PDFParser::ParseFileDirectory,Unexpected object at xref start
huangtiansama opened this issue · 9 comments
I need to parse a PDF file, but it fails to parse
Hi huangtiansama, care to share the pdf so i can figure out why the xref start os read incorrectly?
Hi huangtiansama, care to share the pdf so i can figure out why the xref start os read incorrectly?
OK, but you need to wait Monday
Sure, no problem. Thank you :)
Sure, no problem. Thank you :)
sorry, This file is too large
I've tried some solutions that jump 8 bytes after "endobj" and so far, it works fine
Well could also contact whoevers created the pdf and tell them they create faulty ones.
Is this loading on acrobat?
Wanna try sending to me via email? Gal.kahana@hotmail.com
Wanna try sending to me via email? Gal.kahana@hotmail.com
OK, the email has been sent
ok. looks to be like a consistent 8 byte offset here.
normally i'd tell you to go back to whoever created that PDF and get them to fix the position...which you probably should anyways if you can.
Still. seems like i can create a fairly safe solution so that it's both not horribly unsafe and seems to read the file ok.
implemented here - #277.
if works for you, i can incorporate in the main branch
the solution didn't work. parts of the xref do point to the right positions. this PDF need to be corrected (adobe acrobat can be used for that, by opening and closing). then it can be used with this library