[
https://issues.apache.org/jira/browse/PDFBOX-2005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13950832#comment-13950832
]
Tilman Hausherr commented on PDFBOX-2005:
-----------------------------------------
The cause seems to be the CHARACTER_TO_NAME map in pdfbox.encoding.Encoding
which is generated by iterating through the NAME_TO_CHARACTER map. However
several different names have the same character code so the order is important.
I suspect that the existing implementation profited from an iteration order
that has changed now. I tried to add determinism by using LinkedHashMap but
that doesn't solve it. To do what I think was intended (i.e. that some "less
important" names are ignored if already there), I would have to change more
code and I prefer that somebody else do this.
So for now I will just reinstall JDK 1.7.
> JDK 1.8 build fails in TestTTFParser
> ------------------------------------
>
> Key: PDFBOX-2005
> URL: https://issues.apache.org/jira/browse/PDFBOX-2005
> Project: PDFBox
> Issue Type: Bug
> Affects Versions: 2.0.0
> Reporter: Tilman Hausherr
>
> I installed JDK 1.8 and now the build fails because one of the tests fail:
> Tests run: 1, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 0.278 sec <<<
> FAILURE!
> testPostTable(org.apache.pdfbox.pdmodel.font.TestTTFParser) Time elapsed:
> 0.155 sec <<< FAILURE!
> junit.framework.ComparisonFailure: expected:<parenright[bigg]> but
> was:<parenright[]>
> Did anybody else install JDK 1.8?
--
This message was sent by Atlassian JIRA
(v6.2#6252)
