[
https://issues.apache.org/jira/browse/SOLR-8517?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15116530#comment-15116530
]
Joel Bernstein commented on SOLR-8517:
--------------------------------------
Working with this patch now. I made some changes to how numerics are handled. I
believe all numerics will be converted to longs or doubles during the json
de-serialization. To support SQL properly we need to support getInt, getFloat,
getShort, getByte along with getLong and getDouble. So I changed the Numeric
handling to cast to a Number and return the corresponding primitive type. If we
want to add some safeguards to protect against truncation and loss of precision
we can come back and do that in another ticket.
> Implement minimal set of get* methods in ResultSetImpl for column names
> -----------------------------------------------------------------------
>
> Key: SOLR-8517
> URL: https://issues.apache.org/jira/browse/SOLR-8517
> Project: Solr
> Issue Type: Sub-task
> Components: SolrJ
> Affects Versions: Trunk
> Reporter: Kevin Risden
> Attachments: SOLR-8517.patch, SOLR-8517.patch, SOLR-8517.patch,
> SOLR-8517.patch
>
>
> This is related to the ResultSetImpl for column indices but requires that
> more metadata be based back from the SQL handler in relation to column names.
> The SQL handler already knows about the column names and order but they
> aren't passed back to the client. SQL clients used the column names to
> display so this must be implemented for DBVisualizer to work properly.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
