dbwong commented on PR #1458: URL: https://github.com/apache/phoenix/pull/1458#issuecomment-1178358986
Hi some high level questions first. So the use case where I've seen UUIDs used most is for fast unique client side generated, server side validated generation. In general, the is somewhat simple use case as you can use a uuid generator in a strin gor similar and attempt to push it. The part where Phoenix isn't great is guaranteeing uniqueness for this rowid on the server, there are hacky solutions I'm aware of to achieve this but due to use of UPSERT and lack of data coming back I think there is some difficulty differentiating a put from a update. Does this solve/describe the type of use cases you are interested in? If so I'm thinking a PK constraint as opposed to an actual data type might solve this better. Thoughts? Also maybe @tkhurana might chime in as I think he has good knowledge around some of the details of similar use cases. Note, for a fixed length type having 0x00 or 0xFF as part of the rowkey should be fine for indexing etc as far as i'm aware. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
