[jira] [Commented] (PDFBOX-2268) AES-256 decryptions fails

["ASF subversion and git services (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/PDFBOX-2268?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14094336#comment-14094336 ] 

ASF subversion and git services commented on PDFBOX-2268:
---------------------------------------------------------

Commit 1617537 from [~tilman] in branch 'pdfbox/trunk'
[ https://svn.apache.org/r1617537 ]

PDFBOX-2268, PDFBOX-2269: Support for AES-256 Rev. 5 Decryption (Acrobat 9), relaxed validation of the Perms dictionary, as done by Michele Balistreri

> AES-256 decryptions fails
> -------------------------
>
>                 Key: PDFBOX-2268
>                 URL: https://issues.apache.org/jira/browse/PDFBOX-2268
>             Project: PDFBox
>          Issue Type: Bug
>          Components: PDModel
>    Affects Versions: 2.0.0
>            Reporter: Michele Balistreri
>         Attachments: AES256-fix.diff
>
>
> When opening a document encrypted with AES-256 (owner password only) by providing no password, the isUserPassword method fails, since it tries, indirectly, to decrypt a null pointer by calling computeUserPassword. The result of computeUserPassword would be ignored even if the call succeeded, since it is not need for AES-256 encryption.
> Also, the code validating the Perms dictionary is correct, but unfortunately not even Acrobat seems to write Perms correctly (in my case P = F0C0 and Perms = F2C0), so that check needs to be relaxed. Perhaps logging the issue instead of throwing an exception would be more adequate.
> Provided is a patch for both issues. I understand it is probably suboptimal but I am completely new to the project and have not yet had the time to study all coding conventions. Considering the patch is very small maybe someone can take it as a pointer of what needs to be changes.



--
This message was sent by Atlassian JIRA
(v6.2#6252)