| 2022-10-01 00:00:23 +0200 | beteigeuze | (~Thunderbi@2001:8a0:61b5:6101:f0c:e4e3:bfdc:91df) |
| 2022-10-01 00:02:33 +0200 | zero | (~z@user/zero) |
| 2022-10-01 00:02:54 +0200 | <[exa]> | monadic joke choice is pretty safe here tbh. :D |
| 2022-10-01 00:04:30 +0200 | L29Ah | (~L29Ah@wikipedia/L29Ah) (Ping timeout: 244 seconds) |
| 2022-10-01 00:05:05 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 268 seconds) |
| 2022-10-01 00:05:41 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 00:06:00 +0200 | <monochrom> | Oh heh. |
| 2022-10-01 00:06:16 +0200 | zzz | (~z@user/zero) (Ping timeout: 268 seconds) |
| 2022-10-01 00:10:31 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 260 seconds) |
| 2022-10-01 00:10:53 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 00:10:56 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 00:14:38 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 00:15:46 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 260 seconds) |
| 2022-10-01 00:16:21 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 260 seconds) |
| 2022-10-01 00:18:15 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 00:18:36 +0200 | thyriaen | (~thyriaen@2a02:8109:8340:686c:7383:e0e2:ad95:9fce) (Quit: Leaving) |
| 2022-10-01 00:23:41 +0200 | zmt00 | (~zmt00@user/zmt00) (Read error: Connection reset by peer) |
| 2022-10-01 00:23:42 +0200 | vorpuni | (~pvorp@2001:861:3881:c690:ec7f:c79:dd1b:f81a) |
| 2022-10-01 00:25:38 +0200 | zmt00 | (~zmt00@user/zmt00) |
| 2022-10-01 00:26:43 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 244 seconds) |
| 2022-10-01 00:28:22 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 00:31:59 +0200 | Midjak | (~Midjak@82.66.147.146) (Quit: This computer has gone to sleep) |
| 2022-10-01 00:32:45 +0200 | titibandit | (~titibandi@xdsl-212-8-150-57.nc.de) (Remote host closed the connection) |
| 2022-10-01 00:32:58 +0200 | michalz | (~michalz@185.246.207.221) (Remote host closed the connection) |
| 2022-10-01 00:35:31 +0200 | rosalind | (rosalind@id-194105.uxbridge.irccloud.com) (Quit: bbl) |
| 2022-10-01 00:37:34 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 244 seconds) |
| 2022-10-01 00:39:55 +0200 | zebrag | (~chris@user/zebrag) |
| 2022-10-01 00:41:18 +0200 | rosalind | (rosalind@id-194105.uxbridge.irccloud.com) |
| 2022-10-01 00:41:38 +0200 | ardell | (~ardell@user/ardell) (Quit: Leaving) |
| 2022-10-01 00:55:13 +0200 | gff_ | (~gff@user/gff) (Ping timeout: 246 seconds) |
| 2022-10-01 00:56:01 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 260 seconds) |
| 2022-10-01 00:57:26 +0200 | gff_ | (~gff@user/gff) |
| 2022-10-01 00:59:21 +0200 | vglfr | (~vglfr@145.224.100.164) |
| 2022-10-01 00:59:47 +0200 | Tuplanolla | (~Tuplanoll@91-159-69-34.elisa-laajakaista.fi) |
| 2022-10-01 00:59:49 +0200 | megaTherion | (~therion@unix.io) (Remote host closed the connection) |
| 2022-10-01 00:59:56 +0200 | bitdex | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 01:00:08 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 01:03:12 +0200 | megaTherion | (~therion@unix.io) |
| 2022-10-01 01:03:52 +0200 | edrx | (~Eduardo@2804:18:5805:83e:c9df:9b:886f:e903) |
| 2022-10-01 01:04:21 +0200 | <edrx> | hi all! |
| 2022-10-01 01:04:57 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 244 seconds) |
| 2022-10-01 01:05:50 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 01:05:51 +0200 | chexum | (~quassel@gateway/tor-sasl/chexum) (Remote host closed the connection) |
| 2022-10-01 01:06:07 +0200 | <edrx> | I am trying to translate the definitions of Functor, Applicative, and Monad into some kinds of diagrams that I am very familiar with... two questions: 1) what is the pronounciation of <*>? 2) is there a way to define <*> from (>>=)? how? |
| 2022-10-01 01:07:00 +0200 | <hpc> | @src ap |
| 2022-10-01 01:07:00 +0200 | <lambdabot> | ap = liftM2 id |
| 2022-10-01 01:07:07 +0200 | <hpc> | bah |
| 2022-10-01 01:07:21 +0200 | <ski> | well, i pronounce it `ap' .. for the other question, see ^ |
| 2022-10-01 01:07:24 +0200 | <hpc> | anyhoo, ap = (<*>), so see how that's defined |
| 2022-10-01 01:07:31 +0200 | <ski> | @src liftM2 |
| 2022-10-01 01:07:32 +0200 | <lambdabot> | liftM2 f m1 m2 = do |
| 2022-10-01 01:07:32 +0200 | <lambdabot> | x1 <- m1 |
| 2022-10-01 01:07:32 +0200 | <lambdabot> | x2 <- m2 |
| 2022-10-01 01:07:33 +0200 | <lambdabot> | return (f x1 x2) |
| 2022-10-01 01:08:49 +0200 | <ski> | liftM2 f m0 m1 = m0 >>= \x0 -> m1 >>= \x1 -> return (f x0 x1) |
| 2022-10-01 01:08:54 +0200 | <ski> | = m0 >>= \x0 -> (return . f x0 <$> m1) |
| 2022-10-01 01:09:17 +0200 | <ski> | so |
| 2022-10-01 01:09:31 +0200 | <edrx> | so that's a definition for <*>? |
| 2022-10-01 01:09:43 +0200 | chexum | (~quassel@gateway/tor-sasl/chexum) |
| 2022-10-01 01:09:47 +0200 | <edrx> | I'll need some minutes to type each subexpression of that |
| 2022-10-01 01:09:55 +0200 | <geekosaur> | use `id` for `f` |
| 2022-10-01 01:10:20 +0200 | <ski> | er, sorry, no `return' in the last line |
| 2022-10-01 01:10:26 +0200 | <ski> | = m0 >>= \x0 -> (f x0 <$> m1) |
| 2022-10-01 01:11:34 +0200 | <ski> | mf `ap` mx = mf >>= \f -> f <$> mx |
| 2022-10-01 01:11:35 +0200 | <ski> | = mf >>= (<$> mx) |
| 2022-10-01 01:14:44 +0200 | <ski> | (i used here that `f <$> mx ={synonym}= fmap f mx ={law}= liftM f mx ={definition}= mx >>= \x -> return (f x)' to simplify the `m1 >>= \x1 -> return (f x0 x1)' to `f x0 <$> m1') |
| 2022-10-01 01:15:48 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 01:17:01 +0200 | <edrx> | let me see |
| 2022-10-01 01:18:21 +0200 | <ski> | edrx : it's not uncommon, if you're making both an `Applicative', and a `Monad' instance, to do the former like `pure = return' and `(<*>) = ap', and then do the "real implementation" in `Monad', defining `return' and `(>>=)' (and maybe `(>>)', if you care enough / want something better than the default implementation) |
| 2022-10-01 01:19:21 +0200 | <ski> | similarly, if you're making an `Applicative' instance, you can make the `Functor' instance by `fmap = liftA' (`liftA' is defined in terms of `pure' and `(<*>)'). or, if you already also have a `Monad' instance, you could go directly for `fmap = liftM' |
| 2022-10-01 01:20:19 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 268 seconds) |
| 2022-10-01 01:21:12 +0200 | <ski> | iow, `liftA' and `liftM' are suitable default implementations for `fmap' (/ `(<$>)'), in case you have the "proper" implementation in `Applicative' or `Monad'. and similarly `ap' (and `return') is a suitable default implementation of `(<*>)' (and `pure'), if you're doing `Monad' |
| 2022-10-01 01:23:31 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 01:24:10 +0200 | <ski> | it would also be nice if one could implement `Monad' by defining `join' (and `fmap' in `Functor', possibly via `liftA'), and then `mx >>= k = join (k <$> mx)' .. but alas, there are currently technical difficulties to this (related to being able to derive `Functor',`Applicative',`Monad' instances for `newtypes', iirc) |
| 2022-10-01 01:24:50 +0200 | <ski> | (because sometimes, it's easier to think in terms of `join' than `(>>=)'/`(=<<)' (or even `(>=>)'/`(<=<)')) |
| 2022-10-01 01:28:53 +0200 | chomwitt | (~chomwitt@2a02:587:dc14:f500:3759:458b:6e75:9dca) (Ping timeout: 268 seconds) |
| 2022-10-01 01:29:19 +0200 | nckx | nckhexen |
| 2022-10-01 01:33:19 +0200 | <edrx> | are the symbols "={synonym}=", "={law}=", "={definition}=" common? I'm going to start using them! |
| 2022-10-01 01:33:29 +0200 | rosalind | (rosalind@id-194105.uxbridge.irccloud.com) (Quit: Yawn) |
| 2022-10-01 01:33:50 +0200 | rosalind | (rosalind@id-194105.uxbridge.irccloud.com) |
| 2022-10-01 01:34:53 +0200 | <edrx> | oops |
| 2022-10-01 01:35:08 +0200 | <edrx> | I think I've asked the wrong question =/ |
| 2022-10-01 01:35:50 +0200 | <edrx> | I was looking for a way to define <*> - not <$> - from return and (>>=)... |
| 2022-10-01 01:36:46 +0200 | <geekosaur> | you were shown the definiiton of `ap` which is <*>, thn how to derive it from >>= |
| 2022-10-01 01:37:03 +0200 | <edrx> | gaaah |
| 2022-10-01 01:37:08 +0200 | <edrx> | let me read that again, sorry |
| 2022-10-01 01:38:54 +0200 | rosalind | (rosalind@id-194105.uxbridge.irccloud.com) () |
| 2022-10-01 01:41:31 +0200 | edrx | (~Eduardo@2804:18:5805:83e:c9df:9b:886f:e903) (Ping timeout: 260 seconds) |
| 2022-10-01 01:47:19 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 244 seconds) |
| 2022-10-01 01:50:29 +0200 | <ski> | @tell edrx "are the symbols \"={synonym}=\", \"={law}=\", \"={definition}=\" common? I'm going to start using them!" -- well, `={...}=' is just a symbol i've been using (and others have used similar symbols before) for documenting the *reason*/*justification* for an equality holding, when doing equational reasoning (for reasoning about code, including refactoring code) |
| 2022-10-01 01:50:30 +0200 | <lambdabot> | Consider it noted. |
| 2022-10-01 01:51:28 +0200 | anthezium | (~anthezium@75.164.54.59) |
| 2022-10-01 01:53:31 +0200 | <anthezium> | hey folks, does anyone know what the blessed way to do things like map a closed type family over a type-level list promoted with DataKinds is? Is the functionality from e.g. https://hackage.haskell.org/package/type-list-0.5.0.0 duplicated in a non-deprecated place? |
| 2022-10-01 01:54:20 +0200 | <anthezium> | is there a way to promote term-level map from the prelude, or something? |
| 2022-10-01 01:56:55 +0200 | <ski> | @tell edrx "I was looking for a way to define <*> - not <$> - from return and (>>=)..." -- and i derived, from the definition `ap = liftM2 id', and the definitions of `liftM',`liftM2' (in terms of `do'-notation / `(>>=)' and `return'), (also using the definitions `(<$>) = fmap' and (implicitly) `id x = x'), and the laws `liftM = fmap' and `ap = (<*>)', an alternate definition of `(<*>)'/`ap' in terms |
| 2022-10-01 01:56:55 +0200 | <lambdabot> | Consider it noted. |
| 2022-10-01 01:57:01 +0200 | <ski> | of`(>>=)' and `(<$>)' |
| 2022-10-01 01:57:10 +0200 | <ski> | @tell edrx .. of `(>>=)' and `(<$>)' |
| 2022-10-01 01:57:11 +0200 | <lambdabot> | Consider it noted. |
| 2022-10-01 01:58:02 +0200 | jero98772 | (~jero98772@2800:484:1d80:d8ce:efcc:cbb3:7f2a:6dff) |
| 2022-10-01 01:58:37 +0200 | <moonsheep> | How many bytes can I receive from a TCP stream with the network package? |
| 2022-10-01 01:58:45 +0200 | <moonsheep> | The docs say that "Considering hardware and network realities, the maximum number of bytes to receive should be a small power of 2, e.g., 4096." |
| 2022-10-01 01:58:48 +0200 | L29Ah | (~L29Ah@wikipedia/L29Ah) |
| 2022-10-01 01:58:59 +0200 | <moonsheep> | Yet it is my understanding that TCP can split messages into separate packets and deliver them in order |
| 2022-10-01 01:59:21 +0200 | <moonsheep> | So do I have to manually receive 4k chunks and reassemble them? |
| 2022-10-01 02:00:01 +0200 | vulpine | ghoulpine |
| 2022-10-01 02:01:11 +0200 | <geekosaur> | no, TCP does that for you. you may be waiting a long time to receive a very large series of bytes, though |
| 2022-10-01 02:01:31 +0200 | <geekosaur> | retransmissions aren't instantaneous |
| 2022-10-01 02:05:07 +0200 | <moonsheep> | right, I don't need to receive many bytes anyway, just *exactly* however many I want |
| 2022-10-01 02:07:11 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) (Ping timeout: 260 seconds) |
| 2022-10-01 02:09:06 +0200 | greuben | (~greuben@user/greuben) |
| 2022-10-01 02:10:34 +0200 | zebrag | (~chris@user/zebrag) (Quit: Konversation terminated!) |
| 2022-10-01 02:11:25 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) |
| 2022-10-01 02:11:57 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) |
| 2022-10-01 02:11:59 +0200 | <geekosaur> | recvBuf should work in that case, for TCP sockets. (UDP's another matter, as it won't combine two packets together to fill the buffer) |
| 2022-10-01 02:12:34 +0200 | <geekosaur> | this is why the socket type matters |
| 2022-10-01 02:16:12 +0200 | <moonsheep> | yeah don't worry, I'm using TCP |
| 2022-10-01 02:16:42 +0200 | <moonsheep> | why should I use recvBuf instead of the ByteString recv though? |
| 2022-10-01 02:16:47 +0200 | <moonsheep> | is there any advantage? |
| 2022-10-01 02:18:01 +0200 | dr_merijn | (~dr_merijn@86-86-29-250.fixed.kpn.net) |
| 2022-10-01 02:20:24 +0200 | <geekosaur> | ByteString would be preferable because that's usually what you really want. Going through String for what is presumably binary data (since it's size delimited instead of e.g. line oriented) is dumb |
| 2022-10-01 02:20:41 +0200 | jargon | (~jargon@184.101.88.60) |
| 2022-10-01 02:20:45 +0200 | <geekosaur> | so ByteString and then `binary` to parse it |
| 2022-10-01 02:21:44 +0200 | <geekosaur> | (or attoparsec for complex packets) |
| 2022-10-01 02:22:42 +0200 | <moonsheep> | oh I'm using cereal |
| 2022-10-01 02:22:44 +0200 | <moonsheep> | is that a problem? |
| 2022-10-01 02:22:47 +0200 | <geekosaur> | `network` and its `recvBuf` predate wide use of ByteString |
| 2022-10-01 02:22:55 +0200 | dr_merijn | (~dr_merijn@86-86-29-250.fixed.kpn.net) (Ping timeout: 265 seconds) |
| 2022-10-01 02:23:19 +0200 | <geekosaur> | cereal's mostly obsolete since `binary` started supporting lazy ByteString, but if it's being kept up to date then go ahead |
| 2022-10-01 02:23:22 +0200 | archsinner | (~archsinne@2603-9000-8400-033b-0000-0000-0000-1df1.inf6.spectrum.com) |
| 2022-10-01 02:23:32 +0200 | <moonsheep> | oh I didn't know that |
| 2022-10-01 02:23:38 +0200 | <moonsheep> | I always used cereal as the default my whole life |
| 2022-10-01 02:23:42 +0200 | <moonsheep> | I guess the name's catchier |
| 2022-10-01 02:24:31 +0200 | <moonsheep> | btw, what's the best way to parse variable-length packets? |
| 2022-10-01 02:24:49 +0200 | <moonsheep> | use recv to get the first 4 bytes, decode them as an integer length, and recv as much as that? |
| 2022-10-01 02:27:04 +0200 | <geekosaur> | that's probably not optimal. recv in what chunks you can and queue them, then decode length, then decode data based on length. attoparsec is better than binary/cereal here because it can produce a partial parse if you need to wait for more data |
| 2022-10-01 02:27:38 +0200 | <moonsheep> | right, so I end up having to reassemble them manually anyway. |
| 2022-10-01 02:27:47 +0200 | <moonsheep> | I've always wondered if the stream abstraction was necessary for TCP |
| 2022-10-01 02:28:04 +0200 | <moonsheep> | why not still treat the connection as packets that arrive in order and without corruption? |
| 2022-10-01 02:28:05 +0200 | <geekosaur> | not really, the closest you get to that is telling attoparsec to resume a parse |
| 2022-10-01 02:28:20 +0200 | <moonsheep> | that always felt like a weird design decision for TCP honestly |
| 2022-10-01 02:28:35 +0200 | <moonsheep> | you nearly always end up concatting the packets in the stream anyway |
| 2022-10-01 02:28:47 +0200 | <moonsheep> | except for large file transfers I guess, that makes some sense |
| 2022-10-01 02:28:47 +0200 | <geekosaur> | which is what stream does |
| 2022-10-01 02:29:03 +0200 | <geekosaur> | so I'm not seeing your point |
| 2022-10-01 02:29:19 +0200 | <moonsheep> | well why treat the connection as a stream of bytes, as opposed to as a stream of packets? |
| 2022-10-01 02:29:34 +0200 | <moonsheep> | so you'd do send and recv in terms of packets, not bytes |
| 2022-10-01 02:29:41 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) (Ping timeout: 265 seconds) |
| 2022-10-01 02:29:48 +0200 | <moonsheep> | you need to build that abstraction back on top of TCP anyway, right? |
| 2022-10-01 02:30:18 +0200 | <geekosaur> | no? |
| 2022-10-01 02:30:30 +0200 | <geekosaur> | most TCP protocols are line oriented, not packet |
| 2022-10-01 02:30:43 +0200 | <moonsheep> | well I guess fair enough |
| 2022-10-01 02:30:57 +0200 | <moonsheep> | maybe mine is the exception not the rule |
| 2022-10-01 02:31:23 +0200 | <moonsheep> | it's a chat, so I do have to separate messages as different units |
| 2022-10-01 02:31:25 +0200 | <moonsheep> | or "packets" |
| 2022-10-01 02:31:46 +0200 | <geekosaur> | the sockets interface does support SOCK_SEQPACKET but I don't think TCP/IP does |
| 2022-10-01 02:31:50 +0200 | <geekosaur> | OSI does, I think |
| 2022-10-01 02:31:56 +0200 | <moonsheep> | oh alright |
| 2022-10-01 02:31:58 +0200 | <moonsheep> | anyway thanks for everything |
| 2022-10-01 02:35:30 +0200 | Tuplanolla | (~Tuplanoll@91-159-69-34.elisa-laajakaista.fi) (Quit: Leaving.) |
| 2022-10-01 02:36:33 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) |
| 2022-10-01 02:38:21 +0200 | moonsheep | (~user@user/moonsheep) (Quit: ERC 5.4 (IRC client for GNU Emacs 28.2)) |
| 2022-10-01 02:41:46 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) (Ping timeout: 265 seconds) |
| 2022-10-01 02:47:16 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) (Remote host closed the connection) |
| 2022-10-01 02:48:01 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) (Remote host closed the connection) |
| 2022-10-01 02:48:06 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) |
| 2022-10-01 02:55:44 +0200 | frost | (~frost@user/frost) |
| 2022-10-01 02:56:06 +0200 | neightchan | (~nate@98.45.169.16) (Quit: WeeChat 2.9) |
| 2022-10-01 02:59:21 +0200 | Katarushisu | (~Katarushi@cpc147790-finc20-2-0-cust502.4-2.cable.virginm.net) (Quit: Ping timeout (120 seconds)) |
| 2022-10-01 02:59:41 +0200 | jespada | (~jespada@cpc121060-nmal24-2-0-cust249.19-2.cable.virginm.net) (Ping timeout: 260 seconds) |
| 2022-10-01 02:59:59 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) |
| 2022-10-01 03:00:43 +0200 | Katarushisu | (~Katarushi@86.1.5.247) |
| 2022-10-01 03:02:55 +0200 | jespada | (~jespada@77.97.36.250) |
| 2022-10-01 03:03:11 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) |
| 2022-10-01 03:03:49 +0200 | LoupGris | Lycanthropy |
| 2022-10-01 03:04:23 +0200 | Lycanthropy | Haematophage |
| 2022-10-01 03:08:28 +0200 | jespada_ | (~jespada@cpc121060-nmal24-2-0-cust249.19-2.cable.virginm.net) |
| 2022-10-01 03:08:51 +0200 | jespada | (~jespada@77.97.36.250) (Ping timeout: 265 seconds) |
| 2022-10-01 03:08:51 +0200 | Katarushisu | (~Katarushi@86.1.5.247) (Ping timeout: 265 seconds) |
| 2022-10-01 03:08:59 +0200 | adanwan_ | (~adanwan@gateway/tor-sasl/adanwan) (Ping timeout: 258 seconds) |
| 2022-10-01 03:09:01 +0200 | Katarushisu | (~Katarushi@cpc147790-finc20-2-0-cust502.4-2.cable.virginm.net) |
| 2022-10-01 03:09:20 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 03:10:04 +0200 | wroathe | (~wroathe@206-55-188-8.fttp.usinternet.com) |
| 2022-10-01 03:10:05 +0200 | wroathe | (~wroathe@206-55-188-8.fttp.usinternet.com) (Changing host) |
| 2022-10-01 03:10:05 +0200 | wroathe | (~wroathe@user/wroathe) |
| 2022-10-01 03:10:31 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) |
| 2022-10-01 03:13:41 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 260 seconds) |
| 2022-10-01 03:19:24 +0200 | tcard_ | (~tcard@p1205018-ipngn11201hodogaya.kanagawa.ocn.ne.jp) |
| 2022-10-01 03:19:31 +0200 | gurkenglas | (~gurkengla@p548ac72e.dip0.t-ipconnect.de) (Ping timeout: 248 seconds) |
| 2022-10-01 03:20:06 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 03:21:16 +0200 | beteigeuze | (~Thunderbi@2001:8a0:61b5:6101:f0c:e4e3:bfdc:91df) (Ping timeout: 260 seconds) |
| 2022-10-01 03:22:23 +0200 | tcard | (~tcard@p696070-ipngn8501hodogaya.kanagawa.ocn.ne.jp) (Ping timeout: 265 seconds) |
| 2022-10-01 03:25:12 +0200 | tcard_ | (~tcard@p1205018-ipngn11201hodogaya.kanagawa.ocn.ne.jp) (Quit: Leaving) |
| 2022-10-01 03:26:43 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 265 seconds) |
| 2022-10-01 03:29:38 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 265 seconds) |
| 2022-10-01 03:29:56 +0200 | vorpuni | (~pvorp@2001:861:3881:c690:ec7f:c79:dd1b:f81a) (Remote host closed the connection) |
| 2022-10-01 03:35:19 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 03:35:45 +0200 | harveypwca | (~harveypwc@2601:246:c180:a570:3828:d8:e523:3f67) |
| 2022-10-01 03:43:38 +0200 | lyle | (~lyle@104.246.145.85) (Ping timeout: 265 seconds) |
| 2022-10-01 03:43:52 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) (Ping timeout: 258 seconds) |
| 2022-10-01 03:44:37 +0200 | TonyStone | (~TonyStone@2603-7080-8607-c36a-cc48-2eb7-9785-f03a.res6.spectrum.com) (Quit: Leaving) |
| 2022-10-01 03:44:55 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) |
| 2022-10-01 03:48:53 +0200 | archsinner | (~archsinne@2603-9000-8400-033b-0000-0000-0000-1df1.inf6.spectrum.com) (Quit: archsinner) |
| 2022-10-01 03:49:17 +0200 | archsinner | (~archsinne@2603-9000-8400-033b-0000-0000-0000-1df1.inf6.spectrum.com) |
| 2022-10-01 03:51:51 +0200 | talismanick | (~talismani@c-73-41-86-39.hsd1.ca.comcast.net) (Ping timeout: 265 seconds) |
| 2022-10-01 03:53:39 +0200 | TonyStone | (~TonyStone@cpe-74-76-51-197.nycap.res.rr.com) |
| 2022-10-01 03:56:41 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 265 seconds) |
| 2022-10-01 03:58:56 +0200 | bilegeek_ | (~bilegeek@10.sub-174-208-237.myvzw.com) |
| 2022-10-01 04:00:06 +0200 | razetime | (~quassel@117.254.34.169) |
| 2022-10-01 04:00:35 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 04:01:09 +0200 | bilegeek | (~bilegeek@2600:1008:b018:1dfc:44e1:aaa5:595f:8e17) (Ping timeout: 244 seconds) |
| 2022-10-01 04:01:14 +0200 | archsinner | (~archsinne@2603-9000-8400-033b-0000-0000-0000-1df1.inf6.spectrum.com) (Quit: archsinner) |
| 2022-10-01 04:04:14 +0200 | wroathe | (~wroathe@user/wroathe) (Ping timeout: 244 seconds) |
| 2022-10-01 04:05:01 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 260 seconds) |
| 2022-10-01 04:05:42 +0200 | anthezium | (~anthezium@75.164.54.59) (Ping timeout: 252 seconds) |
| 2022-10-01 04:07:11 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 04:08:34 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) (Remote host closed the connection) |
| 2022-10-01 04:08:51 +0200 | adanwan_ | (~adanwan@gateway/tor-sasl/adanwan) |
| 2022-10-01 04:11:01 +0200 | greuben | (~greuben@user/greuben) (Quit: WeeChat 3.6) |
| 2022-10-01 04:11:53 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 252 seconds) |
| 2022-10-01 04:13:00 +0200 | td_ | (~td@94.134.91.160) (Ping timeout: 264 seconds) |
| 2022-10-01 04:14:44 +0200 | td_ | (~td@94.134.91.190) |
| 2022-10-01 04:15:26 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 04:20:27 +0200 | edrx | (~Eduardo@2804:56c:d2d3:4800:cf7d:b421:4c3a:392e) |
| 2022-10-01 04:20:43 +0200 | <edrx> | @src ap |
| 2022-10-01 04:20:44 +0200 | <lambdabot> | ap = liftM2 id |
| 2022-10-01 04:21:29 +0200 | tcard | (~tcard@2400:4051:5801:7500:19ce:ed82:2ab7:90f9) |
| 2022-10-01 04:22:30 +0200 | <edrx> | question: where did lambdabot find that? a few hours ago I asked some questions here and people answered by asking things to the lambdabot, and I was able to locate most of the answers in the source code of the Prelude... but not this definition for ap =/ |
| 2022-10-01 04:25:49 +0200 | <monochrom> | Does liftM2 ($) help? |
| 2022-10-01 04:26:21 +0200 | <edrx> | in what sense? |
| 2022-10-01 04:27:23 +0200 | <edrx> | for understanding how to define <*> from monad methods? |
| 2022-10-01 04:30:46 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 04:31:26 +0200 | <jackdk> | `($) :: (a -> b) -> a -> b` unifies with `id :: c -> c` with `c ~ (a -> b)` (remember that `(->)` is right-associative) |
| 2022-10-01 04:31:48 +0200 | greuben | (~greuben@user/greuben) |
| 2022-10-01 04:35:01 +0200 | harveypwca | (~harveypwc@2601:246:c180:a570:3828:d8:e523:3f67) (Quit: Leaving) |
| 2022-10-01 04:35:22 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 265 seconds) |
| 2022-10-01 04:35:48 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) (Ping timeout: 268 seconds) |
| 2022-10-01 04:41:10 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 04:45:37 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 252 seconds) |
| 2022-10-01 04:46:50 +0200 | tcard | (~tcard@2400:4051:5801:7500:19ce:ed82:2ab7:90f9) (Quit: Leaving) |
| 2022-10-01 04:48:31 +0200 | tcard | (~tcard@2400:4051:5801:7500:19ce:ed82:2ab7:90f9) |
| 2022-10-01 04:55:16 +0200 | <edrx> | aaaaah |
| 2022-10-01 04:55:18 +0200 | <edrx> | :module +GHC.Base |
| 2022-10-01 04:55:18 +0200 | <edrx> | :t liftM2 |
| 2022-10-01 04:55:18 +0200 | <edrx> | :t liftM2 ($) |
| 2022-10-01 04:55:19 +0200 | <lambdabot> | Monad m => (a1 -> a2 -> r) -> m a1 -> m a2 -> m r |
| 2022-10-01 04:55:20 +0200 | <lambdabot> | Monad m => m (a -> r) -> m a -> m r |
| 2022-10-01 04:55:31 +0200 | machinedgod | (~machinedg@d198-53-218-113.abhsia.telus.net) (Quit: Lost terminal) |
| 2022-10-01 04:55:40 +0200 | <edrx> | lovely =) |
| 2022-10-01 05:07:51 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 05:14:04 +0200 | bilegeek_ | (~bilegeek@10.sub-174-208-237.myvzw.com) (Remote host closed the connection) |
| 2022-10-01 05:14:12 +0200 | <sm> | hey all.. is it possible to re-export one thing from module X, without also leaking orphan instances defined in X ? |
| 2022-10-01 05:14:26 +0200 | bilegeek_ | (~bilegeek@10.sub-174-208-237.myvzw.com) |
| 2022-10-01 05:14:43 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 05:15:16 +0200 | <sm> | eg: import Text.Megaparsec.Custom as X (customErrorBundlePretty) -- re-exporting X, don't want to re-export orphan instances |
| 2022-10-01 05:17:23 +0200 | <jackdk> | I would use newtypes instead of orphan instances and this is one reason why. |
| 2022-10-01 05:19:36 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 264 seconds) |
| 2022-10-01 05:23:27 +0200 | bilegeek_ | bilegeek |
| 2022-10-01 05:28:03 +0200 | xff0x | (~xff0x@2405:6580:b080:900:b7c1:3130:710b:2a7f) (Ping timeout: 248 seconds) |
| 2022-10-01 05:29:01 +0200 | <[Leary]> | sm: I think you can just put customWhatever straight in your export list? |
| 2022-10-01 05:29:26 +0200 | darkstardevx | (~darkstard@192.183.207.94) (Remote host closed the connection) |
| 2022-10-01 05:29:38 +0200 | <sm> | [Leary]: that sounds good, and should avoid re-exporting the orphan instances ? |
| 2022-10-01 05:30:11 +0200 | razetime | (~quassel@117.254.34.169) (Ping timeout: 260 seconds) |
| 2022-10-01 05:30:49 +0200 | darkstardevx | (~darkstard@192.183.207.94) |
| 2022-10-01 05:30:57 +0200 | <[Leary]> | I would imagine so, but I don't actually know. |
| 2022-10-01 05:33:36 +0200 | mmhat | (~mmh@p200300f1c7062333ee086bfffe095315.dip0.t-ipconnect.de) (Quit: WeeChat 3.6) |
| 2022-10-01 05:33:56 +0200 | Everything | (~Everythin@37.115.210.35) () |
| 2022-10-01 05:34:07 +0200 | <sm> | thanks, both |
| 2022-10-01 05:34:39 +0200 | jargon | (~jargon@184.101.88.60) (Remote host closed the connection) |
| 2022-10-01 05:36:41 +0200 | luffy | (~chenqisu1@183.217.201.220) |
| 2022-10-01 05:38:11 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 265 seconds) |
| 2022-10-01 05:40:12 +0200 | xff0x | (~xff0x@2405:6580:b080:900:b7c1:3130:710b:2a7f) |
| 2022-10-01 05:42:26 +0200 | darkstardevx | (~darkstard@192.183.207.94) (Ping timeout: 260 seconds) |
| 2022-10-01 05:42:48 +0200 | jpds1 | (~jpds@gateway/tor-sasl/jpds) (Remote host closed the connection) |
| 2022-10-01 05:43:29 +0200 | jpds1 | (~jpds@gateway/tor-sasl/jpds) |
| 2022-10-01 05:43:59 +0200 | luffy | (~chenqisu1@183.217.201.220) (Ping timeout: 265 seconds) |
| 2022-10-01 05:48:09 +0200 | <[Leary]> | sm: Haskell wiki says: "All instances defined in a module A are imported automatically when importing A, or importing any module that imports A, directly or indirectly." |
| 2022-10-01 05:48:19 +0200 | <[Leary]> | So it looks like I'm wrong; you can't escape the pollution. |
| 2022-10-01 05:48:44 +0200 | <sm> | ack |
| 2022-10-01 05:49:18 +0200 | jero98772 | (~jero98772@2800:484:1d80:d8ce:efcc:cbb3:7f2a:6dff) (Remote host closed the connection) |
| 2022-10-01 05:50:32 +0200 | jargon | (~jargon@184.101.88.60) |
| 2022-10-01 05:50:43 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 05:51:02 +0200 | jargon | (~jargon@184.101.88.60) (Remote host closed the connection) |
| 2022-10-01 05:51:19 +0200 | bilegeek_ | (~bilegeek@10.sub-174-208-237.myvzw.com) |
| 2022-10-01 05:52:25 +0200 | bilegeek | (~bilegeek@10.sub-174-208-237.myvzw.com) (Read error: Connection reset by peer) |
| 2022-10-01 05:53:25 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 05:54:58 +0200 | FinnElija | (~finn_elij@user/finn-elija/x-0085643) (Ping timeout: 258 seconds) |
| 2022-10-01 05:55:19 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 244 seconds) |
| 2022-10-01 05:56:39 +0200 | FinnElija | (~finn_elij@user/finn-elija/x-0085643) |
| 2022-10-01 05:56:42 +0200 | razetime | (~quassel@117.254.34.169) |
| 2022-10-01 05:56:48 +0200 | <EvanR> | instances are very global |
| 2022-10-01 05:57:32 +0200 | <EvanR> | it's like there can be only one instance for a given type in the whole ecosystem |
| 2022-10-01 06:01:17 +0200 | bilegeek_ | (~bilegeek@10.sub-174-208-237.myvzw.com) (Remote host closed the connection) |
| 2022-10-01 06:02:14 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 06:02:16 +0200 | bitdex | (~bitdex@gateway/tor-sasl/bitdex) (Remote host closed the connection) |
| 2022-10-01 06:08:53 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 06:09:22 +0200 | bilegeek | (~bilegeek@10.sub-174-208-237.myvzw.com) |
| 2022-10-01 06:13:37 +0200 | <hololeap> | I'm curious if there's an appreciable difference between `LogicT (State s) a` and `StateT s Logic a`. I just want to throw a counter on for each element in the Logic monad |
| 2022-10-01 06:13:51 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 268 seconds) |
| 2022-10-01 06:15:31 +0200 | <hololeap> | I could also use `zip [1..]` after observeAll |
| 2022-10-01 06:17:18 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 06:18:14 +0200 | dr_merijn | (~dr_merijn@86-86-29-250.fixed.kpn.net) |
| 2022-10-01 06:20:12 +0200 | edrx | (~Eduardo@2804:56c:d2d3:4800:cf7d:b421:4c3a:392e) (Ping timeout: 264 seconds) |
| 2022-10-01 06:22:25 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 252 seconds) |
| 2022-10-01 06:22:47 +0200 | dr_merijn | (~dr_merijn@86-86-29-250.fixed.kpn.net) (Ping timeout: 252 seconds) |
| 2022-10-01 06:30:31 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 246 seconds) |
| 2022-10-01 06:40:01 +0200 | <jackdk> | In fact, if you want to explicitly import instances and silence warnings, the idiom is to `import Foo ()` |
| 2022-10-01 06:41:51 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 06:43:43 +0200 | tmiller | (~tmiller@199.241.26.152) (Quit: WeeChat 3.6) |
| 2022-10-01 06:45:35 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 06:48:27 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 252 seconds) |
| 2022-10-01 06:49:28 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 06:52:26 +0200 | bontaq | (~user@ool-45779fe5.dyn.optonline.net) (Ping timeout: 260 seconds) |
| 2022-10-01 06:55:03 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 252 seconds) |
| 2022-10-01 06:55:52 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) (Remote host closed the connection) |
| 2022-10-01 06:56:56 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 07:06:06 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 244 seconds) |
| 2022-10-01 07:10:43 +0200 | gawen | (~gawen@user/gawen) (Quit: cya) |
| 2022-10-01 07:10:46 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 07:12:03 +0200 | gawen | (~gawen@user/gawen) |
| 2022-10-01 07:15:16 +0200 | frost | (~frost@user/frost) (Ping timeout: 252 seconds) |
| 2022-10-01 07:15:57 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 252 seconds) |
| 2022-10-01 07:17:56 +0200 | Lycurgus | (~juan@user/Lycurgus) |
| 2022-10-01 07:20:00 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 07:21:01 +0200 | Lycurgus | (~juan@user/Lycurgus) (Quit: Exeunt juan@acm.org) |
| 2022-10-01 07:23:22 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 07:23:58 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Remote host closed the connection) |
| 2022-10-01 07:24:23 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 252 seconds) |
| 2022-10-01 07:25:32 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 07:28:36 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 260 seconds) |
| 2022-10-01 07:34:06 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 07:39:06 +0200 | mvk | (~mvk@2607:fea8:5ce3:8500::778c) (Ping timeout: 260 seconds) |
| 2022-10-01 07:39:07 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 246 seconds) |
| 2022-10-01 07:42:58 +0200 | takuan | (~takuan@178-116-218-225.access.telenet.be) |
| 2022-10-01 07:43:04 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 07:44:11 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) (Ping timeout: 252 seconds) |
| 2022-10-01 07:47:19 +0200 | zxx7529 | (~Thunderbi@user/zxx7529) |
| 2022-10-01 07:49:10 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 265 seconds) |
| 2022-10-01 07:49:51 +0200 | k8yun | (~k8yun@user/k8yun) |
| 2022-10-01 08:02:04 +0200 | gmg | (~user@user/gehmehgeh) |
| 2022-10-01 08:02:09 +0200 | k8yun | (~k8yun@user/k8yun) (Ping timeout: 252 seconds) |
| 2022-10-01 08:02:58 +0200 | ksu | (~ksu@user/prtr) |
| 2022-10-01 08:03:25 +0200 | redmp | (~redmp@mobile-166-171-251-80.mycingular.net) |
| 2022-10-01 08:03:45 +0200 | redmp | (~redmp@mobile-166-171-251-80.mycingular.net) (Client Quit) |
| 2022-10-01 08:06:06 +0200 | redmp | (~redmp@mobile-166-171-251-80.mycingular.net) |
| 2022-10-01 08:09:21 +0200 | redmp | (~redmp@mobile-166-171-251-80.mycingular.net) (Client Quit) |
| 2022-10-01 08:09:50 +0200 | vorpuni | (~pvorp@2001:861:3881:c690:f9de:fbcb:2100:7597) |
| 2022-10-01 08:17:01 +0200 | coot | (~coot@2a02:a310:e241:1b00:ec1a:e9df:79ac:66ba) |
| 2022-10-01 08:20:31 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 260 seconds) |
| 2022-10-01 08:22:40 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 08:27:40 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 268 seconds) |
| 2022-10-01 08:32:59 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 08:33:10 +0200 | greuben | (~greuben@user/greuben) (Ping timeout: 265 seconds) |
| 2022-10-01 08:38:40 +0200 | <ksu> | is there a way to prevent _silently_ type defaulting integer constants to Int around functions like length or (!!)? |
| 2022-10-01 08:38:44 +0200 | <ksu> | > (2 ^ 81, 2 ^ 81 + length [], [0] !! (2 ^ 81)) |
| 2022-10-01 08:38:45 +0200 | <lambdabot> | (2417851639229258349412352,0,0) |
| 2022-10-01 08:39:49 +0200 | <ksu> | literals |
| 2022-10-01 08:43:39 +0200 | greuben | (~greuben@user/greuben) |
| 2022-10-01 08:44:56 +0200 | vglfr | (~vglfr@145.224.100.164) (Remote host closed the connection) |
| 2022-10-01 08:45:50 +0200 | vglfr | (~vglfr@145.224.100.164) |
| 2022-10-01 08:48:36 +0200 | Midjak | (~Midjak@82.66.147.146) |
| 2022-10-01 08:50:33 +0200 | vglfr | (~vglfr@145.224.100.164) (Remote host closed the connection) |
| 2022-10-01 08:51:13 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:b833:229b:a9d8:c346) (Remote host closed the connection) |
| 2022-10-01 08:52:17 +0200 | eggplantade | (~Eggplanta@104-55-37-220.lightspeed.sntcca.sbcglobal.net) |
| 2022-10-01 08:53:48 +0200 | vglfr | (~vglfr@145.224.100.164) |
| 2022-10-01 08:55:26 +0200 | <probie> | They're not defaulting, they're being "solved", since the types of `length` and `(!!)` involve `Int`s, and things like `(+)` require both arguments to have the same type |
| 2022-10-01 08:55:45 +0200 | talismanick | (~talismani@2601:200:c100:c9e0::24ac) |
| 2022-10-01 08:58:26 +0200 | paddymahoney | (~paddymaho@cpe9050ca207f83-cm9050ca207f80.cpe.net.cable.rogers.com) (Ping timeout: 260 seconds) |
| 2022-10-01 08:58:47 +0200 | jonathanx | (~jonathan@h-178-174-176-109.A357.priv.bahnhof.se) |
| 2022-10-01 08:58:53 +0200 | <probie> | :t length |
| 2022-10-01 08:58:54 +0200 | <lambdabot> | Foldable t => t a -> Int |
| 2022-10-01 08:58:58 +0200 | <probie> | :t (!!) |
| 2022-10-01 08:58:59 +0200 | <lambdabot> | [a] -> Int -> a |
| 2022-10-01 09:02:22 +0200 | <ksu> | ok, thanks |
| 2022-10-01 09:02:37 +0200 | <ksu> | so, is there a way to prevent it from silently happening? |
| 2022-10-01 09:02:40 +0200 | jonathanx_ | (~jonathan@h-178-174-176-109.A357.priv.bahnhof.se) |
| 2022-10-01 09:03:07 +0200 | jonathanx | (~jonathan@h-178-174-176-109.A357.priv.bahnhof.se) (Ping timeout: 246 seconds) |
| 2022-10-01 09:04:21 +0200 | eggplantade | (~Eggplanta@104-55-37-220.lightspeed.sntcca.sbcglobal.net) (Remote host closed the connection) |
| 2022-10-01 09:06:06 +0200 | bilegeek | (~bilegeek@10.sub-174-208-237.myvzw.com) (Remote host closed the connection) |
| 2022-10-01 09:06:28 +0200 | bilegeek | (~bilegeek@10.sub-174-208-237.myvzw.com) |
| 2022-10-01 09:13:08 +0200 | titibandit | (~titibandi@xdsl-212-8-150-57.nc.de) |
| 2022-10-01 09:14:06 +0200 | waldo | (~waldo@user/waldo) |
| 2022-10-01 09:16:35 +0200 | zxx7529 | (~Thunderbi@user/zxx7529) (Ping timeout: 252 seconds) |
| 2022-10-01 09:16:51 +0200 | jrm | (~jrm@user/jrm) (Quit: ciao) |
| 2022-10-01 09:17:10 +0200 | jrm | (~jrm@user/jrm) |
| 2022-10-01 09:21:46 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 260 seconds) |
| 2022-10-01 09:21:57 +0200 | zxx7529 | (~Thunderbi@user/zxx7529) |
| 2022-10-01 09:22:43 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 09:24:21 +0200 | Vajb | (~Vajb@2001:999:504:1841:9e47:1ec7:a52e:1d57) (Read error: Connection reset by peer) |
| 2022-10-01 09:25:02 +0200 | adanwan_ | (~adanwan@gateway/tor-sasl/adanwan) (Ping timeout: 258 seconds) |
| 2022-10-01 09:25:06 +0200 | lagash_ | (lagash@lagash.shelltalk.net) |
| 2022-10-01 09:25:33 +0200 | Vajb | (~Vajb@hag-jnsbng11-58c3a5-27.dhcp.inet.fi) |
| 2022-10-01 09:25:52 +0200 | lagash | (lagash@lagash.shelltalk.net) (Ping timeout: 260 seconds) |
| 2022-10-01 09:26:19 +0200 | vglfr | (~vglfr@145.224.100.164) (Remote host closed the connection) |
| 2022-10-01 09:26:35 +0200 | vglfr | (~vglfr@145.224.100.164) |
| 2022-10-01 09:26:49 +0200 | vglfr | (~vglfr@145.224.100.164) (Remote host closed the connection) |
| 2022-10-01 09:26:54 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) |
| 2022-10-01 09:27:21 +0200 | vglfr | (~vglfr@145.224.100.164) |
| 2022-10-01 09:34:41 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 09:39:36 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Ping timeout: 258 seconds) |
| 2022-10-01 09:39:49 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 268 seconds) |
| 2022-10-01 09:45:10 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 265 seconds) |
| 2022-10-01 09:45:55 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 09:48:14 +0200 | greuben | (~greuben@user/greuben) (Quit: WeeChat 3.6) |
| 2022-10-01 09:49:34 +0200 | gmg | (~user@user/gehmehgeh) (Ping timeout: 258 seconds) |
| 2022-10-01 09:49:58 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 09:52:04 +0200 | gmg | (~user@user/gehmehgeh) |
| 2022-10-01 09:52:17 +0200 | paddymahoney | (~paddymaho@cpe9050ca207f83-cm9050ca207f80.cpe.net.cable.rogers.com) |
| 2022-10-01 09:54:54 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 09:55:15 +0200 | elkcl | (~elkcl@broadband-37-110-156-162.ip.moscow.rt.ru) (Ping timeout: 248 seconds) |
| 2022-10-01 09:57:52 +0200 | elkcl | (~elkcl@broadband-37-110-156-162.ip.moscow.rt.ru) |
| 2022-10-01 09:58:24 +0200 | CiaoSen | (~Jura@p200300c957332e002a3a4dfffe84dbd5.dip0.t-ipconnect.de) |
| 2022-10-01 09:59:11 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 244 seconds) |
| 2022-10-01 09:59:37 +0200 | meinside | (uid24933@id-24933.helmsley.irccloud.com) |
| 2022-10-01 10:01:04 +0200 | tzh | (~tzh@c-24-21-73-154.hsd1.wa.comcast.net) (Quit: zzz) |
| 2022-10-01 10:01:24 +0200 | <probie> | If you turn on warnings (:set -Wall in GHCi), you'll get warned about 2 ^ 81 defaulting both 2 and 81 to `Integer`, but never about them becoming `Int` if they only valid type they could be is `Int` |
| 2022-10-01 10:01:31 +0200 | Guest9025 | (~Guest90@h-109-228-141-176.A980.priv.bahnhof.se) |
| 2022-10-01 10:03:18 +0200 | <tomjaguarpaw> | Can anyone explain to me why, if I'm in a checkout of this directory and I 'cabal build lib:trial' it tries to build 'trial-example'? https://github.com/kowainik/trial/tree/main/trial |
| 2022-10-01 10:03:27 +0200 | <tomjaguarpaw> | trial-example shouldn't be in the dependency tree at all |
| 2022-10-01 10:04:51 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:b833:229b:a9d8:c346) |
| 2022-10-01 10:05:11 +0200 | notzmv | (~zmv@user/notzmv) (Ping timeout: 250 seconds) |
| 2022-10-01 10:09:36 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:b833:229b:a9d8:c346) (Ping timeout: 260 seconds) |
| 2022-10-01 10:14:59 +0200 | bilegeek | (~bilegeek@10.sub-174-208-237.myvzw.com) (Quit: Leaving) |
| 2022-10-01 10:15:50 +0200 | Guest9025 | (~Guest90@h-109-228-141-176.A980.priv.bahnhof.se) (Quit: Client closed) |
| 2022-10-01 10:17:43 +0200 | ircbrowse_tom | (~ircbrowse@2a01:4f8:1c1c:9319::1) |
| 2022-10-01 10:17:44 +0200 | Server | +Cnt |
| 2022-10-01 10:18:33 +0200 | razetime | (~quassel@117.254.34.169) (Ping timeout: 252 seconds) |
| 2022-10-01 10:21:55 +0200 | ozkutuk | (~ozkutuk@176.240.173.153) (Ping timeout: 265 seconds) |
| 2022-10-01 10:30:43 +0200 | jmdaemon | (~jmdaemon@user/jmdaemon) (Ping timeout: 244 seconds) |
| 2022-10-01 10:35:08 +0200 | razetime | (~quassel@117.254.34.169) |
| 2022-10-01 10:35:28 +0200 | jmdaemon | (~jmdaemon@user/jmdaemon) |
| 2022-10-01 10:37:24 +0200 | Vajb | (~Vajb@hag-jnsbng11-58c3a5-27.dhcp.inet.fi) (Read error: Connection reset by peer) |
| 2022-10-01 10:38:32 +0200 | Vajb | (~Vajb@2001:999:504:1841:9e47:1ec7:a52e:1d57) |
| 2022-10-01 10:46:47 +0200 | raym | (~aritra@user/raym) (Ping timeout: 252 seconds) |
| 2022-10-01 11:00:10 +0200 | the-coot[m] | (~the-cootm@2001:470:69fc:105::95f) (Quit: You have been kicked for being idle) |
| 2022-10-01 11:06:26 +0200 | ozkutuk | (~ozkutuk@176.240.173.153) |
| 2022-10-01 11:06:58 +0200 | Tuplanolla | (~Tuplanoll@91-159-69-34.elisa-laajakaista.fi) |
| 2022-10-01 11:07:45 +0200 | raym | (~aritra@user/raym) |
| 2022-10-01 11:09:56 +0200 | L29Ah | (~L29Ah@wikipedia/L29Ah) (Read error: Connection reset by peer) |
| 2022-10-01 11:14:33 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) |
| 2022-10-01 11:18:42 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) (Remote host closed the connection) |
| 2022-10-01 11:20:39 +0200 | adanwan | (~adanwan@gateway/tor-sasl/adanwan) |
| 2022-10-01 11:21:28 +0200 | acidjnk | (~acidjnk@p200300d6e7137a85945ed0a823b800bd.dip0.t-ipconnect.de) |
| 2022-10-01 11:26:25 +0200 | wootehfoot | (~wootehfoo@user/wootehfoot) |
| 2022-10-01 11:28:37 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 265 seconds) |
| 2022-10-01 11:48:35 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 11:53:16 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 265 seconds) |
| 2022-10-01 11:55:16 +0200 | chomwitt | (~chomwitt@2a02:587:dc14:f500:f42c:42aa:75f3:5bbf) |
| 2022-10-01 11:57:53 +0200 | whatsupdoc | (uid509081@id-509081.hampstead.irccloud.com) (Quit: Connection closed for inactivity) |
| 2022-10-01 12:09:19 +0200 | meinside | (uid24933@id-24933.helmsley.irccloud.com) (Quit: Connection closed for inactivity) |
| 2022-10-01 12:14:17 +0200 | zxx7529 | (~Thunderbi@user/zxx7529) (Quit: zxx7529) |
| 2022-10-01 12:14:25 +0200 | notzmv | (~zmv@user/notzmv) |
| 2022-10-01 12:18:37 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) (Read error: Connection reset by peer) |
| 2022-10-01 12:21:07 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 12:23:59 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 12:25:47 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 252 seconds) |
| 2022-10-01 12:27:45 +0200 | jmdaemon | (~jmdaemon@user/jmdaemon) (Ping timeout: 250 seconds) |
| 2022-10-01 12:28:11 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 250 seconds) |
| 2022-10-01 12:28:17 +0200 | titibandit | (~titibandi@xdsl-212-8-150-57.nc.de) (Remote host closed the connection) |
| 2022-10-01 12:29:44 +0200 | titibandit | (~titibandi@xdsl-212-8-150-57.nc.de) |
| 2022-10-01 12:30:30 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 12:33:51 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) (Ping timeout: 252 seconds) |
| 2022-10-01 12:34:44 +0200 | mmhat | (~mmh@p200300f1c7062333ee086bfffe095315.dip0.t-ipconnect.de) |
| 2022-10-01 12:34:51 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 260 seconds) |
| 2022-10-01 12:42:51 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 12:43:09 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Remote host closed the connection) |
| 2022-10-01 12:47:07 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 244 seconds) |
| 2022-10-01 12:50:21 +0200 | <ski> | ksu : explicit type ascriptions ? |
| 2022-10-01 12:50:24 +0200 | <ski> | there's also |
| 2022-10-01 12:50:26 +0200 | <ski> | @type genericLength |
| 2022-10-01 12:50:27 +0200 | <lambdabot> | Num i => [a] -> i |
| 2022-10-01 12:50:30 +0200 | <ski> | @type genericIndex |
| 2022-10-01 12:50:32 +0200 | <lambdabot> | Integral i => [a] -> i -> a |
| 2022-10-01 12:55:00 +0200 | beteigeuze | (~Thunderbi@2001:8a0:61b5:6101:f0c:e4e3:bfdc:91df) |
| 2022-10-01 12:56:34 +0200 | CiaoSen | (~Jura@p200300c957332e002a3a4dfffe84dbd5.dip0.t-ipconnect.de) (Ping timeout: 246 seconds) |
| 2022-10-01 13:03:29 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) |
| 2022-10-01 13:04:12 +0200 | <ski> | hololeap : do you want backtrackable or non-backtrackable state ? |
| 2022-10-01 13:07:30 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:b833:229b:a9d8:c346) |
| 2022-10-01 13:08:11 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) (Ping timeout: 265 seconds) |
| 2022-10-01 13:08:46 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 13:11:55 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:b833:229b:a9d8:c346) (Ping timeout: 244 seconds) |
| 2022-10-01 13:13:30 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Remote host closed the connection) |
| 2022-10-01 13:19:55 +0200 | razetime | (~quassel@117.254.34.169) (Ping timeout: 268 seconds) |
| 2022-10-01 13:21:08 +0200 | zebrag | (~chris@user/zebrag) |
| 2022-10-01 13:21:18 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 13:25:52 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 244 seconds) |
| 2022-10-01 13:26:15 +0200 | perrierjouet | (~perrier-j@modemcable048.127-56-74.mc.videotron.ca) (Quit: WeeChat 3.6) |
| 2022-10-01 13:26:18 +0200 | razetime | (~quassel@117.254.34.169) |
| 2022-10-01 13:26:41 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 13:26:45 +0200 | perrierjouet | (~perrier-j@modemcable048.127-56-74.mc.videotron.ca) |
| 2022-10-01 13:32:48 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 264 seconds) |
| 2022-10-01 13:44:16 +0200 | lyle | (~lyle@104.246.145.85) |
| 2022-10-01 13:52:13 +0200 | talismanick | (~talismani@2601:200:c100:c9e0::24ac) (Ping timeout: 244 seconds) |
| 2022-10-01 14:02:35 +0200 | phma | (phma@2001:5b0:210b:aca8:ecff:405c:745f:81a2) (Read error: Connection reset by peer) |
| 2022-10-01 14:03:55 +0200 | phma | (~phma@2001:5b0:211b:e328:5893:8a41:b371:1ef) |
| 2022-10-01 14:11:00 +0200 | anon00 | (~anon@46.102.157.245) |
| 2022-10-01 14:23:33 +0200 | <fr33domlover[m]> | Laziness question: I have `f :: Stuff -> URI -> IO ()` which encodes `Stuff` to JSON and sends over HTTP. I'd like to send the same `Stuff` to many URIs. If I partially apply just `x :: Stuff` to `f`, will it compute the conversion to JSON just once and remember it for all runs of the IO action? Or do I need to manually write `g :: BL.ByteString -> URI -> IO()` to make sure conversion is done once in advance? |
| 2022-10-01 14:24:06 +0200 | <fr33domlover[m]> | (I know aeson has both strict and lazy versions, asking what the default for Haskell is) |
| 2022-10-01 14:25:46 +0200 | <hpc> | if you write it so the bytestring is outside the \uri -> whatever expression, it will remember it |
| 2022-10-01 14:27:18 +0200 | <hpc> | that's less of a laziness question and more of a how stuff is garbage-collected question too - you can generalize that if some subexpression only exists in a lambda, it gets forgotten when you leave it |
| 2022-10-01 14:27:50 +0200 | <hpc> | so \x -> \y -> foo (expensive x computation) will always forget |
| 2022-10-01 14:28:03 +0200 | <ski> | f stuff = \uri -> ..json..uri.. |
| 2022-10-01 14:28:05 +0200 | <ski> | where |
| 2022-10-01 14:28:15 +0200 | <ski> | json = stuffToJSON stuff |
| 2022-10-01 14:28:39 +0200 | <hpc> | \x -> let z = expensive x computation in \y -> foo z -- will remember z as long as the outer scope remembers (f x) |
| 2022-10-01 14:28:54 +0200 | <hpc> | this is how the memoization libraries work, but at the top level |
| 2022-10-01 14:29:22 +0200 | <hpc> | fibs = let fibslist = something in \n -> fibslist !! n |
| 2022-10-01 14:29:35 +0200 | <hpc> | fibslist will be remembered as long as fibs is, which at the top-level means forever |
| 2022-10-01 14:32:26 +0200 | <probie> | "which at the top-level means forever" - not so, top level terms can be GC-ed |
| 2022-10-01 14:33:29 +0200 | ellensol | (~ellen@178.78.210.152) |
| 2022-10-01 14:34:42 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) (Remote host closed the connection) |
| 2022-10-01 14:34:43 +0200 | califax | (~califax@user/califx) (Remote host closed the connection) |
| 2022-10-01 14:35:35 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) |
| 2022-10-01 14:35:36 +0200 | acidjnk | (~acidjnk@p200300d6e7137a85945ed0a823b800bd.dip0.t-ipconnect.de) (Ping timeout: 260 seconds) |
| 2022-10-01 14:36:19 +0200 | td_ | (~td@94.134.91.190) (Ping timeout: 252 seconds) |
| 2022-10-01 14:36:36 +0200 | <hpc> | probie: oh, neat |
| 2022-10-01 14:37:41 +0200 | ellensol | (~ellen@178.78.210.152) (Ping timeout: 244 seconds) |
| 2022-10-01 14:38:55 +0200 | califax | (~califax@user/califx) |
| 2022-10-01 14:40:38 +0200 | td_ | (~td@94.134.91.190) |
| 2022-10-01 14:41:13 +0200 | <fr33domlover[m]> | hpc: Thank you! |
| 2022-10-01 14:41:28 +0200 | L29Ah | (~L29Ah@wikipedia/L29Ah) |
| 2022-10-01 14:43:14 +0200 | <probie> | hpc: CAFs (top level thunks) are treated somewhat specially in that they can be GC-ed if there are no "live" references to them (at least that's my understanding - someone with a better knowledge of GHC internals can probably be more precise) |
| 2022-10-01 14:44:06 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 14:46:11 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 14:47:12 +0200 | <hpc> | ah, i will have to keep that in mind |
| 2022-10-01 14:47:59 +0200 | <hpc> | i try to avoid those sorts of things just because i don't like globals, but nice to know it's not a memory leak waiting to happen |
| 2022-10-01 14:48:59 +0200 | cdimitroulas | (~cdimitrou@194.32.120.166) |
| 2022-10-01 14:50:37 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 268 seconds) |
| 2022-10-01 14:50:50 +0200 | <ski> | probie : they revert to thunk state ? |
| 2022-10-01 14:52:01 +0200 | <geekosaur> | that's my recollection |
| 2022-10-01 14:52:39 +0200 | <geekosaur> | special support was added for that some ghc versions back |
| 2022-10-01 14:55:27 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) |
| 2022-10-01 14:55:36 +0200 | waldo | (~waldo@user/waldo) (Ping timeout: 264 seconds) |
| 2022-10-01 14:58:08 +0200 | waleee | (~waleee@h-176-10-137-138.NA.cust.bahnhof.se) |
| 2022-10-01 15:00:29 +0200 | ellensol_ | (~ellen@178-78-210-152.customers.ownit.se) |
| 2022-10-01 15:01:02 +0200 | zer0bitz | (~zer0bitz@2001:2003:f748:2000:493c:f2cd:d7ed:ecef) |
| 2022-10-01 15:03:05 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) (Ping timeout: 252 seconds) |
| 2022-10-01 15:05:25 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) |
| 2022-10-01 15:08:17 +0200 | <geekosaur> | not finding the original MR but https://gitlab.haskell.org/ghc/ghc/-/merge_requests/1232 implies it by existence |
| 2022-10-01 15:09:54 +0200 | <HP-UX> | Started to learn haskell |
| 2022-10-01 15:10:10 +0200 | <HP-UX> | Frist thing I'm reading now: https://andrew.gibiansky.com/blog/haskell/haskell-syntax/ as the syntax was always confusing |
| 2022-10-01 15:10:26 +0200 | <HP-UX> | Will consult the docs soon |
| 2022-10-01 15:10:55 +0200 | <fr33domlover[m]> | > \x -> \y -> foo (expensive x computation) will always forget |
| 2022-10-01 15:10:55 +0200 | <fr33domlover[m]> | hpc why does it always forget? You mean it will do the expensive computation every time and not memoize? I wonder why the compiler can't guess it can memoize because when I look at the code it seems like why not :P |
| 2022-10-01 15:10:57 +0200 | <lambdabot> | error: |
| 2022-10-01 15:10:57 +0200 | <lambdabot> | • Variable not in scope: foo :: t0 -> t2 -> t3 -> t4 -> t5 |
| 2022-10-01 15:10:57 +0200 | <lambdabot> | • Perhaps you meant ‘for’ (imported from Data.Traversable)error: |
| 2022-10-01 15:11:29 +0200 | <geekosaur> | Haskell syntax is actually fairly "regular" (in the sense that pretty much everything has the same form: <declaration> [where <definitions>]) |
| 2022-10-01 15:12:21 +0200 | <hpc> | fr33domlover[m]: same reason things get recomputed in any other language - it's only in the function body and the references to it all go away |
| 2022-10-01 15:12:36 +0200 | <ski> | fr33domlover[m] : each time you call the (inner) lambda (potentially getting a new value for `y'), `expensive x computation' will be evaluated anew |
| 2022-10-01 15:13:29 +0200 | <ski> | there is no automatic memoizing |
| 2022-10-01 15:13:51 +0200 | <ski> | HP-UX : feel free to ask any questions here |
| 2022-10-01 15:16:19 +0200 | <fr33domlover[m]> | ski: but if the expensive computation is pure and doesn't depend on y (e.g. JSON-encoding x), wouldn't the compiler want to optimize by memoizing the computation result just in case the outer lambda is used again (even with a different y)? |
| 2022-10-01 15:16:20 +0200 | causal | (~user@2001:470:ea0f:3:329c:23ff:fe3f:1e0e) (Quit: WeeChat 3.6) |
| 2022-10-01 15:16:51 +0200 | <geekosaur> | you end up wasting a lot of memory that way |
| 2022-10-01 15:16:52 +0200 | <ski> | that could cause space leaks |
| 2022-10-01 15:17:15 +0200 | <probie> | geekosaur: I think it was added around GHC 4 (or something of that vintage), but my memory might be playing tricks on me |
| 2022-10-01 15:17:39 +0200 | <geekosaur> | `reset CAFs` was a lot more recent. maybe we're talking about different things |
| 2022-10-01 15:18:29 +0200 | <ski> | (and you mean "sharing" or maybe "caching", not "memoizing". memoizing refers to having a function remember computed input-output associations, for use in later calls, checking if input has been memoized for the function already) |
| 2022-10-01 15:19:44 +0200 | <fr33domlover[m]> | ski, geekosaur : And if I rewrite `f` as `\x -> let j = encodeJSON x in \ y -> ...` why does it then cache the result of computing `j`? It seems like magic, but maybe means it only happens when I intend and not automatically in the original version of `f`? |
| 2022-10-01 15:19:51 +0200 | <geekosaur> | actually hm, ISTR the one I'm thinking of was related to ghci |
| 2022-10-01 15:20:27 +0200 | <geekosaur> | in general if you `let` something the compiler will take it as a request for sharing |
| 2022-10-01 15:20:43 +0200 | waldo | (~waldo@user/waldo) |
| 2022-10-01 15:21:12 +0200 | <geekosaur> | (but only a request. depending on how it modifies the code, that request may be ignored) |
| 2022-10-01 15:21:31 +0200 | <geekosaur> | that said, lets can be lifted, which increases sharing |
| 2022-10-01 15:21:57 +0200 | <ski> | fr33domlover[m] : it only caches across a single invocation `f x0'. iow if you do `g = f x0' or `map (f x0)' or the like. if you later do `f x1' (whether `x1' is equivalent or even pointer-equal to `x0' or not), then that will still recompute `encodeJSON x1' |
| 2022-10-01 15:25:27 +0200 | <fr33domlover[m]> | ski: So if I define `g = f x0`, it will memoize/share the result of `encodeJSON x0`? (and in the non-let version of `f` it won't?) That's good enough for me, was going to do the `g = f x0` thing ^\_^ |
| 2022-10-01 15:25:41 +0200 | <ski> | yes |
| 2022-10-01 15:26:54 +0200 | <ski> | (but also passing `f x0' as an argument, e.g. to `map' or whatever should be fine) |
| 2022-10-01 15:27:38 +0200 | <fr33domlover[m]> | Thank you :) Feels a bit weird to insert that peculiar `let` into the code and then have to add a comment to explain, because it's not clear that that's the purpose of it. But I suppose I can instead use meaningful function names to make it clear that there's a "prepare" step and "use" step, meant to allow the memoization |
| 2022-10-01 15:28:04 +0200 | <ski> | of course, you could also probably use `where' as i indicated above |
| 2022-10-01 15:28:16 +0200 | <fr33domlover[m]> | Years of haskelling, and somehow it wasn't clear to me until now "^_^ |
| 2022-10-01 15:29:14 +0200 | <ski> | this pattern is sometimes known as "serious (as opposed to trivial) currying", or "run-time compilation" |
| 2022-10-01 15:31:14 +0200 | <ski> | (a typical example would be converting a description of a regular expression into a function that employs an efficient state-based algorithm. so you want to translate from the description to the state machine, before the function accepts an input string to match on) |
| 2022-10-01 15:32:39 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 15:37:26 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 260 seconds) |
| 2022-10-01 15:47:13 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Ping timeout: 258 seconds) |
| 2022-10-01 15:47:34 +0200 | bontaq | (~user@ool-45779fe5.dyn.optonline.net) |
| 2022-10-01 15:48:50 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 15:50:47 +0200 | jakalx | (~jakalx@base.jakalx.net) () |
| 2022-10-01 15:55:24 +0200 | sinity | (~ezo.dev@193.200.83.122) |
| 2022-10-01 16:06:35 +0200 | lyle | (~lyle@104.246.145.85) (Ping timeout: 250 seconds) |
| 2022-10-01 16:12:24 +0200 | Guest|47 | (~Guest|47@1.156.128.208) |
| 2022-10-01 16:12:44 +0200 | Guest|47 | (~Guest|47@1.156.128.208) (Client Quit) |
| 2022-10-01 16:15:43 +0200 | lyle | (~lyle@104.246.145.85) |
| 2022-10-01 16:16:46 +0200 | k8yun | (~k8yun@user/k8yun) |
| 2022-10-01 16:22:38 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 16:22:45 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 16:25:11 +0200 | kenran | (~user@user/kenran) |
| 2022-10-01 16:27:26 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 268 seconds) |
| 2022-10-01 16:28:03 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 268 seconds) |
| 2022-10-01 16:28:41 +0200 | coot | (~coot@2a02:a310:e241:1b00:ec1a:e9df:79ac:66ba) (Quit: coot) |
| 2022-10-01 16:29:35 +0200 | [itchyjunk] | (~itchyjunk@user/itchyjunk/x-7353470) |
| 2022-10-01 16:30:37 +0200 | razetime | (~quassel@117.254.34.169) (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) |
| 2022-10-01 16:33:53 +0200 | waldo | (~waldo@user/waldo) (Ping timeout: 250 seconds) |
| 2022-10-01 16:37:46 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) (Remote host closed the connection) |
| 2022-10-01 16:41:41 +0200 | shapr | (~user@68.54.166.125) (Ping timeout: 244 seconds) |
| 2022-10-01 16:41:51 +0200 | jakalx | (~jakalx@base.jakalx.net) |
| 2022-10-01 16:42:51 +0200 | lyle | (~lyle@104.246.145.85) (Ping timeout: 268 seconds) |
| 2022-10-01 16:43:26 +0200 | jao | (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) |
| 2022-10-01 16:48:24 +0200 | ellensol_ | (~ellen@178-78-210-152.customers.ownit.se) (Ping timeout: 264 seconds) |
| 2022-10-01 16:50:04 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 16:51:49 +0200 | kenran` | (~user@user/kenran) |
| 2022-10-01 16:54:07 +0200 | k8yun | (~k8yun@user/k8yun) (Quit: Leaving) |
| 2022-10-01 16:54:33 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 252 seconds) |
| 2022-10-01 16:55:39 +0200 | lyle | (~lyle@104.246.145.85) |
| 2022-10-01 16:57:16 +0200 | kenran` | (~user@user/kenran) (Remote host closed the connection) |
| 2022-10-01 16:59:46 +0200 | lyle | (~lyle@104.246.145.85) (Ping timeout: 244 seconds) |
| 2022-10-01 16:59:53 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 17:02:14 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 17:04:30 +0200 | Guest2951 | (~Guest29@nmal-25-b2-v4wan-166405-cust169.vm24.cable.virginm.net) |
| 2022-10-01 17:04:43 +0200 | Guest2951 | (~Guest29@nmal-25-b2-v4wan-166405-cust169.vm24.cable.virginm.net) () |
| 2022-10-01 17:08:32 +0200 | mvk | (~mvk@2607:fea8:5ce3:8500::778c) |
| 2022-10-01 17:09:21 +0200 | lisbeths | (uid135845@id-135845.lymington.irccloud.com) |
| 2022-10-01 17:09:28 +0200 | waldo | (~waldo@user/waldo) |
| 2022-10-01 17:11:52 +0200 | machinedgod | (~machinedg@d198-53-218-113.abhsia.telus.net) |
| 2022-10-01 17:12:19 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) (Ping timeout: 258 seconds) |
| 2022-10-01 17:13:09 +0200 | ddellacosta | (~ddellacos@89.45.224.100) |
| 2022-10-01 17:14:09 +0200 | gurkenglas | (~gurkengla@p548ac72e.dip0.t-ipconnect.de) |
| 2022-10-01 17:15:08 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 17:16:32 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Ping timeout: 258 seconds) |
| 2022-10-01 17:16:58 +0200 | tzh | (~tzh@c-24-21-73-154.hsd1.wa.comcast.net) |
| 2022-10-01 17:17:39 +0200 | lyle | (~lyle@104.246.145.85) |
| 2022-10-01 17:24:44 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 17:25:23 +0200 | <fr33domlover[m]> | Cool, the example really helps :) |
| 2022-10-01 17:26:09 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) |
| 2022-10-01 17:26:48 +0200 | ddellacosta | (~ddellacos@89.45.224.100) (Ping timeout: 264 seconds) |
| 2022-10-01 17:31:57 +0200 | lyle | (~lyle@104.246.145.85) (Ping timeout: 252 seconds) |
| 2022-10-01 17:38:51 +0200 | cdimitroulas | (~cdimitrou@194.32.120.166) (Ping timeout: 265 seconds) |
| 2022-10-01 17:42:14 +0200 | bb010g | (~bb010g@2001:470:69fc:105::9a5) |
| 2022-10-01 17:46:02 +0200 | kenran | (~user@user/kenran) (Remote host closed the connection) |
| 2022-10-01 17:49:27 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) (Ping timeout: 268 seconds) |
| 2022-10-01 17:51:17 +0200 | jero98772 | (~jero98772@2800:484:1d80:d8ce:3490:26c5:1782:da8c) |
| 2022-10-01 17:51:36 +0200 | jespada_ | (~jespada@cpc121060-nmal24-2-0-cust249.19-2.cable.virginm.net) (Ping timeout: 260 seconds) |
| 2022-10-01 17:51:47 +0200 | mmhat | (~mmh@p200300f1c7062333ee086bfffe095315.dip0.t-ipconnect.de) (Quit: WeeChat 3.6) |
| 2022-10-01 17:52:58 +0200 | kimjetwav | (~user@2607:fea8:235e:b600:fded:756d:e09c:6c2d) (Remote host closed the connection) |
| 2022-10-01 17:54:58 +0200 | jespada | (~jespada@nmal-24-b2-v4wan-166357-cust1764.vm24.cable.virginm.net) |
| 2022-10-01 18:00:07 +0200 | Bulby[m] | (~bulbyvrma@2001:470:69fc:105::1:fe0a) (Quit: You have been kicked for being idle) |
| 2022-10-01 18:00:33 +0200 | mixphix | (~cigsender@bras-base-otwaon237cw-grc-11-174-91-129-69.dsl.bell.ca) |
| 2022-10-01 18:03:59 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 265 seconds) |
| 2022-10-01 18:05:12 +0200 | <hololeap> | ski, that's an interesting question. it doesn't matter to me in this case but I'm interested in what each would look like. |
| 2022-10-01 18:06:52 +0200 | __monty__ | (~toonn@user/toonn) |
| 2022-10-01 18:07:22 +0200 | wootehfoot | (~wootehfoo@user/wootehfoot) (Ping timeout: 265 seconds) |
| 2022-10-01 18:07:45 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) |
| 2022-10-01 18:21:43 +0200 | cyphase | (~cyphase@user/cyphase) (Ping timeout: 246 seconds) |
| 2022-10-01 18:23:05 +0200 | johnw | (~johnw@2600:1700:cf00:db0:dd60:bc61:a3c9:2dac) (Ping timeout: 250 seconds) |
| 2022-10-01 18:23:36 +0200 | cyphase | (~cyphase@user/cyphase) |
| 2022-10-01 18:24:24 +0200 | jargon | (~jargon@184.101.88.60) |
| 2022-10-01 18:29:07 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) (Ping timeout: 265 seconds) |
| 2022-10-01 18:29:34 +0200 | nitr0gen | (~nitr0gen@2409:4040:e8f:1ec2:6a7:469b:8f63:5372) |
| 2022-10-01 18:36:07 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:509f:aac5:1bc5:5910) |
| 2022-10-01 18:38:57 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Ping timeout: 258 seconds) |
| 2022-10-01 18:40:19 +0200 | kdaishi | (~Thunderbi@mobile-access-5672fd-223.dhcp.inet.fi) |
| 2022-10-01 18:40:58 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 18:41:37 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 18:43:36 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 264 seconds) |
| 2022-10-01 18:46:35 +0200 | jinsun__ | (~jinsun@user/jinsun) |
| 2022-10-01 18:46:35 +0200 | jinsun | Guest4802 |
| 2022-10-01 18:46:35 +0200 | jinsun__ | jinsun |
| 2022-10-01 18:48:04 +0200 | jinsun__ | (~jinsun@user/jinsun) |
| 2022-10-01 18:48:04 +0200 | jinsun | Guest8606 |
| 2022-10-01 18:48:04 +0200 | jinsun__ | jinsun |
| 2022-10-01 18:50:20 +0200 | Guest4802 | (~jinsun@user/jinsun) (Ping timeout: 244 seconds) |
| 2022-10-01 18:50:51 +0200 | Guest8606 | (~jinsun@user/jinsun) (Ping timeout: 244 seconds) |
| 2022-10-01 18:55:13 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Quit: leaving) |
| 2022-10-01 18:55:55 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 18:56:15 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 18:56:24 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Client Quit) |
| 2022-10-01 18:57:10 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 18:58:28 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Client Quit) |
| 2022-10-01 18:59:05 +0200 | acidjnk | (~acidjnk@p200300d6e7137a32f90b84488ac8cf0c.dip0.t-ipconnect.de) |
| 2022-10-01 19:02:56 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 19:03:03 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Client Quit) |
| 2022-10-01 19:03:26 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 265 seconds) |
| 2022-10-01 19:03:28 +0200 | <EvanR> | what is the rationale behind Show1 |
| 2022-10-01 19:03:28 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 19:03:49 +0200 | <EvanR> | I mean, intuitively, it involves being able to show * -> *'s, but on the surface that makes no sense |
| 2022-10-01 19:04:00 +0200 | <EvanR> | to me |
| 2022-10-01 19:05:03 +0200 | <geekosaur> | I understand it as more giving you a Show for an f a when you have only the f (with Show a being a prerequisite) |
| 2022-10-01 19:06:49 +0200 | kdaishi | (~Thunderbi@mobile-access-5672fd-223.dhcp.inet.fi) (Ping timeout: 265 seconds) |
| 2022-10-01 19:06:51 +0200 | <geekosaur> | or, if you prefer, an assertion that this f does not prevent showing f a |
| 2022-10-01 19:07:16 +0200 | <EvanR> | Show1 f => Show (Fix f) ... |
| 2022-10-01 19:07:19 +0200 | <EvanR> | instance |
| 2022-10-01 19:07:34 +0200 | <EvanR> | so the a is..... |
| 2022-10-01 19:07:45 +0200 | <geekosaur> | hrm |
| 2022-10-01 19:07:57 +0200 | geekosaur | doesn't understand Fix enough to say there |
| 2022-10-01 19:09:05 +0200 | jakalx | (~jakalx@base.jakalx.net) (Error from remote client) |
| 2022-10-01 19:09:40 +0200 | <geekosaur> | but it does seem to fit what I said, presuming Fix f can do something useful in some context |
| 2022-10-01 19:10:14 +0200 | <EvanR> | somehow it makes sense now. It lets you show * -> *'s |
| 2022-10-01 19:10:39 +0200 | <EvanR> | now about deriving it |
| 2022-10-01 19:10:53 +0200 | dmwit | (~dmwit@2600:4040:264a:df00:422e:910b:6593:d3fb) |
| 2022-10-01 19:12:15 +0200 | <dmwit> | Does anybody know what (if anything) I have to be careful about when passing hasktorch Tensors between threads? |
| 2022-10-01 19:14:18 +0200 | jakalx | (~jakalx@base.jakalx.net) |
| 2022-10-01 19:15:42 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 19:19:20 +0200 | Topsi | (~Topsi@dyndsl-095-033-018-041.ewe-ip-backbone.de) |
| 2022-10-01 19:21:50 +0200 | machinedgod | (~machinedg@d198-53-218-113.abhsia.telus.net) (Quit: Lost terminal) |
| 2022-10-01 19:22:43 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Quit: restarting computer, will come back soon) |
| 2022-10-01 19:23:50 +0200 | dmwit | (~dmwit@2600:4040:264a:df00:422e:910b:6593:d3fb) (Ping timeout: 252 seconds) |
| 2022-10-01 19:28:12 +0200 | pavonia | (~user@user/siracusa) (Quit: Bye!) |
| 2022-10-01 19:32:55 +0200 | Midjak | (~Midjak@82.66.147.146) (Ping timeout: 265 seconds) |
| 2022-10-01 19:35:11 +0200 | justsomeguy | (~justsomeg@user/justsomeguy) |
| 2022-10-01 19:36:29 +0200 | ChaiTRex | (~ChaiTRex@user/chaitrex) (Remote host closed the connection) |
| 2022-10-01 19:37:06 +0200 | ChaiTRex | (~ChaiTRex@user/chaitrex) |
| 2022-10-01 19:38:13 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 19:43:44 +0200 | <EvanR> | can't derive instance of Show1, it's not a stock class (Eq, Show, etc). Try enabling DeriveAnyClass |
| 2022-10-01 19:44:30 +0200 | <EvanR> | it produces an instance with no liftShowsPrec what's the point |
| 2022-10-01 19:44:53 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) (Ping timeout: 258 seconds) |
| 2022-10-01 19:45:06 +0200 | jinsun | (~jinsun@user/jinsun) (Read error: Connection reset by peer) |
| 2022-10-01 19:45:33 +0200 | <geekosaur> | I think you have to go via Generics to get any of the *1 classes |
| 2022-10-01 19:45:49 +0200 | <geekosaur> | although there might be a proposal for some of them |
| 2022-10-01 19:46:11 +0200 | jinsun | (~jinsun@user/jinsun) |
| 2022-10-01 19:49:41 +0200 | <geekosaur> | meh, don't see it. unless it's been accepted but not implemented yet |
| 2022-10-01 19:51:57 +0200 | leungbk | (~user@2603-8000-1201-2dd2-d72d-d70e-f2fb-2233.res6.spectrum.com) |
| 2022-10-01 19:56:45 +0200 | stiell_ | (~stiell@gateway/tor-sasl/stiell) |
| 2022-10-01 19:59:49 +0200 | <geekosaur> | don't see it there either, sigh |
| 2022-10-01 20:01:23 +0200 | mrianbloom | (uid350277@id-350277.ilkley.irccloud.com) (Quit: Connection closed for inactivity) |
| 2022-10-01 20:07:56 +0200 | <EvanR> | by the time I write the Show1 instance, it will be ready xD |
| 2022-10-01 20:08:41 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 260 seconds) |
| 2022-10-01 20:11:50 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:509f:aac5:1bc5:5910) (Remote host closed the connection) |
| 2022-10-01 20:13:57 +0200 | archsinner | (~archsinne@2603-9000-8708-30f2-0000-0000-0000-070f.inf6.spectrum.com) |
| 2022-10-01 20:14:59 +0200 | bontaq | (~user@ool-45779fe5.dyn.optonline.net) (Ping timeout: 268 seconds) |
| 2022-10-01 20:15:51 +0200 | eggplantade | (~Eggplanta@2600:1700:38c5:d800:509f:aac5:1bc5:5910) |
| 2022-10-01 20:19:57 +0200 | rockymarine | (~rocky@user/rockymarine) |
| 2022-10-01 20:29:28 +0200 | nitr0gen | (~nitr0gen@2409:4040:e8f:1ec2:6a7:469b:8f63:5372) (Ping timeout: 246 seconds) |
| 2022-10-01 20:34:24 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) |
| 2022-10-01 20:35:32 +0200 | jinsun | Guest7186 |
| 2022-10-01 20:35:32 +0200 | jinsun__ | (~jinsun@user/jinsun) |
| 2022-10-01 20:35:32 +0200 | Guest7186 | (~jinsun@user/jinsun) (Killed (molybdenum.libera.chat (Nickname regained by services))) |
| 2022-10-01 20:35:32 +0200 | jinsun__ | jinsun |
| 2022-10-01 20:43:07 +0200 | jinsun__ | (~jinsun@user/jinsun) |
| 2022-10-01 20:43:07 +0200 | jinsun | Guest9672 |
| 2022-10-01 20:43:07 +0200 | jinsun__ | jinsun |
| 2022-10-01 20:43:32 +0200 | justsomeguy | (~justsomeg@user/justsomeguy) (Quit: WeeChat 3.5) |
| 2022-10-01 20:44:00 +0200 | Guest9672 | (~jinsun@user/jinsun) (Ping timeout: 244 seconds) |
| 2022-10-01 20:44:00 +0200 | leungbk | (~user@2603-8000-1201-2dd2-d72d-d70e-f2fb-2233.res6.spectrum.com) (Ping timeout: 244 seconds) |
| 2022-10-01 20:45:49 +0200 | leungbk | (~user@cpe-142-129-149-172.socal.res.rr.com) |
| 2022-10-01 20:47:50 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) (Ping timeout: 265 seconds) |
| 2022-10-01 20:48:36 +0200 | nitr0gen | (~nitr0gen@2409:4040:e8f:1ec2:495b:6751:e4ab:2105) |
| 2022-10-01 20:53:21 +0200 | edrx | (~Eduardo@2804:18:580b:d1fe:c1d0:9f8d:e583:678c) |
| 2022-10-01 20:53:55 +0200 | cyphase | (~cyphase@user/cyphase) (Ping timeout: 248 seconds) |
| 2022-10-01 20:55:24 +0200 | cyphase | (~cyphase@user/cyphase) |
| 2022-10-01 20:56:53 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Quit: Be back in a sec) |
| 2022-10-01 20:57:59 +0200 | Lord_of_Life | (~Lord@user/lord-of-life/x-2819915) (Ping timeout: 265 seconds) |
| 2022-10-01 20:58:33 +0200 | Lord_of_Life | (~Lord@user/lord-of-life/x-2819915) |
| 2022-10-01 20:58:41 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 20:59:58 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Client Quit) |
| 2022-10-01 21:00:41 +0200 | docter_d | (~docter_d@2001:9e8:33e6:2e00:2dee:1861:4e3d:6a1d) |
| 2022-10-01 21:00:54 +0200 | sinity | (~ezo.dev@193.200.83.122) () |
| 2022-10-01 21:01:20 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 21:09:03 +0200 | <edrx> | hi all! I am trying to understand classes, instances, and inheritance, and I wrote this to have a minimalistic example to work on: |
| 2022-10-01 21:09:05 +0200 | <edrx> | http://angg.twu.net/HASKELL/Classes1.hs.html#class |
| 2022-10-01 21:09:43 +0200 | <edrx> | when I try ":load Classes1.hs" GHC yields this error: `m1' is not a (visible) method of class `C2' |
| 2022-10-01 21:10:43 +0200 | <edrx> | does that mean that if I make m1 visible in C2 I can override it? how do I make it visible? |
| 2022-10-01 21:11:07 +0200 | <geekosaur> | you're trying to use typeclasses as OOP. they aren't |
| 2022-10-01 21:11:20 +0200 | <edrx> | aaah |
| 2022-10-01 21:11:41 +0200 | <geekosaur> | % :info Ord |
| 2022-10-01 21:11:41 +0200 | <yahb2> | type Ord :: * -> Constraint ; class Eq a => Ord a where ; compare :: a -> a -> Ordering ; (<) :: a -> a -> Bool ; (<=) :: a -> a -> Bool ; (>) :: a -> a -> Bool ; (>=) :: a -> a -> Bool ;... |
| 2022-10-01 21:11:47 +0200 | <edrx> | yes, I don't have a good mental model for them yet... I'm trying to build it |
| 2022-10-01 21:11:51 +0200 | <hololeap> | can I pull in a library module as well as Main.hs when using runhaskell? |
| 2022-10-01 21:13:58 +0200 | <hololeap> | I'm having trouble with getting NoBuffering to stick since I split my Main.hs into an executable and a library, controlled by cabal |
| 2022-10-01 21:14:12 +0200 | raym | (~aritra@user/raym) (Ping timeout: 264 seconds) |
| 2022-10-01 21:14:19 +0200 | <hololeap> | so I would like to try again with runhaskell, but it needs to know about my library module |
| 2022-10-01 21:14:25 +0200 | <geekosaur> | https://paste.tomsmeding.com/WReEoZJb seems to work |
| 2022-10-01 21:15:32 +0200 | <geekosaur> | if it's a compiled library then you may need to use --ghc-arg=--package --ghc-arg=pkgnamehere |
| 2022-10-01 21:16:30 +0200 | <geekosaur> | or if you;re using a cabal library then cabal exec runhaskell … |
| 2022-10-01 21:16:56 +0200 | <hololeap> | ok, that works, great |
| 2022-10-01 21:18:06 +0200 | cyphase | (~cyphase@user/cyphase) (Ping timeout: 244 seconds) |
| 2022-10-01 21:18:22 +0200 | <geekosaur> | edrx, so a "superclass" is better thought of as a prerequisite. above I showed you (part of) how Ord is declared. it has Eq as a prerequisite, so it can derive other comparison operators using (==), (/=), and either compare or (<) |
| 2022-10-01 21:18:58 +0200 | <geekosaur> | you can't "override" things in the prerequisite, which is one reason the "superclass" terminology is bad |
| 2022-10-01 21:21:53 +0200 | <geekosaur> | it's more correct to say that you can *use* things from the prerequisite. you can't redefine them, though. |
| 2022-10-01 21:22:08 +0200 | <hololeap> | hm, yeah there is no buffering when I use runhaskell, but there is when the bin is compiled. I have `forM_ [stdout,stderr] (`hSetBuffering` NoBuffering)` at the top of my main function |
| 2022-10-01 21:22:44 +0200 | <geekosaur> | do you want to include stdin there? |
| 2022-10-01 21:23:10 +0200 | cyphase | (~cyphase@user/cyphase) |
| 2022-10-01 21:23:19 +0200 | coot | (~coot@213.134.165.79) |
| 2022-10-01 21:23:31 +0200 | <c_wraith> | I thought stderr defaulted to no buffering, too |
| 2022-10-01 21:24:21 +0200 | <geekosaur> | edrx, one place that demonstrates why treating typeclasses as OOP is bad: when we added Applicative as a prerequisite for Monad, many programs backformed it by saying `instance Applicative MyMonad where { pure = return; (<*>) = ap }` |
| 2022-10-01 21:24:49 +0200 | <geekosaur> | `return` is part of the Monad class, `ap` is defined for Monad instead of Applicative |
| 2022-10-01 21:25:01 +0200 | <geekosaur> | so it works backwards from what you'd expect from OOP |
| 2022-10-01 21:25:33 +0200 | <geekosaur> | we're defining the so-called "superclass" based on the "subclass" instead of the other way around! |
| 2022-10-01 21:26:03 +0200 | <edrx> | geekosaur: aah, I had just stumbled on that by googling... https://itecnote.com/tecnote/haskell-defining-a-new-monad-in-haskell-raises-no-instance-for-applic… |
| 2022-10-01 21:27:34 +0200 | kenran | (~user@user/kenran) |
| 2022-10-01 21:27:54 +0200 | <hololeap> | geekosaur: no, it's just supposed to use sourceProcessFromStreams from conduit-extra to move stdout/stderr from a subprocess directly to stdout/stderr for the main process |
| 2022-10-01 21:27:55 +0200 | <hololeap> | http://sprunge.us/Ht4OLD |
| 2022-10-01 21:28:14 +0200 | <geekosaur> | this does not, by the way, mean that the relationship is backwards; it means that it's not truly a superclass-subclass relationship but something entirely different |
| 2022-10-01 21:28:44 +0200 | lisbeths | (uid135845@id-135845.lymington.irccloud.com) (Quit: Connection closed for inactivity) |
| 2022-10-01 21:29:17 +0200 | <hololeap> | so you can run a process that displays a real-time percentage or something and have it displayed in real-time through the haskell binary |
| 2022-10-01 21:29:43 +0200 | <geekosaur> | did you also unbuffer the process's output? |
| 2022-10-01 21:29:48 +0200 | <hololeap> | the problem is, when I compile the binary, it waits until the subprocess completes before showing any of the stdout/stderr |
| 2022-10-01 21:30:01 +0200 | raym | (~aritra@user/raym) |
| 2022-10-01 21:30:27 +0200 | <hololeap> | hm, no I'm not sure where to get a reference to its handle if I'm using sourceProcessFromStreams, but that may be the answer |
| 2022-10-01 21:30:41 +0200 | <geekosaur> | (note that this does not mean calling hSetBuffering on the handle for the other process! it makes its own decision as to buffering, so you have to use GNU stdbuf or use ptys instead of pipes or etc.) |
| 2022-10-01 21:31:07 +0200 | kenran` | (~user@user/kenran) |
| 2022-10-01 21:31:09 +0200 | <hololeap> | oh, well it works correctly when I run it with `cabal exec runhaskell` |
| 2022-10-01 21:31:36 +0200 | <geekosaur> | that doesn't really sound like it'd affect the other process |
| 2022-10-01 21:32:06 +0200 | <geekosaur> | the forM_ looks correct to me so I don't know why it wouldn't be taking effect |
| 2022-10-01 21:32:26 +0200 | <geekosaur> | unless something does output before you run it in which case, well, oops |
| 2022-10-01 21:32:38 +0200 | <hololeap> | is there a flag to cabal/ghc to use the same buffering strategy that runhaskell uses? |
| 2022-10-01 21:32:53 +0200 | <geekosaur> | I don't think so |
| 2022-10-01 21:33:32 +0200 | <hololeap> | this might have something to do with whatever conduit is doing behind the scenes |
| 2022-10-01 21:33:57 +0200 | <edrx> | well, I was trying to build a minimal example of classes and instances because the standard examples - the ones from the Prelude - come with tons of extras, like laws... |
| 2022-10-01 21:34:59 +0200 | <geekosaur> | we prefer the ones with laws, yes. but that's just preference (try to figure out laws for Num and friends sometime) |
| 2022-10-01 21:36:14 +0200 | <geekosaur> | I think the usual recommendation is to study Functor and replicate it without looking at the source |
| 2022-10-01 21:37:16 +0200 | <geekosaur> | in practice, a class declaration declares a record that you can't see, and an instance declaration specifies a possible value for that record based on the type(s) the instance is declared for |
| 2022-10-01 21:38:10 +0200 | <geekosaur> | when you have a function with a constraint `(Foo a) => …`, Haskell quietly passes the appropriate record value based on what type `a` represents to that function |
| 2022-10-01 21:38:56 +0200 | fluxit- | (~fluxit@techsmix.net) (Quit: Bye!) |
| 2022-10-01 21:39:08 +0200 | <geekosaur> | when a class has a so-called "superclass", it means the instance record will (also) contain another instance record representing the appropriate "superclass" instance record |
| 2022-10-01 21:39:51 +0200 | <geekosaur> | this is why you can't redefine "superclass" methods in a "subclass": the instance can only pass on the "superclass" record, it can't modify it |
| 2022-10-01 21:40:26 +0200 | <geekosaur> | (I am describing how ghc does it. Some other Haskell compiler might do it differently, but at the moment the only real Haskell compiler is ghc.) |
| 2022-10-01 21:40:30 +0200 | kenran` | (~user@user/kenran) (Remote host closed the connection) |
| 2022-10-01 21:40:50 +0200 | nitr0gen | (~nitr0gen@2409:4040:e8f:1ec2:495b:6751:e4ab:2105) (Ping timeout: 244 seconds) |
| 2022-10-01 21:41:10 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Remote host closed the connection) |
| 2022-10-01 21:41:21 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) |
| 2022-10-01 21:41:34 +0200 | <geekosaur> | technically all it should have to pass is a type witness (that is, proof of the type it's being used with), but the simplest such witness is just the dictionary for that type |
| 2022-10-01 21:42:05 +0200 | <kenran> | geekosaur: thanks for the advice about not using named pipes and going the socket route instead. Just had to add one different effect interpreter, replace the read/write with socket operations and everthing immediately worked. No weird bugs anymore :) |
| 2022-10-01 21:42:19 +0200 | <geekosaur> | np |
| 2022-10-01 21:42:54 +0200 | <kenran> | I'd still love to know what's wrong with named pipes on macOS in Haskell, but alas... I can always dig up the program code and try it again another time. |
| 2022-10-01 21:43:23 +0200 | <edrx> | is it common to have a class that defines the types of some methods but not default values for those methods? I tried to do that in my minimal example and got a warning - or error? - that said that I had to use a pragma if I really wanted to do that... |
| 2022-10-01 21:43:34 +0200 | <geekosaur> | I'd be interested in knowing if FreeBSD has the same issue, since much of that code was taken from FreeBSD |
| 2022-10-01 21:44:20 +0200 | <geekosaur> | edrx, usually we don't specify defaults, in fact. Can you show an example of a declaration that raises an error? |
| 2022-10-01 21:44:24 +0200 | rockymarine | (~rocky@user/rockymarine) (Ping timeout: 268 seconds) |
| 2022-10-01 21:44:46 +0200 | <edrx> | sure, one min |
| 2022-10-01 21:44:57 +0200 | docter_d | (~docter_d@2001:9e8:33e6:2e00:2dee:1861:4e3d:6a1d) (Ping timeout: 268 seconds) |
| 2022-10-01 21:45:10 +0200 | <kenran> | geekosaur: oh, was it? Yeah, that'd be interesting as well. One more datapoint I have is that it seems to happen on x86 and arm macOS. |
| 2022-10-01 21:45:52 +0200 | <geekosaur> | that doesn't surprise me, the implementation of named pipes wouldn't be platform dependent beyond basics like page size which is probably the same for both |
| 2022-10-01 21:47:43 +0200 | acidjnk_new | (~acidjnk@p200300d6e7137a32a8f89e31b171299c.dip0.t-ipconnect.de) |
| 2022-10-01 21:47:48 +0200 | acidjnk | (~acidjnk@p200300d6e7137a32f90b84488ac8cf0c.dip0.t-ipconnect.de) (Ping timeout: 264 seconds) |
| 2022-10-01 21:49:52 +0200 | <EvanR> | edrx, often when you're thinking about default values, you're thinking about records of implementation/configuration instead of type classes |
| 2022-10-01 21:50:15 +0200 | fef | (~thedawn@user/thedawn) |
| 2022-10-01 21:50:28 +0200 | <EvanR> | where the record type defines the interface and particular record values define the implementation, and can be mixed and matched for purposes |
| 2022-10-01 21:50:38 +0200 | <edrx> | geekosaur: hm, I modified my test program and now I need to fix something else first... but your hint of "think that each type has a dictionary" has made many things clearer... |
| 2022-10-01 21:50:39 +0200 | <EvanR> | then you can create a default record and customize it |
| 2022-10-01 21:54:34 +0200 | tired | (~tired@user/tired) (Quit: /) |
| 2022-10-01 21:55:13 +0200 | tired | (~tired@user/tired) |
| 2022-10-01 21:56:02 +0200 | <edrx> | EvanR: you mean records in this sense? https://www.haskell.org/definition/haskell2010.pdf#page=46 |
| 2022-10-01 21:57:00 +0200 | <edrx> | (that looks like a good way to think about dictionaries of methods, btw) |
| 2022-10-01 21:57:21 +0200 | <EvanR> | data Employee = Employee { name :: String, salary :: Int } |
| 2022-10-01 21:57:27 +0200 | <edrx> | yep |
| 2022-10-01 21:58:01 +0200 | <EvanR> | data Driver m = Driver { putChar :: Char -> m (), getChar :: m Char } |
| 2022-10-01 22:01:14 +0200 | nate3 | (~nate@98.45.169.16) |
| 2022-10-01 22:01:26 +0200 | pavonia | (~user@user/siracusa) |
| 2022-10-01 22:02:34 +0200 | anon00 | (~anon@46.102.157.245) (Quit: Konversation terminated!) |
| 2022-10-01 22:04:03 +0200 | nitr0gen | (~nitr0gen@103.235.2.145) (Quit: Quit) |
| 2022-10-01 22:04:38 +0200 | <geekosaur> | I think it's actually a tuple in implementation, but you can think of a tuple as an anonymous record |
| 2022-10-01 22:05:10 +0200 | <geekosaur> | ("anonymous product type" if we're being all theoretical about it) |
| 2022-10-01 22:06:30 +0200 | tired | (~tired@user/tired) (Quit: /) |
| 2022-10-01 22:06:31 +0200 | nate3 | (~nate@98.45.169.16) (Ping timeout: 260 seconds) |
| 2022-10-01 22:07:27 +0200 | <edrx> | =) |
| 2022-10-01 22:07:54 +0200 | tired | (~tired@user/tired) |
| 2022-10-01 22:09:05 +0200 | Midjak | (~Midjak@82.66.147.146) |
| 2022-10-01 22:09:13 +0200 | ddellacosta | (~ddellacos@143.244.47.90) |
| 2022-10-01 22:10:37 +0200 | <edrx> | btw, this video about Core Haskell - <http://www.youtube.com/watch?v=fty9QL4aSRc>, "Vladislav Zavialov - Haskell to Core: Understanding Haskell Features Through Their Desugaring" has helped me a LOT |
| 2022-10-01 22:11:57 +0200 | <edrx> | it explains classes starting on 20:12. I need to watch that part again |
| 2022-10-01 22:12:04 +0200 | <geekosaur> | yes, it can be useful to trace a Haskell program through the desugarer to Core and then down to STG (which is even simpler) |
| 2022-10-01 22:12:18 +0200 | <edrx> | what is STG? |
| 2022-10-01 22:13:28 +0200 | <geekosaur> | it's the graph representing a Haskell program, which is simplified at runtime by the RTS's graph reduction machine |
| 2022-10-01 22:13:43 +0200 | <edrx> | ok! |
| 2022-10-01 22:13:45 +0200 | <geekosaur> | which is the simplest way to implement lazy evaluation |
| 2022-10-01 22:13:55 +0200 | <geekosaur> | you can look "graph reduction" up on wikipedia |
| 2022-10-01 22:14:34 +0200 | <geekosaur> | I do *not* recommend reading the STG papers for this; they're less about the why or the high level how than about the low level details of efficient graph reduction on a CPU |
| 2022-10-01 22:16:51 +0200 | archsinner | (~archsinne@2603-9000-8708-30f2-0000-0000-0000-070f.inf6.spectrum.com) (Quit: archsinner) |
| 2022-10-01 22:17:09 +0200 | <sclv> | re graph reduction this video may be of interest as well https://www.youtube.com/watch?v=UPy7TXgrK1A |
| 2022-10-01 22:17:15 +0200 | waldo | (~waldo@user/waldo) (Quit: quit) |
| 2022-10-01 22:17:25 +0200 | <geekosaur> | anyway the STG representation of a program is basically case and let |
| 2022-10-01 22:17:40 +0200 | <geekosaur> | everything else has gone away |
| 2022-10-01 22:17:54 +0200 | archsinner | (~archsinne@35.136.124.156) |
| 2022-10-01 22:18:16 +0200 | <geekosaur> | (or has become a value) |
| 2022-10-01 22:18:59 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) (Ping timeout: 258 seconds) |
| 2022-10-01 22:19:31 +0200 | <edrx> | sclv: thanks! |
| 2022-10-01 22:21:08 +0200 | bitdex_ | (~bitdex@gateway/tor-sasl/bitdex) |
| 2022-10-01 22:22:02 +0200 | <geekosaur> | there's also |
| 2022-10-01 22:22:06 +0200 | <geekosaur> | @hackage stgi |
| 2022-10-01 22:22:06 +0200 | <lambdabot> | https://hackage.haskell.org/package/stgi |
| 2022-10-01 22:22:17 +0200 | <edrx> | I was going to mention that! https://wiki.haskell.org/Ministg |
| 2022-10-01 22:22:37 +0200 | <geekosaur> | which you can use to visually trace reduction of a graph which is a simplified form of the one ghc uses |
| 2022-10-01 22:22:42 +0200 | <edrx> | it looks nice, but it is still a bit above my level... |
| 2022-10-01 22:22:51 +0200 | <edrx> | (or a lot above =/) |
| 2022-10-01 22:22:51 +0200 | leungbk | (~user@cpe-142-129-149-172.socal.res.rr.com) (Ping timeout: 260 seconds) |
| 2022-10-01 22:23:36 +0200 | <geekosaur> | yeh, at the very least make sure you understand the Core level, then start poking at how the Core representation becomes the STG representation |
| 2022-10-01 22:23:50 +0200 | <geekosaur> | then you can use stgi to see how that is evaluated |
| 2022-10-01 22:24:21 +0200 | <geekosaur> | sadly it's not quite complete enough that someone could hook it into ghc and produce a visual trace of actual ghc code |
| 2022-10-01 22:31:15 +0200 | fef | (~thedawn@user/thedawn) (Ping timeout: 258 seconds) |
| 2022-10-01 22:31:37 +0200 | <edrx> | gosh =( this - https://0x0.st/o4DR.txt - gives me this error: No instance for (C1 DD2) arising from the superclasses of an instance declaration in the instance declaration for `C2 DD2' |
| 2022-10-01 22:32:54 +0200 | <edrx> | building a good mental model of how classes work is turning out much harder than I expected ☹️ |
| 2022-10-01 22:33:12 +0200 | <geekosaur> | remember that I called it a prerequisite instead of a superclass |
| 2022-10-01 22:33:36 +0200 | <geekosaur> | you have to declare the instance for C1 before you can declare the instance for C2 |
| 2022-10-01 22:33:44 +0200 | <edrx> | let me try |
| 2022-10-01 22:33:46 +0200 | michalz | (~michalz@185.246.207.203) |
| 2022-10-01 22:33:48 +0200 | <geekosaur> | it does *not* do that for you |
| 2022-10-01 22:34:10 +0200 | archsinner | (~archsinne@35.136.124.156) (Quit: archsinner) |
| 2022-10-01 22:34:59 +0200 | <edrx> | sorry, can you explain? my "instance C1 DD1" is two lines above "instance C2 DD2"... |
| 2022-10-01 22:35:08 +0200 | archsinner | (~archsinne@2603-9000-8708-30f2-0000-0000-0000-070f.inf6.spectrum.com) |
| 2022-10-01 22:35:30 +0200 | <geekosaur> | but do you have an `instance C1 DD2`? |
| 2022-10-01 22:35:39 +0200 | <edrx> | aaah! trying! |
| 2022-10-01 22:36:07 +0200 | <edrx> | works! thanks! =) |
| 2022-10-01 22:37:59 +0200 | <edrx> | wait =( |
| 2022-10-01 22:38:06 +0200 | <edrx> | if I need these two lines: |
| 2022-10-01 22:38:08 +0200 | <edrx> | instance C1 DD2 |
| 2022-10-01 22:38:08 +0200 | <edrx> | instance C2 DD2 |
| 2022-10-01 22:38:36 +0200 | <edrx> | then what is the role of the "C1 a =>" in "class C1 a => C2 a where ..."? |
| 2022-10-01 22:39:33 +0200 | archsinner | (~archsinne@2603-9000-8708-30f2-0000-0000-0000-070f.inf6.spectrum.com) (Client Quit) |
| 2022-10-01 22:39:34 +0200 | <geekosaur> | "you must have a C1 instance in order to have a C2 instance" |
| 2022-10-01 22:39:49 +0200 | <geekosaur> | which is why I called it prerequisite instead of superclass |
| 2022-10-01 22:39:58 +0200 | <edrx> | aaaaahhhhh!!!! |
| 2022-10-01 22:42:16 +0200 | <geekosaur> | I'll refer you back tyo the Ord class I showed you: in order to do Ordering operations on a type, you first need to be able to do equality (Eq typeclass) on it |
| 2022-10-01 22:42:42 +0200 | <geekosaur> | this is different from OOP notions of "superclass", which is why I call it "prerequisite" instead |
| 2022-10-01 22:43:12 +0200 | <edrx> | yes, now it's clear! |
| 2022-10-01 22:47:15 +0200 | gmg | (~user@user/gehmehgeh) (Quit: Leaving) |
| 2022-10-01 22:48:11 +0200 | coot | (~coot@213.134.165.79) (Quit: coot) |
| 2022-10-01 22:48:55 +0200 | <edrx> | is there a syntax for joining "instance C1 DD2" and "instance C2 DD2" into a single "instance" declaration? is the "where" after it going to be able to define methods for both C1 and C2? |
| 2022-10-01 22:49:11 +0200 | <geekosaur> | there is not and it cannot |
| 2022-10-01 22:49:43 +0200 | <edrx> | ok, makes some sense |
| 2022-10-01 22:49:59 +0200 | <geekosaur> | we don't normally relate things that way |
| 2022-10-01 22:50:02 +0200 | <edrx> | it will make more sense after I understand more details, I think |
| 2022-10-01 22:51:16 +0200 | <ski> | "you have to declare the instance for C1 before you can declare the instance for C2" -- well, you can actually do it syntactically later, in the source (because all declarations in a module can be mutually recursive, or out-of-order). but conceptually, it still holds that you need to declare the prerequisite/"superclass" instances, before you'll be allowed to declare the instance for the particular class |
| 2022-10-01 22:51:16 +0200 | <geekosaur> | Applicative to Monad is close, but even there there's an exception: ZipLists are Appplicative but not Monad. but otherwise it *would* be convenient to combine them because not only do you need Applicative to get Monad, but they have to be related in a specific way to work properly |
| 2022-10-01 22:52:03 +0200 | <monochrom> | EvanR: You may like to know that you can either go "instance Show1 f => Show (Fix f)" or turn on QuantifiedConstraints, UndecidableInstances and go "instance (forall a. Show a => Show (f a)) => Show (Fix f)" |
| 2022-10-01 22:52:51 +0200 | <EvanR> | yeah the Show1 f is what I'm still working on |
| 2022-10-01 22:52:56 +0200 | coot | (~coot@2a02:a310:e241:1b00:ec1a:e9df:79ac:66ba) |
| 2022-10-01 22:53:18 +0200 | <dolio> | Show1 predates quantified constraints. |
| 2022-10-01 22:53:49 +0200 | <monochrom> | Show1 is probably motivated by lacking QuantifiedConstraints back then. |
| 2022-10-01 22:53:58 +0200 | coot | (~coot@2a02:a310:e241:1b00:ec1a:e9df:79ac:66ba) (Client Quit) |
| 2022-10-01 22:54:02 +0200 | jonathanx_ | (~jonathan@h-178-174-176-109.A357.priv.bahnhof.se) (Remote host closed the connection) |
| 2022-10-01 22:54:04 +0200 | <dolio> | Right. |
| 2022-10-01 22:54:23 +0200 | jonathanx_ | (~jonathan@h-178-174-176-109.A357.priv.bahnhof.se) |
| 2022-10-01 22:54:25 +0200 | <ski> | (oh, a pet peeve of mine : when people have two classes (perhaps with a common superclass), and claim that when constraints for both are in force for a type, that means that the methods of the two classes are expected to interact in a certain way according to some laws. this does not make sense, imho. you could have party A make one instance, party B make another instance, and then party C try to use both |
| 2022-10-01 22:54:31 +0200 | <ski> | instances, and get confused when they don't cohere. the proper way to solve this is to make a common subclass (possibly with zero methods), that specifies the intended interaction laws) |
| 2022-10-01 22:54:44 +0200 | <monochrom> | You can almost read liftShowsPrec's type "(Int -> a -> ShowS) -> ([a] -> ShowS) -> Int -> f a -> ShowS" as "Show a => Int -> f a -> ShowS". |
| 2022-10-01 22:55:28 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 22:57:14 +0200 | <dolio> | I think originally it was literally just a proxy for the quantified constraint, but it's somewhat more useful/flexible to lift a showing function than lift a Show instance (effectively). |
| 2022-10-01 23:00:55 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 244 seconds) |
| 2022-10-01 23:03:35 +0200 | chomwitt | (~chomwitt@2a02:587:dc14:f500:f42c:42aa:75f3:5bbf) (Remote host closed the connection) |
| 2022-10-01 23:03:48 +0200 | <edrx> | ...aha, and this explains how functor laws, monad laws, etc, work: https://downloads.haskell.org/ghc/latest/docs/users_guide/exts/rewrite_rules.html =) |
| 2022-10-01 23:05:20 +0200 | dmwit | (~dmwit@2600:4040:264a:df00:422e:910b:6593:d3fb) |
| 2022-10-01 23:05:28 +0200 | talismanick | (~talismani@2601:200:c100:c9e0::24ac) |
| 2022-10-01 23:05:42 +0200 | ellensol | (~ellen@178-78-210-152.customers.ownit.se) |
| 2022-10-01 23:08:16 +0200 | takuan | (~takuan@178-116-218-225.access.telenet.be) (Remote host closed the connection) |
| 2022-10-01 23:08:17 +0200 | jargon | (~jargon@184.101.88.60) (Remote host closed the connection) |
| 2022-10-01 23:08:24 +0200 | machinedgod | (~machinedg@d198-53-218-113.abhsia.telus.net) |
| 2022-10-01 23:08:35 +0200 | LukeHoersten | (~LukeHoers@user/lukehoersten) |
| 2022-10-01 23:10:54 +0200 | Haematophage | (~colere@about/linux/staff/sauvin) (Ping timeout: 265 seconds) |
| 2022-10-01 23:10:59 +0200 | Colere | (~colere@about/linux/staff/sauvin) |
| 2022-10-01 23:13:09 +0200 | Furor | (~colere@about/linux/staff/sauvin) |
| 2022-10-01 23:13:42 +0200 | <ski> | hololeap : `StateT s (LogicT m) a' would not give you non-backtrackable state (being state that survives backtracking). not quite sure atm if `LogicT m (StateT s m) a' would, or if maybe you'd have to splice the `StateT s' inbetween the two `ContT's in `forall o. ContT () (ContT o m) a' |
| 2022-10-01 23:13:51 +0200 | <ski> | (hololeap : and if the only difference between that and `LogicT o (StateT s m) a' is that the latter also produces the final state .. hm, i guess it depends on whether it (the inner `ContT') does nqCPSy things or not) |
| 2022-10-01 23:14:08 +0200 | wroathe | (~wroathe@206-55-188-8.fttp.usinternet.com) |
| 2022-10-01 23:14:08 +0200 | wroathe | (~wroathe@206-55-188-8.fttp.usinternet.com) (Changing host) |
| 2022-10-01 23:14:08 +0200 | wroathe | (~wroathe@user/wroathe) |
| 2022-10-01 23:14:32 +0200 | __monty__ | (~toonn@user/toonn) (Quit: leaving) |
| 2022-10-01 23:14:52 +0200 | <ski> | edrx : btw, "Interface-passing style" by Fare in 2010-02-17 at <https://fare.livejournal.com/155094.html> might be interesting |
| 2022-10-01 23:16:15 +0200 | Colere | (~colere@about/linux/staff/sauvin) (Ping timeout: 252 seconds) |
| 2022-10-01 23:16:18 +0200 | dmwit | (~dmwit@2600:4040:264a:df00:422e:910b:6593:d3fb) (Ping timeout: 252 seconds) |
| 2022-10-01 23:19:50 +0200 | <edrx> | ski: it has lots of good links! thanks! =) |
| 2022-10-01 23:21:08 +0200 | <ski> | edrx : no, laws associated with type class methods are not (necessarily) rewrite `RULES' |
| 2022-10-01 23:21:57 +0200 | <edrx> | where can I find more about that? |
| 2022-10-01 23:22:06 +0200 | <geekosaur> | edrx, rewrite rules are about optimization, not laws. they often *assume* law-abiding instances, but they are not in and of themselves laws |
| 2022-10-01 23:22:35 +0200 | <geekosaur> | usually useful laws can't actually be written in Haskell |
| 2022-10-01 23:23:40 +0200 | <ski> | edrx : anyway, a Java interface `Widget' more or less corresponds to `exists a. Widget a *> a' in Haskell, where `Widget a' would describe a type class where every method takes (exactly) one `a' as input (and otherwise does not reference `a') .. but to explain that more, i'd have to explain `exists' and `*>' |
| 2022-10-01 23:23:41 +0200 | <geekosaur> | there are other languages where you *can* write out the laws, such as Agda and Idris. (at a different price: you must not only carry around proofs for the laws, but you may take a speed hit during compilation or sometimes runtime) |
| 2022-10-01 23:24:08 +0200 | kenran | (~user@user/kenran) (Remote host closed the connection) |
| 2022-10-01 23:26:28 +0200 | <ski> | methods in Java classes & interfaces implicitly take an object (`this') as input (and can reach into its innards). methods in Haskell type classes don't. however, they may take multiple `a's as in put ("binary methods") or produce `a's as output ("clone methods") (possibly being overloaded *only* on the result type, e.g. `readsPrec :: Read a => Int -> String -> [(a,String)]',`minBound,maxBound :: Bounded a |
| 2022-10-01 23:26:34 +0200 | <ski> | => a',`return :: Monad m => a -> m a',..) |
| 2022-10-01 23:27:29 +0200 | leungbk | (~user@cpe-142-129-149-172.socal.res.rr.com) |
| 2022-10-01 23:28:02 +0200 | jargon | (~jargon@184.101.88.60) |
| 2022-10-01 23:28:09 +0200 | <ski> | (being able to take multiple `a's as input may make you think a little about multi-methods (although therey's no run-time dispatching based on the particular kind of arguments you get, with type classes). it should probably make you think more about Abstract Data Types, where you can have `merge'-operations that take multiple data structures, and are able to see through the abstraction for all of them |
| 2022-10-01 23:28:15 +0200 | <ski> | (unlike in OO)) |
| 2022-10-01 23:30:30 +0200 | <ski> | edrx : did i already link you to the "On Understanding .." papers ? |
| 2022-10-01 23:31:43 +0200 | <edrx> | if you did then I saved only the links, not the papers... can you send me them again? |
| 2022-10-01 23:32:24 +0200 | <ski> | "On Understanding Types, Data Abstraction, and Polymorphism" by Luca Cardelli,Peter Wegner in 1985-12 at <http://lucacardelli.name/Papers/OnUnderstanding.A4.pdf> ; "On Understanding Data Abstraction, Revisited" by William R. Cook in 2009-10 at <https://www.cs.utexas.edu/~wcook/Drafts/2009/essay.pdf> |
| 2022-10-01 23:32:34 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) |
| 2022-10-01 23:35:13 +0200 | <edrx> | thanks! |
| 2022-10-01 23:35:50 +0200 | <edrx> | the paper by Cardelli and Wegner was one of the first papers on Type Theory that I stumbled on... but it didn't make any sense to me at that time =P |
| 2022-10-01 23:36:01 +0200 | <ski> | first one talks about different notions of polymorphism, and also (iirc) talks a bit about abstract types, how they relate to existential quantification |
| 2022-10-01 23:36:43 +0200 | <ski> | while later paper talks a bit more about OO objects, and possibly about how that (in a different way) relates to existential quantification |
| 2022-10-01 23:36:53 +0200 | <ski> | TaPL also talks about these things, in some chapters |
| 2022-10-01 23:37:05 +0200 | burnsidesLlama | (~burnsides@client-8-86.eduroam.oxuni.org.uk) (Ping timeout: 244 seconds) |
| 2022-10-01 23:37:34 +0200 | gnawk | (~gnawk@76-196-232-104.lightspeed.sndgca.sbcglobal.net) |
| 2022-10-01 23:37:55 +0200 | <ski> | @where TaPL |
| 2022-10-01 23:37:55 +0200 | <lambdabot> | "Types and Programming Languages" by Benjamin C. Pierce in 2002-02-01 at <https://www.cis.upenn.edu/~bcpierce/tapl/> |
| 2022-10-01 23:39:26 +0200 | vorpuni | (~pvorp@2001:861:3881:c690:f9de:fbcb:2100:7597) (Remote host closed the connection) |
| 2022-10-01 23:40:17 +0200 | <ski> | edrx : oh, you might also find "Polymorphic Type Inference" by Michael I. Schwartzbach in 1995-03 at <https://cs.au.dk/~amoeller/mis/typeinf.p(s|df)>,<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.57.1493> helpful |
| 2022-10-01 23:40:57 +0200 | <edrx> | totally unrelated question... |
| 2022-10-01 23:42:16 +0200 | michalz | (~michalz@185.246.207.203) (Remote host closed the connection) |
| 2022-10-01 23:43:16 +0200 | <edrx> | thanks! =) |
| 2022-10-01 23:43:25 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Quit: leaving) |
| 2022-10-01 23:44:26 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 23:44:44 +0200 | <edrx> | in one of the courses that I teach I am experimenting with diagrams like the ones here, in pages 7 and 8 - http://angg.twu.net/LATEX/2022-1-C2-infs-e-sups.pdf#page=7 |
| 2022-10-01 23:46:13 +0200 | <edrx> | in which I do a set comprehensions and quantifications with some extra annotations - for example, instead of just ∀ x ∈ {7,8,9}. 2.5 ≤ f(x) |
| 2022-10-01 23:47:03 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) (Client Quit) |
| 2022-10-01 23:47:14 +0200 | gnawk | (~gnawk@76-196-232-104.lightspeed.sndgca.sbcglobal.net) (Remote host closed the connection) |
| 2022-10-01 23:47:41 +0200 | <ski> | i'm not sure what the `em's are intended to mean |
| 2022-10-01 23:47:51 +0200 | <edrx> | the extra annotations can say things like "draw the result of 2.5 ≤ f(x) at the point (x,f(x)) using the colors that I use for truth-values" |
| 2022-10-01 23:47:58 +0200 | Maeda | (~Maeda@91-161-10-149.subs.proxad.net) |
| 2022-10-01 23:48:26 +0200 | <edrx> | "em" means "in", but I think that my explanation here is going to be clearer |
| 2022-10-01 23:48:59 +0200 | <ski> | mhm |
| 2022-10-01 23:49:58 +0200 | kdaishi | (~Thunderbi@mobile-access-6df0e5-30.dhcp.inet.fi) |
| 2022-10-01 23:50:59 +0200 | jpds1 | (~jpds@gateway/tor-sasl/jpds) (Ping timeout: 258 seconds) |
| 2022-10-01 23:51:04 +0200 | titibandit | (~titibandi@xdsl-212-8-150-57.nc.de) (Remote host closed the connection) |
| 2022-10-01 23:51:05 +0200 | <edrx> | suppose that we start with: |
| 2022-10-01 23:51:10 +0200 | <edrx> | [(x,y) | x <- [1..3], y <- [x..3], x+y<6] |
| 2022-10-01 23:51:51 +0200 | <edrx> | I would like to add a clause to it that means "plot (x,y)" - the code could look like this, |
| 2022-10-01 23:52:19 +0200 | <edrx> | [(x,y) | x <- [1..3], y <- [x..3], x+y<6, _ <- plot x y] |
| 2022-10-01 23:53:31 +0200 | jero98772 | (~jero98772@2800:484:1d80:d8ce:3490:26c5:1782:da8c) (Ping timeout: 246 seconds) |
| 2022-10-01 23:53:34 +0200 | <edrx> | but to produce the diagrams that I need these plotting clauses would have to add plot information to an extra list... in other languages I could do things like |
| 2022-10-01 23:54:00 +0200 | <edrx> | [(x,y) | x <- [1..3], y <- [x..3], x+y<6, _ <- table.insert(plotinfo, {x,y})] |
| 2022-10-01 23:54:30 +0200 | <edrx> | where the table.insert would modify that array plotinfo |
| 2022-10-01 23:55:41 +0200 | <edrx> | how would you do that in Haskell? in my beginner's view this seems to need several comprehensions running in parallel, or several "collect" clauses in a Common Lisp "loop"... |
| 2022-10-01 23:57:19 +0200 | leungbk | (~user@cpe-142-129-149-172.socal.res.rr.com) (Ping timeout: 252 seconds) |
| 2022-10-01 23:58:47 +0200 | LukeHoersten | (~LukeHoers@user/lukehoersten) (Quit: My MacBook has gone to sleep. ZZZzzz…) |