[
https://issues.apache.org/jira/browse/PDFBOX-2563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14244998#comment-14244998
]
John Hewson commented on PDFBOX-2563:
-------------------------------------
We need a generic way to find substitute fonts, rather than hardcoding them. We
only hardcoded the standard 14 fonts and use heuristics to find all other
missing fonts.
There's a problem with trying to find substitutes using the Unicode characters,
due to [Han unification|http://www.unicode.org/faq/han_cjk.html], where the
same glyph might look very different in Japanese vs Korean vs Chinese.
TTF fonts don't explicitly label the languages which they contain. One idea I
could suggest it to sniff the languages in the 'name' table, which is the only
place in a TTF which is language-specific. We need to categorise each font on
disk C, J, or K, or none. I'd be interested to know how Evince is doing this.
> [PATCH] Use cmap for Type0/TTF fallback
> ---------------------------------------
>
> Key: PDFBOX-2563
> URL: https://issues.apache.org/jira/browse/PDFBOX-2563
> Project: PDFBox
> Issue Type: Bug
> Components: Rendering
> Affects Versions: 2.0.0
> Reporter: simon steiner
> Assignee: John Hewson
> Fix For: 2.0.0
>
> Attachments: VariousKFontsNotembeded218.PDF, ttfcmapfallback.patch
>
>
> java -jar ~/pdf-box-svn/app/target/pdfbox-app-2.0.0-SNAPSHOT.jar PDFToImage
> VariousKFontsNotembeded218.PDF
> This patch addresses some of the issues in PDFBOX-2509.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
