[
https://issues.apache.org/jira/browse/PDFBOX-538?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12765917#action_12765917
]
Andreas Lehmkühler commented on PDFBOX-538:
-------------------------------------------
Thanks for the detailed explanation. Obviously there is another possible value
of V=4 since pdf 1.5 in the pdf-reference v1.7. IMO we have to take that value
also into amount to be 1.5 compliant. If we really support that value or just
throw an exception or log an error/warning that's another story.
> CryptographyException on Adobe Distiller generated file
> -------------------------------------------------------
>
> Key: PDFBOX-538
> URL: https://issues.apache.org/jira/browse/PDFBOX-538
> Project: PDFBox
> Issue Type: Bug
> Components: Parsing
> Affects Versions: 0.8.0-incubator
> Reporter: Luc Claes
>
> Well... this is actually *not* a PDFBox bug, but an Adobe Distiller (7.0.x)
> one...
> The following exception is thrown in
> StandardSecurityHandler.computeEncryptedKey:
> org.apache.pdfbox.exceptions.CryptographyException: Error: length should be 5
> when revision is two actual=16
> The PDF file is accepted by various readers and contains the following object:
> 2106 0 obj
> <<
> /Filter /Standard
> /V 1
> /R 2
> /Length 128
> /P -12
> /O (f¥—èàã؇�æa§�»eücmý/Æ6ïàY÷[*™6)
> /U (å£átÂ>c÷sÉcû^KÞ•À›¶YjCò~)
> >>
> endobj
> Referenced by an encrypt tag:
> /Encrypt 2106 0 R
> PDFBox rejects (rightly) such an object with version 1, revision 2 and a
> length != 5*8
> That said, we are parsing many of those files to feed a Lucene index. Could
> the validation be relaxed ?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
