multiple xrefs in pdf cause page extraction return only the first page

[meldonization]

PDF file: bad_page_number.pdf

• pdfminer extracts only the first page of the file
• parser.bufpos stop at size 10325
• it looks like this file contains two %%EOF lines

from pdfminer.pdfparser import PDFParser
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfpage import PDFPage
from pdfminer.pdfinterp import PDFResourceManager
from pdfminer.pdfinterp import PDFPageInterpreter
from pdfminer.pdfdevice import PDFDevice
from pdfminer.layout import LAParams
from pdfminer.converter import PDFPageAggregator
from pdf2html import *
fpath = '~/bad_page_number.pdf.pdf'
fp = open(fpath, 'rb')
parser = PDFParser(fp)
laparams = LAParams()
document = PDFDocument(parser)
rsrcmgr = PDFResourceManager()
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
print(len(list(PDFPage.create_pages(document))))

Could you please look into the problem please?

[bmteller]

This can be fixed by changing line.startwith(b'trailer') to line.strip().startswith(b'trailer'). https://github.com/pdfminer/pdfminer.six/blob/develop/pdfminer/pdfdocument.py#L102

the document has an xref table where the trailer keyword is indented.

0003575084 00000 n
 trailer
<</Size 973
>>

[pietermarsman]

Looks like a valid solution. Feel free to implement if you have the time.

[jstockwin]

I can take a look at this

[pietermarsman]

@jstockwin I accidentally merged the PR while the build was broken. Could you see what the issue was and recreate the PR?

[jstockwin]

Huh, oops. Sure, I will take another look...