[
https://issues.apache.org/jira/browse/PDFBOX-821?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13027194#comment-13027194
]
Lars Torunski commented on PDFBOX-821:
--------------------------------------
RandomAccessBuffer usage was introduced with PDFBOX-948. Hopefully switching
the default to in-memory scratch buffers in 1.5.0 doesn't raised a problem.
> RandomAccessBuffer returns wrong values for single byte reads, patch attached
> -----------------------------------------------------------------------------
>
> Key: PDFBOX-821
> URL: https://issues.apache.org/jira/browse/PDFBOX-821
> Project: PDFBox
> Issue Type: Bug
> Components: Parsing
> Affects Versions: 1.3.1
> Reporter: Timo Boehme
> Attachments: pdfbox_issue821.patch
>
>
> RandomAccessBuffer contains a critical bug which had no effect so far since
> this class is currently not used. The problem is in method read() (single
> byte read). If the byte read is larger than 127 the method returns the signed
> byte value instead of the unsigned values 128-255. Thus a component using
> this method will assume end-of-stream.
> The solution is to 'AND' the value with 0xff. Patch is attached.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
