[
https://issues.apache.org/jira/browse/PDFBOX-1648?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14157965#comment-14157965
]
Kevin Pearcey edited comment on PDFBOX-1648 at 10/3/14 1:31 PM:
----------------------------------------------------------------
This looks like a similar case to PDFBOX-2035, which further highlights that
the CMap parser was only using whitespace as delimiters in some cases where in
fact the full range of allowed PDF delimiter characters should be used.
I have updated the patch to the latest 1.8.8-SNAPSHOT and will attach it in the
hope it may make it to the src, and we can stop patching every release we
upgrade to.
was (Author: [email protected]):
This looks like a similar case to PDFBOX-2035, which further highlights that
the CMap parser was only using whitespace as delimiters in some cases where in
fact the full range of allowed PDF delimiter characters should be used.
I have updated the patch to the latest 1.8.9-SNAPSHOT and will attach it in the
hope it may make it to the src, and we can stop patching every release we
upgrade to.
> FontBox can't load CMaps with no spaces between tokens
> ------------------------------------------------------
>
> Key: PDFBOX-1648
> URL: https://issues.apache.org/jira/browse/PDFBOX-1648
> Project: PDFBox
> Issue Type: Improvement
> Components: FontBox
> Affects Versions: 1.7.0, 1.8.2
> Reporter: Max Gilead
> Priority: Minor
> Labels: patch, test
> Attachments: CMapNoWhitespace.txt, CMapWithWhitespace.txt,
> cmap-with-no-whitespace.patch, cmap.diff
>
>
> FontBox (tested versions 1.7.0 and 1.8.2) can't load CMap entries if they're
> not properly separated by whitespace making it problematic to extract the
> text from such PDFs. Test case and a proposed patch attached. The patch is
> tiny but in case it matters: it is in public domain, no credit necessary,
> please feel free to relicence it as you see fit.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
