pdftotext.Error: Poppler error creating document

[prakritidev]

while using pdftotext with multiprocessing module on ec2

('read pdf file', '1004.5293.pdf')
Traceback (most recent call last):
  File "main.py", line 44, in <module>
    result = pool.map(pdf_extract, filenames)
  File "/usr/lib64/python2.7/multiprocessing/pool.py", line 251, in map
    return self.map_async(func, iterable, chunksize).get()
  File "/usr/lib64/python2.7/multiprocessing/pool.py", line 567, in get
    raise self._value
pdftotext.Error: Poppler error creating document

My code:

def pdf_extract(dirs):
    paths, filename = dirs
    file = filename.replace(".pdf", ".txt")
    if file in have:
        print("file alreafy extracted!!")
    else:
	print("read pdf file", filename)
        with open(os.path.join(paths, filename), "rb") as f:
            pdf = pdftotext.PDF(f)
            prin(len(pdf))
        text = "\n\n".join(pdf)
        print("converted file")
        file = filename.replace(".pdf", ".txt")
        with open(txt_path+file, "w") as f:
            f.writelines(text)
            f.close()
            print("saved file")
        time.sleep(0.01)

Link : arxiv paper

[jalan]

Can you attach the PDF in question, 1004.5293.pdf?

[prakritidev]

Its a research paper from arxiv.org

[jalan]

Okay, I found it. Works fine here:

>>> import pdftotext
>>> f = open("1004.5293.pdf", "rb")
>>> pdf = pdftotext.PDF(f)
>>> len(pdf)
34

What version of poppler do you have?

[prakritidev]

I'm using Version: 3:4.8.5-1, I had 1 million pdf files and I got this error for many files. I'll try to do this again. I'm not sure if I'm doing something wrong in my function. Please let me know if you think I'm doing something wrong while reading pdf files.

Thanks for the help.

[jalan]

3:4.8.5-1 doesn't look like a poppler version. It should look something like 0.63.0. You can see some common versions at poppler.freedesktop.org under the Packaged Versions section.

Your code looks fine to me, but if you're using an older version of poppler, errors like this are more common.

[anuragladdha-ml]

how do I install poppler on windows? After I download a zip, which directory it needs to be placed at.

[bhakti-visotrust]

I'm getting this error when running pdftotext.PDF(f) on AWS Lambda (Linux) but it works successfully with the same document when running on MacOS.

using pdftotext==2.2.2, poppler-utils==0.1.0

[AlexisH]

I'm getting this error too when running pdftotext.PDF(f) on Ubuntu 20.04.5 LTS. Was working fine last month. No code change.
I upgraded the pdftotext package from 2.1.5 to 2.2.2 but the issue is still there.

Edit: nevermind. I found one of the read files was actually not a pdf file but a png image of the document...

[dvtate]

Same issue with same file, Arch Linux, poppler 23.12.0-1, and pdftotext 2.2.2-4