Problem converting pdf to png about pdf2image HOT 5 CLOSED

That sounds very specific, do you think you could send me the PDF to create a test case?

from pdf2image.

I sent you the PDF by email (I cannot share it publicly).

from pdf2image.

If you print like so

while index < len(data):
       file_size = data[index:].index(b'IEND') + 8 # 4 bytes for IEND + 4 bytes for CRC
       print(f"file_size {file_size} index {data[index:].index(b'IEND')} len data {len(data)}")
       images.append(Image.open(BytesIO(data[index:index+file_size])))
       index += file_size

you'll notice the data[index:].index(b'IEND') is in the middle of the file. However, calling Image.open(BytesIO(data)) works just fine, since for this file there is only one page

from pdf2image.

I was able to reproduce the issue, I will update the package this weekend or earlier.

In the meantime, while I don't know your use case, but you can probably provide an output_folder (since the resulting file are written to disk you wouldn't have this issue):

import tempfile

with tempfile.TemporaryDirectory() as path:
     images_from_path = convert_from_path('/home/belval/example.pdf', output_folder=path)
     # Do something here

Or convert using the default ppm format and then save as PNG using Pillow.

for i, img in enumerate(convert_from_bytes(pdf_data)):
    img.save(f"{i}.png")

Since PPM is a raw format, you won't have any loss in quality.

from pdf2image.

I fixed the issue and uploaded a new package to pypi. You should not be able to reproduce the bug with version 1.4.2 (at least I was not).

Thank you for taking the time to report the issue.

from pdf2image.