2025/12/03

Newest at the top

2025-12-03 01:44:00 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-12-03 01:41:44 +0100CiaoSen(~Jura@2a02:8071:64e1:da0:5a47:caff:fe78:33db) (Ping timeout: 244 seconds)
2025-12-03 01:35:45 +0100Googulator11(~Googulato@2a01-036d-0106-479c-d9ec-010d-f188-ffcb.pool6.digikabel.hu) (Quit: Client closed)
2025-12-03 01:35:44 +0100Googulator88(~Googulato@85-238-68-117.pool.digikabel.hu)
2025-12-03 01:33:29 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 265 seconds)
2025-12-03 01:28:13 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-12-03 01:20:42 +0100 <geekosaur> now I'm being reminded of C source code formatted with formfeeds in comments between functions, thanks
2025-12-03 01:20:07 +0100lambda_gibbon(~lambda_gi@2603:7080:ee00:37d8:11e:138e:d914:c117) (Ping timeout: 264 seconds)
2025-12-03 01:19:57 +0100 <ski> i also not that seldom divide the source into pages (separated by form feeds) of size between thirtythree and sixtysix lines (c.f. "Riastradh's Lisp Style Rules" <https://mumble.net/~campbell/scheme/style.txt>)
2025-12-03 01:18:49 +0100peterbecich(~Thunderbi@172.222.148.214) (Ping timeout: 255 seconds)
2025-12-03 01:18:38 +0100 <haskellbridge> <loonycyborg> while in YAML they're illegal altogether
2025-12-03 01:18:27 +0100 <haskellbridge> <loonycyborg> I think it emits warning if you stick "\t" somewhere already
2025-12-03 01:18:09 +0100 <geekosaur> (case in point, brittany's origins — and its demise)
2025-12-03 01:17:47 +0100 <haskellbridge> <loonycyborg> if there was one true way then compiler would enforce it :P
2025-12-03 01:17:18 +0100 <geekosaur> and you will not convince everyone in the Haskell community to use One True Formatter
2025-12-03 01:17:09 +0100 <jackdk> EvanR: Yes, the configurable formatters I'm aware of can store the project config in version control next to the source.
2025-12-03 01:17:06 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 256 seconds)
2025-12-03 01:16:54 +0100 <geekosaur> my response to said autoformatter fans is that as soon as there's more than one such the purpose is already defeated
2025-12-03 01:16:04 +0100 <ski> (exception is if there's no `module ... where' part)
2025-12-03 01:15:29 +0100 <ski> of course
2025-12-03 01:15:24 +0100lambda_gibbon(~lambda_gi@2603:7080:ee00:37d8:11e:138e:d914:c117)
2025-12-03 01:14:23 +0100 <iqubic> Wait... are you serieous?
2025-12-03 01:13:44 +0100xff0x(~xff0x@2405:6580:b080:900:b577:52ee:470:5943) (Ping timeout: 244 seconds)
2025-12-03 01:13:44 +0100skiindents by two spaces (including the whole module body, from the `where')
2025-12-03 01:13:11 +0100 <EvanR> so like project1 is somehow set up with config 1, project 2 config 2 and everyone can actually keep it straight
2025-12-03 01:12:49 +0100 <EvanR> really
2025-12-03 01:12:27 +0100 <jackdk> (with autoformatter fans, not with you)
2025-12-03 01:12:19 +0100 <jackdk> EvanR: I disagree, because you can still maintain consistency within a codebase, which is where I feel it's most important.
2025-12-03 01:12:15 +0100 <int-e> I meant adoptation, though both work
2025-12-03 01:11:59 +0100 <ski> i would not trust an autoformatter, without closely checking out its opinions
2025-12-03 01:11:57 +0100 <int-e> but if you don't have options then you're defeating adaptation :-P
2025-12-03 01:11:20 +0100 <EvanR> autoformatter fans I've talked to would say if there are configuration options it's defeating the purpose (to have everyone use the same format)
2025-12-03 01:11:18 +0100 <ski> (re tabs vs. spaces, "Yet Another Tabs v. Spaces Debate - I mix tabs and spaces" by dmwit at <http://dmwit.com/tabs/> (also <https://3.bp.blogspot.com/-kX7Gs0lFG_0/V9GIDlF59cI/AAAAAAAAG9E/8OXtloszZRMwd0_NjkWGk6qYedy_0m6jgCL…>) is another opinion)
2025-12-03 01:11:11 +0100 <ski> iqubic : <https://github.com/tibbe/haskell-style-guide/blob/master/haskell-style.md> is one resource, that might be useful to check and ponder
2025-12-03 01:10:10 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-12-03 01:09:30 +0100 <jackdk> ormolu's choices are mostly well-argued but somehow it manages to emit the least aesthetic code I've seen. this is not just an ormulu problem, btw - it seems that each new nix language formatter discovers new frontiers in uglifying code =(. Regardless, I still use autoformatters in a lot of projects just to avoid having style discussions. If I had the time and inclination, I'd look at twiddling the knobs on fourmolu to do more of what i wanted.
2025-12-03 01:06:22 +0100 <geekosaur> likewise fwiw. I think I'd prefer stylish-haskell if someone held a gun to my head and forced me to use a formatter
2025-12-03 01:06:22 +0100Tuplanolla(~Tuplanoll@91-152-225-194.elisa-laajakaista.fi) Tuplanolla
2025-12-03 01:05:46 +0100Googulator95(~Googulato@85-238-68-117.pool.digikabel.hu) (Quit: Client closed)
2025-12-03 01:05:44 +0100Googulator11(~Googulato@2a01-036d-0106-479c-d9ec-010d-f188-ffcb.pool6.digikabel.hu)
2025-12-03 01:01:24 +0100Anarchos(~Anarchos@91-161-254-16.subs.proxad.net) ()
2025-12-03 01:00:48 +0100bitdex(~bitdex@gateway/tor-sasl/bitdex) bitdex
2025-12-03 01:00:26 +0100bitdex(~bitdex@gateway/tor-sasl/bitdex) (Remote host closed the connection)
2025-12-03 00:59:55 +0100 <iqubic> I see.
2025-12-03 00:59:31 +0100 <EvanR> nothing
2025-12-03 00:59:14 +0100 <iqubic> EvanR: What do you use?
2025-12-03 00:59:01 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 246 seconds)
2025-12-03 00:58:57 +0100 <EvanR> I looked at ormolu's output and it didn't match my expectations of style
2025-12-03 00:54:22 +0100merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-12-03 00:52:27 +0100 <iqubic> Is there anywhere I can go to see what they all do? Is there a comparison anywhere?