You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Yury Gerzhedovich (Jira)" <ji...@apache.org> on 2023/11/16 07:55:00 UTC

[jira] [Assigned] (IGNITE-20813) Sql. Introduce InternalSqlRow as resulting row of QueryProcessor

     [ https://issues.apache.org/jira/browse/IGNITE-20813?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yury Gerzhedovich reassigned IGNITE-20813:
------------------------------------------

    Assignee: Yury Gerzhedovich

> Sql. Introduce InternalSqlRow as resulting row of QueryProcessor
> ----------------------------------------------------------------
>
>                 Key: IGNITE-20813
>                 URL: https://issues.apache.org/jira/browse/IGNITE-20813
>             Project: Ignite
>          Issue Type: Improvement
>          Components: sql
>            Reporter: Konstantin Orlov
>            Assignee: Yury Gerzhedovich
>            Priority: Major
>              Labels: ignite-3
>
> As for now, the method signature of 
> {{QueryProcessor#querySingleAsync}} looks like this:
> {code:java}
>     CompletableFuture<AsyncSqlCursor<List<Object>>> querySingleAsync(
>             SqlProperties properties,
>             IgniteTransactions transactions,
>             @Nullable InternalTransaction transaction,
>             String qry,
>             Object... params
>     );
> {code}
>  
> The problem is that client code may not need de-serialised row right here right now (for instance, thin client handler will send those rows down the channel anyway), yet current signature force us to do deserialisation from a storage format.
> Let's change the signature to return something that extends InternalTuple, but also provides a way to get column value by having only index of column of interest (e.g. {{@Nullable Object get(int idx)}}).



--
This message was sent by Atlassian Jira
(v8.20.10#820010)