[
https://issues.apache.org/jira/browse/PDFBOX-2009?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
John Hewson closed PDFBOX-2009.
-------------------------------
Resolution: Not a Problem
It turns out that the section in the PDF spec about the BOM does not apply to
content stream text - instead it applies to bookmarks, annotations, etc. The
code size of content stream text should actually be determined by the CMap of
the current Font.
> PDFStreamEngine.processEncodedText incorrectly handling UTF-16 text with BOM
> FEFF
> ---------------------------------------------------------------------------------
>
> Key: PDFBOX-2009
> URL: https://issues.apache.org/jira/browse/PDFBOX-2009
> Project: PDFBox
> Issue Type: Bug
> Components: Text extraction
> Affects Versions: 2.0.0
> Reporter: Philip Helger
> Fix For: 2.0.0
>
> Attachments: test-properties.pdf
>
>
> When having a text print operation like
> <FEFF21222193219103B103A003A6> Tj
> than the PDFStreamEngine.processEncodedText does not handle this correctly.
> Am I correct that if a BOM was determined, the codelength should be set to 2
> (and not be changed)? Or should alternatively simply the BOM be skipped?
> It may be related to PDFBOX-920
--
This message was sent by Atlassian JIRA
(v6.2#6252)
