[
https://issues.apache.org/jira/browse/LUCENE-3071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13029471#comment-13029471
]
Robert Muir commented on LUCENE-3071:
-------------------------------------
Hi Olivier, at a glance the patch looks really great to me, thanks!
I took a quick look (not in enough detail) but had these thoughts, neither of
which I think are really mandatory for this feature, just ideas:
* do you think it would be cleaner if we made it a separate tokenizer? (e.g.
ReversePath....). The main logic of the tokenizer seems to be completely split,
depending on whether you are reversing or not.
* i think its possible in the future we could simplify the way finalOffset is
being tracked, such that we just accumulate it on every read(), and then
correctOffset a single time in end(). (i don't think this has much to do with
your patch, just looking at the code in general).
> PathHierarchyTokenizer adaptation for urls: splits reversed
> -----------------------------------------------------------
>
> Key: LUCENE-3071
> URL: https://issues.apache.org/jira/browse/LUCENE-3071
> Project: Lucene - Java
> Issue Type: New Feature
> Components: contrib/analyzers
> Reporter: Olivier Favre
> Priority: Minor
> Attachments: LUCENE-3071.patch, LUCENE-3071.patch, ant.log.tar.bz2
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> {{PathHierarchyTokenizer}} should be usable to split urls the a "reversed"
> way (useful for faceted search against urls):
> {{www.site.com}} -> {{www.site.com, site.com, com}}
> Moreover, it should be able to skip a given number of first (or last, if
> reversed) tokens:
> {{/usr/share/doc/somesoftware/INTERESTING/PART}}
> Should give with 4 tokens skipped:
> {{INTERESTING}}
> {{INTERESTING/PART}}
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
