[
https://issues.apache.org/jira/browse/PDFBOX-1922?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14002149#comment-14002149
]
Tilman Hausherr commented on PDFBOX-1922:
-----------------------------------------
Sorry I don't understand you or misunderstood you. I don't alter anything in
the PDF. What I was doing was reading the version from the header, which wasn't
done with the NonSeq parser. Next thing would be to read the version from the
trailer in the NonSeq parser (which apparently isn't done, but I'm not finished
with investigating this). Both would be put in the version variable of
COSDocument. Or do you mean that internally, COSDocument should keep both
versions separately, and then rewrite both in case a PDF is saved?
> NonSequentialParser not reading version in header
> -------------------------------------------------
>
> Key: PDFBOX-1922
> URL: https://issues.apache.org/jira/browse/PDFBOX-1922
> Project: PDFBox
> Issue Type: Bug
> Components: Parsing
> Affects Versions: 1.8.4, 1.8.5, 1.8.6, 2.0.0
> Reporter: Tim Allison
> Priority: Minor
> Attachments: testAnnotations.pdf
>
>
> On older pdf files, the NonSequentialParser returns a different PDFVersion
> than the traditional parser via the call:
> document.getDocument().getVersion().
> I think this is because NonSequentialParser is not reading the header but
> relying on the default 1.4 if no version is found in the trailer.
> Many apologies if this is a duplicate. I feel like I've seen this issue
> elsewhere, but I can't track it down.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
