[
https://issues.apache.org/jira/browse/PDFBOX-3162?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15106830#comment-15106830
]
Philip Helger commented on PDFBOX-3162:
---------------------------------------
With 2.0.0 RC3 the exception stack trace changes slightly and I managed to get
it up in the debugger.
The exception is now in TrueTypeFont.getUnicodeCmap (true).
The ttf (type TrueTypeFont) to be embedded has the following tables:
* fpgm
* cvt
* glyf
* loca
* gasp
* hmtx
* hhea
* maxp
No "cmap" table present... The screenshot tries to capture as many variables as
possible - hth
> IllegalStateException in TTFSubsetter
> -------------------------------------
>
> Key: PDFBOX-3162
> URL: https://issues.apache.org/jira/browse/PDFBOX-3162
> Project: PDFBox
> Issue Type: Bug
> Components: FontBox
> Affects Versions: 2.0.0
> Environment: 2.0.0-RC2
> Reporter: Philip Helger
> Attachments: variables-on-error.png
>
>
> Hi encountered a rare exception with an empty TTF subset:
> {code}
> ==> [1] caused by java.lang.IllegalStateException: subset is empty
> 1.: org.apache.fontbox.ttf.TTFSubsetter.writeToStream(TTFSubsetter.java:921)
> 2.:
> org.apache.pdfbox.pdmodel.font.TrueTypeEmbedder.subset(TrueTypeEmbedder.java:304)
> 3.: org.apache.pdfbox.pdmodel.font.PDType0Font.subset(PDType0Font.java:162)
> 4.: org.apache.pdfbox.pdmodel.PDDocument.save(PDDocument.java:1102)
> 5.: com.helger.pdflayout.PageLayoutPDF.renderTo(PageLayoutPDF.java:276)
> {code}
> Unfortunately I don't know yet what was causing the problem, but I will
> provide you with more details on Monday (if necessary).
> If there is nothing to subset - I think the call should simply be ignored???
> Or maybe this is a problem because of the "uni" name bug (in 2.0.0-RC2) I
> reopened lately?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
