2025/11/13

Newest at the top

2025-11-13 13:37:56 +0100 <kuribas`> [exa]: I have "Hashable key => Hashable (Wild key)"
2025-11-13 13:37:54 +0100 <[exa]> kuribas`: select everything with "foo" at position 1, everything with 3 at position 3, and intersect?
2025-11-13 13:37:26 +0100 <[exa]> kuribas`: how do you evaluate a wildcard on the hashmaps btw
2025-11-13 13:36:09 +0100 <kuribas`> Basically I have "data Wild a = Specific a | Wildcard", and map "Map (Wild key1, Wild key2, ...) val"
2025-11-13 13:35:41 +0100 <lucabtz> [exa] yeah i was speaking about something distinct from multimap
2025-11-13 13:35:28 +0100annamalai(~annamalai@157.33.249.99) (Ping timeout: 255 seconds)
2025-11-13 13:35:07 +0100qqe_(~qqq@185.54.21.203) (Quit: Lost terminal)
2025-11-13 13:34:42 +0100__monty__(~toonn@user/toonn) toonn
2025-11-13 13:34:20 +0100 <kuribas`> Storing * as NULL is a bit inelegant...
2025-11-13 13:33:00 +0100deptype_(~deptype@2406:b400:3a:73c2:ffa7:a8a3:7599:8304)
2025-11-13 13:32:39 +0100deptype_(~deptype@2406:b400:3a:73c2:4944:536a:6ac6:1483) (Remote host closed the connection)
2025-11-13 13:32:32 +0100tromp(~textual@2001:1c00:3487:1b00:7d:cf52:961a:9343)
2025-11-13 13:32:27 +0100merijn(~merijn@77.242.116.146) (Ping timeout: 252 seconds)
2025-11-13 13:30:30 +0100 <kuribas`> Assuming NULL means "star" here...
2025-11-13 13:30:13 +0100 <kuribas`> LIMIT 1
2025-11-13 13:30:07 +0100 <kuribas`> "SELECT val FROM my_patterns WHERE (string_key == "foo" OR string_key IS NULL) AND (..) ORDER BY (string_key, string_key2, int_key) ASC NULLS LAST"
2025-11-13 13:28:08 +0100tromp(~textual@2001:1c00:3487:1b00:7d:cf52:961a:9343) (Quit: My iMac has gone to sleep. ZZZzzz…)
2025-11-13 13:23:35 +0100 <kuribas`> merijn: how would you query my keys, like ("foo", *, 3) in sqlite? With a CTE?
2025-11-13 13:16:40 +0100ChanServ+v lambdabot
2025-11-13 13:16:40 +0100lambdabot(~lambdabot@haskell/bot/lambdabot) lambdabot
2025-11-13 13:14:58 +0100Googulator89Googulator
2025-11-13 13:12:50 +0100deptype_(~deptype@2406:b400:3a:73c2:4944:536a:6ac6:1483)
2025-11-13 13:12:37 +0100deptype_(~deptype@2406:b400:3a:73c2:fdd5:1805:7d92:70ad) (Remote host closed the connection)
2025-11-13 13:00:59 +0100lambdabot(~lambdabot@haskell/bot/lambdabot) (Remote host closed the connection)
2025-11-13 13:00:04 +0100Nachtgespenst(~user@user/siracusa) siracusa
2025-11-13 12:59:23 +0100merijn(~merijn@77.242.116.146) merijn
2025-11-13 12:54:28 +0100 <[exa]> also if you'd get the O(n) worst-case with this one, you'd be equivalently screwed with the hashmaps ("replace your hash")
2025-11-13 12:53:57 +0100 <[exa]> lucabtz: not really, multimaps can have many same keys and you select the whole range
2025-11-13 12:52:24 +0100deptype_(~deptype@2406:b400:3a:73c2:fdd5:1805:7d92:70ad)
2025-11-13 12:52:05 +0100deptype_(~deptype@2406:b400:3a:73c2:8206:7870:138f:c565) (Remote host closed the connection)
2025-11-13 12:42:47 +0100merijn(~merijn@77.242.116.146) (Ping timeout: 256 seconds)
2025-11-13 12:42:01 +0100biberu(~biberu@user/biberu) biberu
2025-11-13 12:36:58 +0100merijn(~merijn@77.242.116.146) merijn
2025-11-13 12:33:41 +0100xff0x(~xff0x@2405:6580:b080:900:7b8c:bddf:6c13:ed0c)
2025-11-13 12:32:15 +0100deptype_(~deptype@2406:b400:3a:73c2:8206:7870:138f:c565)
2025-11-13 12:32:02 +0100deptype_(~deptype@2406:b400:3a:73c2:d739:473:9e2d:bf26) (Remote host closed the connection)
2025-11-13 12:31:52 +0100 <lucabtz> and it still would be better space wise
2025-11-13 12:30:35 +0100xff0x(~xff0x@2405:6580:b080:900:7cd4:5734:7947:6d90) (Quit: xff0x)
2025-11-13 12:30:01 +0100Lycurgus(~juan@user/Lycurgus) Lycurgus
2025-11-13 12:29:04 +0100 <lucabtz> the worst case complexity would still be O(n), maybe the worst case would incredibly unlickely given the key space is so big though
2025-11-13 12:27:46 +0100 <lucabtz> but the tree would need to contain bins are leaves no?
2025-11-13 12:25:13 +0100trickard__trickard
2025-11-13 12:25:13 +0100trickard(~trickard@cpe-62-98-47-163.wireline.com.au) (Ping timeout: 264 seconds)
2025-11-13 12:24:39 +0100 <[exa]> but don't :D
2025-11-13 12:24:19 +0100trickard__(~trickard@cpe-62-98-47-163.wireline.com.au)
2025-11-13 12:24:18 +0100 <[exa]> technically you don't need short hashes because you don't need to minimize the hash table, so you can always have say 64b hash, so collisions won't hurt too much, and at worst you simply find the one key that you wanted from the tree range that the search returns
2025-11-13 12:24:08 +0100xff0x(~xff0x@2405:6580:b080:900:7cd4:5734:7947:6d90)
2025-11-13 12:24:06 +0100merijn(~merijn@77.242.116.146) (Ping timeout: 256 seconds)
2025-11-13 12:22:54 +0100 <[exa]> multimap™
2025-11-13 12:21:20 +0100 <lucabtz> [exa] yeah i didnt think of that, though it wouldnt work, what about hash collisions?