Conversation

Notices

1. pwm (pwm@crlf.ninja)'s status on Monday, 29-Apr-2024 05:31:54 JST pwm
   Have I mentioned that I HATE the pleroma db lately?
   •  (mint@ryona.agency)'s status on Monday, 29-Apr-2024 05:31:53 JST 

     in reply to
     @pwm Who doesn't?
   • Executive Vice President of Racism, Inc. (not_benis@cawfee.club)'s status on Monday, 29-Apr-2024 05:37:15 JST Executive Vice President of Racism, Inc.

     in reply to

     • 
     @mint @pwm my thoughts exactly
      likes this.
   • pwm (pwm@crlf.ninja)'s status on Monday, 29-Apr-2024 05:58:05 JST pwm

     in reply to

     • 
     @mint I guess I could manually create the index I need but sheesh
     query goes from instantaneous to 20s just from adding an order by in order to paginate on the client side
      likes this.
   •  (mint@ryona.agency)'s status on Monday, 29-Apr-2024 05:59:18 JST 

     in reply to
     @pwm Are you trying to optimize pleromer itself or making something separate that grabs data from its DB directly?
   • pwm (pwm@crlf.ninja)'s status on Monday, 29-Apr-2024 08:15:48 JST pwm

     in reply to

     • 
     @mint I was pulling post text out of the db for a separate application, see dsmc logs.
     Just yanking posts out is fast but not deterministic, which I need in order to paginate through query results in the client program, which I'm doing to not pull 64000 posts in one go.
     I'm not sure if I shouldn't just suck it up, technically I already did what I wanted, it was just too slow for me to be satisfied.
      likes this.