Summary: | Corrupted PDF causes abort with 'Call to Object where the object was type 11, not the expected type 8' | ||
---|---|---|---|
Product: | poppler | Reporter: | Tom Parker <freedesktop> |
Component: | general | Assignee: | poppler-bugs <poppler-bugs> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | normal | ||
Priority: | medium | ||
Version: | unspecified | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
Tom Parker
2008-07-01 10:43:25 UTC
Corrupted document (accidentally generated during normal use) is available at http://tevp.net/sdarticle-1.pdf (6.2mb - the 1mb upload limit sucks for people without webspace BTW). The crash is with poppler 0.8.4. Fixed, thanks for reporting. The patch (http://cgit.freedesktop.org/poppler/poppler/commit/?id=0189ff8b86de18486f7397076f7a0fbf133a1a33) doesn't appear to fix things, and I'm still getting the same crash+stacktrace. Going through with gdb says the problem is on line 974 of glib/poppler-page.cc, and that the check there is only for thumb.isNull() but the thumb is an error thumb and so thumb.isError() should be checked there as well. Doing this fixes the bug for me, as evince is now able to show the rest of the document. I now get lots of errors of the form "Error (379611): Bad 'Length' attribute in stream" on the command line and "Weird Page Contents", but given it's a partially corrupted document this isn't entirely surprising. Right, the problem was also on the glib frontend, should be fixed now |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.