[
https://issues.apache.org/jira/browse/LUCENENET-101?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12530425
]
Karol Rybak commented on LUCENENET-101:
---------------------------------------
Sorry i should have been more specific. There is a problem with opening java
lucene indexes not with searching them. Java uses base 36 for naming segments
files and because lucene.net uses base 16 it's not capable of opening them.
I've made a quick fix, however I have used a class found on codeproject for
base conversion so i'm not sure i can submit that. Problematic conversions are
at Index/IndexFileNames line 86 and Index/SegmentInfos lines 165 and 257. Also
i have only tested reading indexes not writing them and reading in lucene java.
> Using incorrect base when opening index
> ---------------------------------------
>
> Key: LUCENENET-101
> URL: https://issues.apache.org/jira/browse/LUCENENET-101
> Project: Lucene.Net
> Issue Type: Bug
> Environment: Not relevant
> Reporter: Karol Rybak
>
> Lucene.Net is using base 16 for generation while Lucene java is using base
> 36, there is no simple workaround since .net Convert.ToInt64 only supports
> base of 2, 8, 10, or 16. A function to convert bases is required.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
