Hi Menglei, It is fixed in IGNITE-5252 [1] and already merged to master, Fix will available from 2.1 release coming.
[1] https://issues.apache.org/jira/browse/IGNITE-5252 On Wed, Jul 19, 2017 at 12:14 AM, menglei <[email protected]> wrote: > Hi all, > > Currently we are heavily using the standard SQL query functionality > provided > by Ignite. We have a lot join across the cache and need to get many columns > from the query. > > But right now the interface of SQL resultset (QueryCursor in ignite) has > limited exposure of the API, which is an iterator of the rows and we can > only able to use the index to get the columns or the object we want. There > is no support for the column name based retrieve operation like JDBC. We > also don't want to use JDBC which is not fully supported yet. > > The interesting part is the QueryCursorEx and the QueryCursorImpl have the > method to get at least the metadata which we can make use of. But I am not > confident to use the QueryCursorEx (it may be changed) in all our queries > since the QueryCursor is the right API to use. > > Could someone give us some insight for how to get the metadata? > > Thanks, > Menglei > > > > -- > View this message in context: http://apache-ignite-users. > 70518.x6.nabble.com/Ignite-SQL-queries-s-metadata-tp15085.html > Sent from the Apache Ignite Users mailing list archive at Nabble.com. > -- Best regards, Andrey V. Mashenkov
