to 70000 as you said...still no chance . Can you try with this one
https://drive.google.com/file/d/13Srl0JfQ-ZCKGnf95ZxFu0JdeH2P9igx/view?usp=sharing

from pdf2image.

I'm getting this issue even with a PDF with less than 200 pages (less than 500 KiB).

@Belval is it possible to return the list of images as a generator instead (if that would solve this issue)?

Basically, I just want the image of one page and convert_from_path(path)[0] is freezing my computer.

Edit: Tried the output_folder="/tmp" and fmt="jpg" parameter as suggested above and now my computer doesn't freeze. That's okay. But I need to manage all the unnecessary files generated. It would have been great if that was handled by pdf2image itself. 🙂

from pdf2image.

As for why the generated images are not handled by pdf2image, that's because of how Pillow handles images. You see, when you do Image.open(), the image is just lazy-loaded until you actually use it. If pdf2image cleaned the temp directory, the image converted would become unusable.

from pdf2image.

Ah, I see. No problem. I used tempfile.TemporaryDirectory as you suggested and it's working fine. 👍 😃

Just out of curiosity, would it be possible to return a generator instead of all pages at once?

from pdf2image.

Unfortunately no, as pdftoppm returns all pages at once and this is just a thin wrapper with added multithreading.

from pdf2image.

I am unable to open the pdf with 2nd password.

from pdf2image.

I just try to convert a pdf file less than 50 pages to jpg, there are no jpg files output also, but png. what should i do

from pdf2image.

By the way, pdf2image's version is 1.6, and what version of poppler is better?

from pdf2image.

This issue is now pretty old and a lot have changed since it was originally opened. Feel free to open another issue if you still have issues with PDFs that have a lot of pages.

from pdf2image.

anyone managed to solve the issue with large pdfs with a lot of pages?
i am trying to loop each page so generator would be a good solution.

anyone figured it out?

from pdf2image.