You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-dev@lucene.apache.org by "Yousef Ourabi (JIRA)" <ji...@apache.org> on 2007/10/04 18:44:50 UTC
[jira] Commented: (SOLR-373) Build a JDBC driver to access Solr
[ https://issues.apache.org/jira/browse/SOLR-373?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12532478 ]
Yousef Ourabi commented on SOLR-373:
------------------------------------
Thomas,
[ I'm not a solr-dev, these are just my thoughts / opinions ]
I think that's a really horrible idea. That would promote the idea that solr is a relational database, and not a search index (inverted index). Here are my
>> This would help people only knowing SQL syntax
I think the query syntax is actually simpler than sql, anyone can pick it up in an hour or so of playing around with the documentation.
>> In the first run I would not allow any joins.
That's the entire point of a relational database, see first paragraph, solr/lucene is not a relational database.
-Yousef
> Build a JDBC driver to access Solr
> ----------------------------------
>
> Key: SOLR-373
> URL: https://issues.apache.org/jira/browse/SOLR-373
> Project: Solr
> Issue Type: New Feature
> Components: search, update
> Reporter: Thomas Peuss
> Priority: Minor
>
> Hello!
> Would it be a good thing to have a JDBC driver to access Solr? This would help people only knowing SQL syntax. And you can use things like Hibernate on top of that. Maybe this is a good alternative to Solrj. In the first run I would not allow any joins.
> I think of a SQL syntax like
> SELECT * FROM SOLR WHERE name='Albert' OR name='Einstein'
> which gets translated to a Solr query like
> q=name:Albert OR name:Einstein&fl=*
> What do you think?
> CU
> Thomas
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.