Continue sqlc migration #385
Conversation
This stack of pull requests is managed by Graphite. Learn more about stacking. |
| SELECT | ||
| * | ||
| FROM | ||
| group_messages | ||
| WHERE | ||
| group_id = @group_id | ||
| ORDER BY |
There was a problem hiding this comment.
I'm using this fancy case statement for sorting for now. It seems to work fine, although I'm a little worried about perf. Sometimes using a case statement will prevent the query planner from using indexes. Will try and test
There was a problem hiding this comment.
Curious about how you would do this. Would pgbench and tools that run thorough SQL transcripts be considered a practical approach?
neekolas
force-pushed
the
04-25-continue_sqlc_migration
branch
from
1163957 to
c2f0240
Compare
| SELECT | ||
| * | ||
| FROM | ||
| group_messages | ||
| WHERE | ||
| group_id = @group_id | ||
| ORDER BY |
There was a problem hiding this comment.
Curious about how you would do this. Would pgbench and tools that run thorough SQL transcripts be considered a practical approach?
|
@37ng pgbench is great. I was thinking just the simplest approach of running the query with and without the CASE statement through |
@neekolas That's pretty cool! |
|
@37ng did the test in Retool and the query plan looks identical |
tl;dr
What's next
No urgency at all on these, but at some point I'd like to get rid of the rest of Bun in the repo and remove the dependency. That would involve:
bun.Migratewith another migration tool and porting over our migrationsauthzdatabase withsqlcqueries (should be super easy, there's maybe 2 queries)