Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 .. 5022
— 502,152 events total
2020-11-16 20:28:44 × ggole quits (~ggole@2001:8003:8119:7200:a813:5a33:e96f:8fc6) (Quit: Leaving)
2020-11-16 20:28:55 <dminuoso> But.. I just stumbled over a very convincing example of dfeuer that suggests catchPure is really bad
2020-11-16 20:28:58 × quarters quits (~quarters@104-0-129-162.lightspeed.irvnca.sbcglobal.net) (Changing host)
2020-11-16 20:28:58 → quarters joins (~quarters@unaffiliated/quarters)
2020-11-16 20:29:16 <dminuoso> (let a = a in a) `seq` error "hallo!"
2020-11-16 20:29:32 <monochrom> Oh that one is even more sinister.
2020-11-16 20:29:35 → vicfred joins (~vicfred@unaffiliated/vicfred)
2020-11-16 20:29:51 <dminuoso> And he says that in the presence of catchPure this could *sometimes* be defined, othertimes not..
2020-11-16 20:29:58 → kritzefitz joins (~kritzefit@212.86.56.80)
2020-11-16 20:30:15 <dolio> Yeah, because GHC also assumes that 'pure exceptions' are equal to non-termination.
2020-11-16 20:30:16 <dminuoso> (obviously, because seq makes no assumption about the ordering)
2020-11-16 20:30:46 <monochrom> "div 1 0" is the kind that we all agree is an exception, the only contention is on language design: should this be bottom or not.
2020-11-16 20:30:51 <dminuoso> dolio: I think its not even that, Id say it's just because `seq` does not guarantee the left side to be evaluated first.
2020-11-16 20:31:05 <dminuoso> So its up to whatever the implementation does in the situation
2020-11-16 20:31:21 <dminuoso> (Which could vary from place to place if the optimizer was up for it)
2020-11-16 20:31:48 <monochrom> "a = a" is a non-termination that is clearly bottom but some compiler optimization can make it detectable, so now you're looking at fragile code optimization getting turned on or off.
2020-11-16 20:32:00 × fendor quits (~fendor@178.115.131.226.wireless.dyn.drei.com) (Remote host closed the connection)
2020-11-16 20:32:04 <dminuoso> oh, or that
2020-11-16 20:33:05 → fendor joins (~fendor@178.115.131.226.wireless.dyn.drei.com)
2020-11-16 20:35:19 <dolio> Anyhow, if you're just using it to paper over a poorly designed API, then that's one thing.
2020-11-16 20:35:52 <dolio> That can't really be justified as treating pure exceptions as actual control flow pervasively.
2020-11-16 20:36:36 → cosimone joins (~cosimone@2001:b07:ae5:db26:d849:743b:370b:b3cd)
2020-11-16 20:36:41 <dminuoso> Right. It's just that edward made that comment a while ago, and it kept lingering in my head because I couldn't quite argue to myself why he was right.
2020-11-16 20:36:52 <_deepfire> Is there a way to pass 'ghci' flags, when it's invoked via cabal repl?
2020-11-16 20:37:17 <merijn> --repl-options
2020-11-16 20:37:28 × vicfred quits (~vicfred@unaffiliated/vicfred) (Quit: Leaving)
2020-11-16 20:37:42 × xff0x quits (~fox@2001:1a81:529e:a500:b7fe:796e:1fbb:7036) (Ping timeout: 260 seconds)
2020-11-16 20:37:56 → boxscape joins (86ab2c1f@gateway/web/cgi-irc/kiwiirc.com/ip.134.171.44.31)
2020-11-16 20:38:16 <dminuoso> But this last example is motivating enough to me. An example of how/why `catchPure` could have a bad impact on my quality of life
2020-11-16 20:38:21 → xff0x joins (~fox@2001:1a81:529e:a500:7f56:5167:60df:e24a)
2020-11-16 20:38:40 <dolio> Yeah, just `let a = a in a` can get you unpredictable results.
2020-11-16 20:38:48 <dminuoso> Right
2020-11-16 20:39:12 <dminuoso> Having fallen prey to not respecting the unsafeness of unsafePerformIO once makes me appreciate being careful to do things in pure code you're not supposed to.
2020-11-16 20:40:06 <_deepfire> merijn: I'm using cabal 3.0, but cabal repl TARGET -O0 --repl-options=-ghci-script=.ghci.local -- doesn't have effect
2020-11-16 20:40:09 × thc202 quits (~thc202@unaffiliated/thc202) (Quit: thc202)
2020-11-16 20:40:43 <dminuoso> monochrom, dolio, edwardk: Thank you all for your insights and arguments.
2020-11-16 20:40:57 × Varis quits (~Tadas@unaffiliated/varis) (Remote host closed the connection)
2020-11-16 20:41:12 <merijn> _deepfire: Maybe "cabal repl" does something else that renders -ghci-script in effective
2020-11-16 20:42:00 <_deepfire> merijn: thank you, nevertheless -- I've missed that option in the long list of cabal's options.. : -)
2020-11-16 20:42:26 × luke quits (~luke@bitnomial/staff/luke) (Quit: sleep)
2020-11-16 20:42:27 <merijn> _deepfire: I know that because the original way was using --ghc-options and I broke that :p
2020-11-16 20:43:01 <_deepfire> merijn: Oh, yes, I've tried --ghc-options and --ghci-options (which is logical, but doesn't parse..)
2020-11-16 20:43:47 → Varis joins (~Tadas@unaffiliated/varis)
2020-11-16 20:47:41 <monochrom> Hey --repl-option=-fdefer-type-errors solves my problem, it doesn't [re-]build every hackaage dependency. Thanks merijn
2020-11-16 20:50:03 <monochrom> To be fair, manually entering ":set -fdefer-type-error" isn't that bad.
2020-11-16 20:50:39 × heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Remote host closed the connection)
2020-11-16 20:50:43 × knupfer quits (~Thunderbi@i5E86B4E0.versanet.de) (Ping timeout: 256 seconds)
2020-11-16 20:50:53 Lord_of_Life_ is now known as Lord_of_Life
2020-11-16 20:51:25 × Iceland_jack quits (~user@31.124.48.169) (Ping timeout: 240 seconds)
2020-11-16 20:55:20 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-11-16 20:55:26 × Jesin quits (~Jesin@pool-72-66-101-18.washdc.fios.verizon.net) (Quit: Leaving)
2020-11-16 20:57:17 × mputz quits (~Thunderbi@dslb-084-058-211-084.084.058.pools.vodafone-ip.de) (Quit: mputz)
2020-11-16 20:57:33 → olligobber joins (olligobber@gateway/vpn/privateinternetaccess/olligobber)
2020-11-16 20:57:42 <tomjaguarpaw> c
2020-11-16 20:58:09 <koz_> I'm in 'StateT s (Either e) a', and have a function 'f :: s -> Either e s'. Is there something similar to 'modify' that would let me pass 'f', to produce a result of type 'StateT s (Either e) ()'?
2020-11-16 20:58:16 <merijn> monochrom: That's why I added --repl-options :p
2020-11-16 20:58:42 <monochrom> Oh you. Many thanks and <3 <3 <3
2020-11-16 20:58:46 <merijn> monochrom: I added flag filtering for ghc-options so the world wouldn't be rebuild, then that broke ghci's use of ghc-options, so I added --repl-options to fix it
2020-11-16 20:59:06 <merijn> repl-options never hit the configuration/hash and thus won't affect dependencies
2020-11-16 20:59:19 <monochrom> Yeah, I notice that cabal build --ghc-option=-ferror-span does not affect the hash.
2020-11-16 20:59:26 × wroathe quits (~wroathe@c-73-24-27-54.hsd1.mn.comcast.net) (Ping timeout: 264 seconds)
2020-11-16 20:59:30 <merijn> monochrom: It's a bit hacky and ad hoc atm
2020-11-16 20:59:30 → wroathe_ joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-11-16 20:59:57 <merijn> It makes for an impressive "power of monoids" showcase, though ;)
2020-11-16 21:00:01 × Frost quits (~Frost@185.163.110.116) ()
2020-11-16 21:00:09 <monochrom> That one I understand. I am thankful for the dirty but necessary work.
2020-11-16 21:01:00 <merijn> monochrom: If you notice something rebuilding that you think shouldn't you should make a Cabal issue and tag me in it
2020-11-16 21:01:08 → ystael joins (~ystael@160.sub-174-242-87.myvzw.com)
2020-11-16 21:01:08 <merijn> monochrom: https://github.com/haskell/cabal/blob/master/Cabal/src/Distribution/Simple/Program/GHC.hs#L49-L304 ;)
2020-11-16 21:01:41 <monochrom> Oh, the only current one that affects me is https://github.com/haskell/cabal/issues/6906
2020-11-16 21:02:07 <merijn> I'm particularly proud of the 'from' and 'to' version checks and how they combine monoidally
2020-11-16 21:02:08 <monochrom> because I would love so much to have ~/.cabal/config contain "ghc-options: -haddock"
2020-11-16 21:02:37 × invaser quits (~Thunderbi@31.148.23.125) (Ping timeout: 256 seconds)
2020-11-16 21:02:42 <merijn> And I still think we need "Monoid m => Bool -> m -> m" in base, but it's hard to find a good name for it
2020-11-16 21:03:01 <merijn> I've been using various versions of mif, mwhen, etc. in my own code, but a good name is hard
2020-11-16 21:03:30 <int-e> :t StateT . (((,) () <$>) .)
2020-11-16 21:03:31 <lambdabot> Functor m => (s -> m s) -> StateT s m ()
2020-11-16 21:04:14 <koz_> int-e: Lol, that's impressively APLy. Thanks!
2020-11-16 21:04:56 <koz_> We should add that to transformers as 'modifyInner'.
2020-11-16 21:05:02 <koz_> Or modifyM or something.
2020-11-16 21:05:18 × wroathe_ quits (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net) (Ping timeout: 256 seconds)
2020-11-16 21:06:23 × coot quits (~coot@37.30.49.253.nat.umts.dynamic.t-mobile.pl) (Quit: coot)
2020-11-16 21:06:29 × heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Remote host closed the connection)
2020-11-16 21:07:08 × geekosaur quits (82659a09@host154-009.vpn.uakron.edu) (Remote host closed the connection)
2020-11-16 21:07:47 → wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-11-16 21:08:57 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-11-16 21:11:09 <int-e> :t \f -> traverse lift (Identity f) >>= put . runIdentity -- hmm
2020-11-16 21:11:10 <lambdabot> (MonadTrans t, Monad m, MonadState s (t m)) => m s -> t m ()
2020-11-16 21:11:47 <koz_> That's the mtl version. :P
2020-11-16 21:12:00 hackage phonetic-languages-examples 0.6.0.0 - A generalization of the uniqueness-periods-vector-examples functionality. https://hackage.haskell.org/package/phonetic-languages-examples-0.6.0.0 (OleksandrZhabenko)
2020-11-16 21:12:50 × wpcarro quits (sid397589@gateway/web/irccloud.com/x-yrrbrsudnjmjbakp) (Read error: Connection reset by peer)
2020-11-16 21:13:06 → wpcarro joins (sid397589@gateway/web/irccloud.com/x-jyvrhcovkjnuufaq)
2020-11-16 21:13:10 × ryjm quits (sid383513@gateway/web/irccloud.com/x-aavthkgvqaggakcc) (Ping timeout: 256 seconds)
2020-11-16 21:13:16 × natim87 quits (sid286962@gateway/web/irccloud.com/x-uifctbbgftadercs) (Read error: Connection reset by peer)
2020-11-16 21:13:16 × ReinH__ quits (sid179972@gateway/web/irccloud.com/x-sdlypohklafvyaji) (Read error: Connection reset by peer)
2020-11-16 21:13:24 × aristid quits (sid1599@gateway/web/irccloud.com/x-ckvrkhwfnpsjpxkc) (Read error: Connection reset by peer)
2020-11-16 21:13:24 × bradparker quits (sid262931@gateway/web/irccloud.com/x-hcrjjmdklhbtocmc) (Read error: Connection reset by peer)
2020-11-16 21:13:27 → natim87 joins (sid286962@gateway/web/irccloud.com/x-mexmpkxgonbwfktv)
2020-11-16 21:13:33 × Cir0X quits (sid221743@gateway/web/irccloud.com/x-ouxxoxkrtirtgzgf) (Read error: Connection reset by peer)
2020-11-16 21:13:35 → aristid joins (sid1599@gateway/web/irccloud.com/x-ieumvqzkylmookgw)
←Prev  Next→
Page 1 .. 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 .. 5022

All times are in UTC.