You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been poking at the sqlalchemy connections compiled cache. It seems to work just fine for normal queries, but anything that goes through executemany does not seem to get cached. My test case is bulk insert, which seems to be suspiciously slower than when using multiple cached single queries. After taking a look at aiomysql's saconnection, the executemany does not actually even implement the cache.
So, is this omission intentional or just a forgotten feature ? I can't think of a case where the caching would not work, but on the other hand, I'm a noob when it comes to *sql drivers :)
The text was updated successfully, but these errors were encountered:
GitMate.io thinks possibly related issues are #290 (Cannot make db -> python value conversion working with custom SA columns), #49 (create_pool can not work with async/await feature), #46 ('trans.rollback()' is not working correct), #158 (aiomysql enabled cache?), and #86 (Inconsistant results after insert using sa).
I've been poking at the sqlalchemy connections compiled cache. It seems to work just fine for normal queries, but anything that goes through executemany does not seem to get cached. My test case is bulk insert, which seems to be suspiciously slower than when using multiple cached single queries. After taking a look at aiomysql's saconnection, the executemany does not actually even implement the cache.
So, is this omission intentional or just a forgotten feature ? I can't think of a case where the caching would not work, but on the other hand, I'm a noob when it comes to *sql drivers :)
The text was updated successfully, but these errors were encountered: