You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think I have this same issue - so wanted to share what I found while trying to troubleshoot in case it adds useful context. Looks like #645 may be the same thing too. In my case the following lines in the PDF are causing an infinite loop (using v1.24):
464 0 obj
<</Dest(þÿ s e c S 0 0 4)/Next 463 0 R/Parent 399 0 R/Prev 465 0 R/Title(þÿ P a r t A\t & \t .)>>
endobj
465 0 obj
<</Dest(þÿ s e c S 0 0 3)/Next 464 0 R/Parent 399 0 R/Prev 466 0 R/Title(þÿ P a r t B)>>
endobj
In the Tokenizer it seems that the backslash in the title causes it to go into Literal.readBackslashSequence() where it fails to see the >> that should end the entry. It then proceeds to read in the following row as part of the same object, which sets the Next value back to 464 and causes an infinite loop where it keeps reloading the same garbled object. In my case the PDF gets stuck in this loop.
I note that the PDF attached to this issue and the one attached to #645 both have backslashes in the Title property of a dictionary entry. I tested the fix in PR #652 and it works for my issue too.
Patch integration tests and added regression test files for #652:
- patched the result of pdf-hul-76-372051162.pdf; and
- added regression tests for #645 and #646.
Using both UI and CLI (on Windows environment, tested on different machines) the attached file nevers seems to reach a result for validation.
897714407.pdf
The text was updated successfully, but these errors were encountered: