Jackie-Jiang commented on PR #8557: URL: https://github.com/apache/pinot/pull/8557#issuecomment-1111289197
> > > > High level question: should we reuse the current SQL endpoint (HTTP GET), or introduce a new one. How do other systems handle it? > > > > > > > > > For most of the DB, there would be just only one endpoint for executing all statements. So I feel we should just extend the existing SQL endpoint. > > > > > > Agree that we should use one single endpoint to handle all statements. My concern is that the single SQL endpoint should probably be a `POST` request instead of the current `GET` request. I'd suggest adding a new `POST` endpoint for all statements, and keep the current `GET` endpoint which can only be used to query data > > GET endpoint is not used much, so far I think only pinot-admin Query command is still using GET api, all the other endpoint are using POST. > > For endpoint perspective, do you mean we should create a new POST endpoint, say `/sql/v2` for all SQL semantics and keep existing endpoint just for DQL? Oh, just realize we already have the POST endpoint. What I meant is to make POST (reusing the existing one) to accept all SQL semantics, and keep GET only accepting DQL. We may add a check in the GET endpoint to prevent user using GET to change the table. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org
