Program Stuck with a pdf about pdftotext

[jalan]

You can use the multiprocessing module to perform long-running tasks with a timeout. Here is a commented example script I called readpdf.py:

import multiprocessing
import sys
import time

import pdftotext


def main():
    path = sys.argv[1]
    text = read_pdf(path)
    if text is None:
        sys.exit("timed out trying to read the PDF")
    print(text)


def read_pdf(path, timeout=5):
    # A queue to pass data between processes
    q = multiprocessing.Queue()

    # Start a new process
    p = multiprocessing.Process(target=read_pdf_process, args=(path, q,))
    p.start()

    # Wait some time for it to finish
    p.join(timeout=timeout)

    if p.is_alive():
        # The process is still running but we are out of time, so terminate it
        p.terminate()

    # Based on the exit code, either get the text that was extracted, or None
    # if we weren't able to finish
    p.join()
    if p.exitcode == 0:
        text = q.get()
    else:
        text = None

    q.close()
    return text


def read_pdf_process(path, q):
    with open(path, "rb") as f:
        pdf = pdftotext.PDF(f)
    text = "\n\n".join(pdf)
    q.put(text)
    q.close()


if __name__ == "__main__":
    main()

On a typical PDF, you get the text:

$ python readpdf.py abcde.pdf 
abcde.

On your troublesome PDF, it will time out after five seconds:

$ python readpdf.py 2014-ASHP-Handbook-web-edition.pdf 
timed out trying to read the PDF

I hope it helps.

[jalan]

Closing, since this is something you should solve in your own code. I shouldn't change this module itself to do this on my side, since spinning up another process would be expensive and surprising to most users.