00:31:29 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-4850-gd9e8576752 00:54:00 Monster database of master branch on crawl.develz.org updated to: 0.30-a0-347-gf58ed5266c 04:23:06 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-4850-gd9e8576752 07:28:10 -!- theWeaver is now known as ioPan 09:17:39 -!- mhcerri7 is now known as mhcerri 14:38:47 New branch created: persistent_db (2 commits) 13https://github.com/crawl/crawl/tree/persistent_db 14:38:48 03advil02 07[persistent_db] * 0.30-a0-348-g79de549745: feat: mark some potentially blocking functions 10(8 minutes ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/79de549745c6 14:38:48 03advil02 07[persistent_db] * 0.30-a0-349-gdc488dc59d: refactor: rewrite webtiles db code 10(7 minutes ago, 4 files, 486+ 310-) 13https://github.com/crawl/crawl/commit/dc488dc59d06 14:45:38 <06advil> I will probably merge that soon, but it's a pretty big change, so I wanted to put it in a branch for at least a little bit just in case someone else looks at it 15:07:16 -!- ioPan is now known as theWeaver 15:16:13 -!- ChaosWitch is now known as Stella 15:16:51 -!- Stella is now known as Stella[OotC 15:16:55 -!- Stella[OotC is now known as Stella[OotC] 15:23:59 <09gammafunk> hah, when designing beem/cerebot/et als use of the sqlite db, I decided to load the db completely into memory via an appropriate dict representation and then do any updates to the in-memory cache as well as the db, so data is only loaded once and so access doesn't involve any sqlite calls. It wasn't really clear to me that this was the wisest approach, and it did add some complexities. But part of the reason for doing this was that 15:24:00 webtiles use of sqlite was my reference point and it was obvious that webtiles' approach would be completely inappropriate for lookups to data that happen pretty frequently (e.g. with chat messages) 15:25:13 <09gammafunk> in theory you might run out of memory, but for all these bots the db size is just pretty small, since the number of users is in the thousands 15:25:48 <06advil> yeah, that seems very reasonable if the db is small and there is no chance of concurrent access 15:25:58 <06advil> (the latter being the biggest concern) 15:26:57 <06advil> I guess also if you don't need something like a rollback mechanism 15:27:03 <09gammafunk> right, yeah, no chance of that given everything I'm doing is one process with an asyncio event loop 15:28:08 <09gammafunk> unit tests, wow 15:28:18 <06advil> haha 15:28:32 <06advil> tbh I got tired of trying to test all that stuff over and over again through the ui 15:28:40 <06advil> though of course that still needs some testing too 15:28:46 <09gammafunk> excellent test password as well 15:28:57 <06advil> thx 15:29:18 <06advil> gotta make the db code less boring somehow 16:23:48 Unstable branch on underhound.eu updated to: 0.30-a0-347-gf58ed5266c (34)