Newest at the top
| 2026-03-07 20:41:58 +0100 | <monochrom> | The Forbidden GHC Sutra |
| 2026-03-07 20:37:45 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 20:29:01 +0100 | Sgeo | (~Sgeo@user/sgeo) Sgeo |
| 2026-03-07 20:26:51 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 255 seconds) |
| 2026-03-07 20:21:58 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 20:21:48 +0100 | <haskellbridge> | <zb> yes! |
| 2026-03-07 20:16:28 +0100 | zb | (jobnbg@bot.kapsi.fi) zb |
| 2026-03-07 20:14:55 +0100 | <EvanR> | luckily O(15000) is really efficient |
| 2026-03-07 20:13:28 +0100 | DetourNe- | DetourNetworkUK |
| 2026-03-07 20:11:44 +0100 | DetourNetworkUK | (~DetourNet@user/DetourNetworkUK) (Read error: Connection reset by peer) |
| 2026-03-07 20:11:37 +0100 | elarks | (~elarks@user/yerrii) yerrii |
| 2026-03-07 20:11:17 +0100 | DetourNe- | (~DetourNet@user/DetourNetworkUK) DetourNetworkUK |
| 2026-03-07 20:11:07 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 264 seconds) |
| 2026-03-07 20:08:10 +0100 | stackdroid18 | (~stackdroi@user/stackdroid) stackdroid |
| 2026-03-07 20:06:12 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 20:01:32 +0100 | Square | (~Square@user/square) (Remote host closed the connection) |
| 2026-03-07 19:59:40 +0100 | <haskellbridge> | <sm> wizardry! |
| 2026-03-07 19:59:00 +0100 | <c_wraith> | some arcane bit of lore. (I think it's actually called the User Guide? Eh. Whatever.) |
| 2026-03-07 19:58:17 +0100 | <haskellbridge> | <sm> what is this "manual" you speak of... |
| 2026-03-07 19:57:25 +0100 | <c_wraith> | You still need to understand *why* you'd use it, but it still provides an incredible reference for what all the syntax does. |
| 2026-03-07 19:56:32 +0100 | <c_wraith> | perhaps shockingly, the ghc user's manual is actually where I learned most of what I know about template haskell. |
| 2026-03-07 19:55:10 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 245 seconds) |
| 2026-03-07 19:54:30 +0100 | tromp | (~textual@2001:1c00:3487:1b00:e975:d7be:a717:768f) |
| 2026-03-07 19:54:19 +0100 | <[exa]> | yeah I'm not sure there's any actual tutorial, it's kinda materialized pre-compilation computation, IMO very good for such usecases |
| 2026-03-07 19:50:23 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 19:49:53 +0100 | <haskellbridge> | <zb> (i’ll start from https://wiki.haskell.org/Template_Haskell) |
| 2026-03-07 19:49:45 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 245 seconds) |
| 2026-03-07 19:46:36 +0100 | <haskellbridge> | <zb> would you have any easy pointers to where i could read more about templatehaskell preprocessing? |
| 2026-03-07 19:45:22 +0100 | <[exa]> | also perhaps generating the alternatives in the process so that you don't have O(15k) of code |
| 2026-03-07 19:45:22 +0100 | <haskellbridge> | ... maximum number of iterations to n) |
| 2026-03-07 19:45:16 +0100 | <haskellbridge> | <zb> for the record: it works! or i think it does! though the compiling takes a couple of minutes, which is a bit inconvenient, and the first lookup i then did in my ghci session felt a bit slow, maybe like 0.5s. but then the next lookups don’t feel slow anymore. i did get a warning when compiling/loading the module: Pattern match checker exceeded (2000000) iterations in an equation for ‘g000’. (Use -fmax-pmcheck-iterations=n to set the... |
| 2026-03-07 19:45:06 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 19:44:55 +0100 | <[exa]> | zb: if you totally need some such thing to happen you might want to have some templatehaskell preprocessing step that chews the 15k alternatives into something more tangible for ghc |
| 2026-03-07 19:34:05 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 272 seconds) |
| 2026-03-07 19:33:05 +0100 | kadobanana | (~mud@user/kadoban) (Ping timeout: 245 seconds) |
| 2026-03-07 19:30:51 +0100 | tromp | (~textual@2001:1c00:3487:1b00:e975:d7be:a717:768f) (Quit: My iMac has gone to sleep. ZZZzzz…) |
| 2026-03-07 19:26:23 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 19:25:00 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 264 seconds) |
| 2026-03-07 19:24:19 +0100 | <monochrom> | Ah yeah, that. "f 1 2 3 = 4; f 1 3 7 = 2" is likely "f x y z = case x of 1 -> {cases for y and z under x=1}". But really, the benefit is negligible until you ask "can I have a function of 15k arguments?" >:) |
| 2026-03-07 19:23:14 +0100 | <c_wraith> | Though I guess that still has as many leaves as there are equations, so it's not really doing much besides eliminating redundant checks |
| 2026-03-07 19:22:05 +0100 | <c_wraith> | GHC *might* make it a trie at best - collapsing all the prefixes together. I wouldn't expect anything more clever than that. |
| 2026-03-07 19:21:08 +0100 | driib3180 | (~driib@176.57.184.141) (Client Quit) |
| 2026-03-07 19:17:46 +0100 | merijn | (~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn |
| 2026-03-07 19:17:43 +0100 | driib3180 | (~driib@176.57.184.141) driib |
| 2026-03-07 19:16:41 +0100 | <haskellbridge> | <zb> kind of expecting, really, since defining the (f 1)’s shouldn’t be too many lines! |
| 2026-03-07 19:15:17 +0100 | <haskellbridge> | <zb> ah... i was hoping for the compiler make it into some clever hash lookups or something :D |
| 2026-03-07 19:12:58 +0100 | CallipygousPepe | (~reuben@user/CallipygousPepe) (Quit: Konversation terminated!) |
| 2026-03-07 19:12:42 +0100 | <monochrom> | 15k lines of "f 1 2 3 = 4" is likely compiled to exhaustive search over 15k cases. |
| 2026-03-07 19:11:10 +0100 | <haskellbridge> | <zb> (this would be charset conversion tables, which i guess i would want to somehow compile into the module for efficiency. the specific module would define only the one 15k equations function) |
| 2026-03-07 19:10:49 +0100 | <EvanR> | the einstein equation for GR expands to like... a lot of equations |