Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 817 818 819 820 821 822 823 824 825 826 827 .. 5022
— 502,152 events total
2020-10-23 12:10:42 <merijn> infinity0: That problem is unsolvable
2020-10-23 12:10:43 → Yangster joins (c053e4f4@192.83.228.244)
2020-10-23 12:10:50 <hc> infinity0: when you ask a hacker a simple question, you might get a long and possible philospophical answer ;p
2020-10-23 12:10:56 <infinity0> fair enough...
2020-10-23 12:10:59 × Yangster quits (c053e4f4@192.83.228.244) (Remote host closed the connection)
2020-10-23 12:11:04 <merijn> "I'm repeatedly computing the same thing, how do I make it fast?" 'well...stop doing that'
2020-10-23 12:11:09 → ggole joins (~ggole@2001:8003:8119:7200:553e:28c1:9eff:faae)
2020-10-23 12:11:23 <merijn> If you're encoding the same string, just, like, save the result and reuse it?
2020-10-23 12:11:49 <merijn> And if you're *not* encoding the same string, then I fail to see how you could convert it faster than "converting it"
2020-10-23 12:12:14 × ubert quits (~Thunderbi@p200300ecdf10db38e6b318fffe838f33.dip0.t-ipconnect.de) (Remote host closed the connection)
2020-10-23 12:12:32 → ubert joins (~Thunderbi@p200300ecdf10db38e6b318fffe838f33.dip0.t-ipconnect.de)
2020-10-23 12:12:36 × cristi_ quits (~cristi@86.121.125.90) (Quit: cristi_)
2020-10-23 12:14:07 × hekkaidekapus quits (~tchouri@gateway/tor-sasl/hekkaidekapus) (Quit: hekkaidekapus)
2020-10-23 12:14:07 <infinity0> i'm repeatedly decoding a stream, that's the thing, and it could either be A or B
2020-10-23 12:14:14 <arahael> merijn: When someone wants to "focus" on doing something efficiently, they don't always mean in computer time, even if they say that. ;)
2020-10-23 12:14:18 × nbloomf quits (~nbloomf@2600:1700:ad14:3020:6d96:aec8:3874:14ea) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2020-10-23 12:14:45 × justsomeguy quits (~justsomeg@unaffiliated/--/x-3805311) (Ping timeout: 240 seconds)
2020-10-23 12:16:01 <merijn> infinity0: Incidentally, replacing Char8.unpack with "T.unpack . decodeUtf8With lenientDecode" will give you the exact same type with a *much* more robust and *much* less bug-prone behaviour
2020-10-23 12:16:13 <infinity0> it's reasonable that a decoding API gives me (leftovers, result), but the two different APIs have different types for "leftovers"
2020-10-23 12:16:14 → hekkaidekapus joins (~tchouri@gateway/tor-sasl/hekkaidekapus)
2020-10-23 12:16:28 → raehik joins (~raehik@cpc95906-rdng25-2-0-cust156.15-3.cable.virginm.net)
2020-10-23 12:16:42 <infinity0> merijn: thanks, yes i thought something like that must be possible, unfortunately it still would be inefficient due to the thunk build-up as if i understand right
2020-10-23 12:16:59 <merijn> (since it will 1. work correctly for any UTF-8 unicode, 2. gracefully handle errors if it's *not* UTF-8. As opposed to Char8 which will just silently corrupt your data)
2020-10-23 12:17:18 <merijn> infinity0: Any conversion to String is *always* going to be inefficient
2020-10-23 12:17:27 <merijn> Given that string easily takes up ~24 byte per character
2020-10-23 12:17:32 <merijn> and ruins your cache
2020-10-23 12:18:17 <[exa]> merijn: like, the users at least... :D
2020-10-23 12:18:33 <infinity0> well laziness could help with that, if only pack was "smart" enough to recognise an "unpack (...)" thunk and actually destructure it instead of building on top of it
2020-10-23 12:18:56 <merijn> [exa]: Well, there's the expectation that users do, but it doesn't hold as you see
2020-10-23 12:19:00 <arahael> infinity0: You mean fusion? :)
2020-10-23 12:19:03 <merijn> infinity0: That's not possible with ByteString
2020-10-23 12:19:18 <[exa]> merijn: moderation/banhammer is not available?
2020-10-23 12:19:28 <merijn> infinity0: To create a ByteString it *must* be copied in full
2020-10-23 12:19:40 <infinity0> sorry, i am talking about Lazy.ByteString here
2020-10-23 12:19:44 <merijn> [exa]: In theory, sure, in practice there's no process
2020-10-23 12:19:57 <merijn> infinity0: Lazy ByteString is just [Strict.ByteString]
2020-10-23 12:20:34 <[exa]> well then it's marvelous that the hackage is in such a good state
2020-10-23 12:20:39 <merijn> The chunk size depending on how/what creates the lazy list
2020-10-23 12:20:49 <infinity0> right, you only need to copy the first part of it, when creating one lazily
2020-10-23 12:21:23 <infinity0> arahael: that only happens at build-time not run-time conditionally AIUI. i'm pretty sure what i mentioned is impossible with today's haskell but i thought perhaps there might be some other strartegy i didn't know about
2020-10-23 12:21:41 → Sarma joins (~Amras@unaffiliated/amras0000)
2020-10-23 12:22:15 <arahael> infinity0: When I deal with haskell, I rarely try to think about compiletime vs runtime. I just accept that all the types are resolved at compile time.
2020-10-23 12:22:51 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-10-23 12:23:20 <infinity0> well, i also care about runtime performance
2020-10-23 12:23:40 → nbloomf joins (~nbloomf@2600:1700:ad14:3020:6d96:aec8:3874:14ea)
2020-10-23 12:24:23 <merijn> infinity0: Which 2 libraries are this?
2020-10-23 12:24:42 <infinity0> Serialise vs Read
2020-10-23 12:24:48 <merijn> oof
2020-10-23 12:24:55 <merijn> As in the Read typeclass?
2020-10-23 12:25:01 <infinity0> yeah
2020-10-23 12:25:05 <merijn> And you care about performance?
2020-10-23 12:25:20 <merijn> Because Read's performance is notoriously *god awful*
2020-10-23 12:25:30 <merijn> Just, like, incredibly bad
2020-10-23 12:25:55 <infinity0> right, for interactive user session is fine, but having it degrade on a O(n^2) basis sounds bad even for an interactive user session
2020-10-23 12:25:58 <merijn> What are you trying to read?
2020-10-23 12:27:21 <infinity0> a stream of a bunch of objects, that could either be Serialise or Read
2020-10-23 12:27:28 → brisbin joins (~patrick@pool-173-49-158-4.phlapa.fios.verizon.net)
2020-10-23 12:27:33 × heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 260 seconds)
2020-10-23 12:28:07 <arahael> When I had to do a Read thing, I made my own read typeclass.
2020-10-23 12:28:34 <merijn> infinity0: And you don't control what creates this input (i.e. there's no way to replace Read with something else?)
2020-10-23 12:29:00 <infinity0> i could replace Read with perhaps JSON, that is another option i guess. mostly Read is convenient to derive and carries no dependencies
2020-10-23 12:29:41 <merijn> If you use Serialise already anyway, why not just derive Serialise?
2020-10-23 12:29:46 <infinity0> the underlying question regarding "runtime fusion" was interesting though
2020-10-23 12:30:12 <infinity0> i am already deriving Serialise. the idea is to have a fast+binary option but also a easy-friendly-manual option
2020-10-23 12:30:27 × acidjnk_new3 quits (~acidjnk@p200300d0c7237805143004c29cad477d.dip0.t-ipconnect.de) (Ping timeout: 260 seconds)
2020-10-23 12:30:48 × nbloomf quits (~nbloomf@2600:1700:ad14:3020:6d96:aec8:3874:14ea) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2020-10-23 12:31:19 → xerox_ joins (~xerox@unaffiliated/xerox)
2020-10-23 12:32:54 → geekosaur joins (82659a09@host154-009.vpn.uakron.edu)
2020-10-23 12:33:07 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-10-23 12:33:18 × knupfer quits (~Thunderbi@200116b8245a88002178c45bdf988b17.dip.versatel-1u1.de) (Remote host closed the connection)
2020-10-23 12:33:26 → knupfer joins (~Thunderbi@200116b8245a88004cde9423a5581dc5.dip.versatel-1u1.de)
2020-10-23 12:35:04 → carlomagno joins (~cararell@148.87.23.13)
2020-10-23 12:35:27 → nbloomf joins (~nbloomf@2600:1700:ad14:3020:3801:1e49:4cbd:104)
2020-10-23 12:35:33 → machinedgod joins (~machinedg@24.105.81.50)
2020-10-23 12:36:18 × carlomagno1 quits (~cararell@148.87.23.5) (Ping timeout: 260 seconds)
2020-10-23 12:37:12 → whatisRT joins (~whatisRT@2002:5b41:6a33:0:ada0:9073:7dbb:6b46)
2020-10-23 12:37:25 → ensyde joins (~ensyde@2600:1702:2e30:1a40:9c62:9bf3:3478:5d05)
2020-10-23 12:37:36 × heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 260 seconds)
2020-10-23 12:37:56 × nbloomf quits (~nbloomf@2600:1700:ad14:3020:3801:1e49:4cbd:104) (Client Quit)
2020-10-23 12:38:13 × fresheyeball quits (~isaac@c-71-237-105-37.hsd1.co.comcast.net) (Quit: WeeChat 2.7.1)
2020-10-23 12:38:22 → cristi_ joins (~cristi@86.121.125.90)
2020-10-23 12:42:07 × ensyde quits (~ensyde@2600:1702:2e30:1a40:9c62:9bf3:3478:5d05) (Ping timeout: 260 seconds)
2020-10-23 12:43:23 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-10-23 12:45:08 → solonarv joins (~solonarv@astrasbourg-552-1-23-6.w90-13.abo.wanadoo.fr)
2020-10-23 12:46:47 × ubert quits (~Thunderbi@p200300ecdf10db38e6b318fffe838f33.dip0.t-ipconnect.de) (Ping timeout: 260 seconds)
2020-10-23 12:47:48 × heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 258 seconds)
2020-10-23 12:50:05 → mirrorbird joins (~psutcliff@2a00:801:42b:7891:16b1:e53f:55b2:15e1)
2020-10-23 12:50:39 × cristi_ quits (~cristi@86.121.125.90) (Quit: cristi_)
2020-10-23 12:55:59 → b_b1 joins (~b_b@185.244.214.217)
2020-10-23 12:57:17 → thir joins (~thir@p200300f27f19de00a929a56a6a990c9a.dip0.t-ipconnect.de)
2020-10-23 13:01:45 × whald quits (~trem@2a02:810a:8100:11a6:1053:b508:9293:f7ba) (Ping timeout: 272 seconds)
2020-10-23 13:01:55 × thir quits (~thir@p200300f27f19de00a929a56a6a990c9a.dip0.t-ipconnect.de) (Ping timeout: 240 seconds)
2020-10-23 13:04:48 × st8less quits (~st8less@2603:a060:11fd:0:dd24:d259:2e39:f97e) (Quit: WeeChat 2.7.1)
2020-10-23 13:05:54 × geekosaur quits (82659a09@host154-009.vpn.uakron.edu) (Ping timeout: 245 seconds)
2020-10-23 13:06:29 → whald joins (~trem@ip4d15893f.dynamic.kabel-deutschland.de)
2020-10-23 13:07:10 × olligobber quits (olligobber@gateway/vpn/privateinternetaccess/olligobber) (Remote host closed the connection)
2020-10-23 13:08:54 × Lowl3v3l quits (~Lowl3v3l@dslb-002-203-195-108.002.203.pools.vodafone-ip.de) (Remote host closed the connection)
2020-10-23 13:09:16 → Lowl3v3l joins (~Lowl3v3l@dslb-002-203-195-108.002.203.pools.vodafone-ip.de)
2020-10-23 13:12:00 × elliott_ quits (~elliott_@pool-108-51-141-12.washdc.fios.verizon.net) (Ping timeout: 272 seconds)
2020-10-23 13:13:17 × GyroW_ quits (~GyroW@unaffiliated/gyrow) (Quit: Someone ate my pie)
←Prev  Next→
Page 1 .. 817 818 819 820 821 822 823 824 825 826 827 .. 5022

All times are in UTC.