Logs: freenode/#haskell
| 2021-05-19 14:17:14 | → | stree joins (~stree@68.36.8.116) |
| 2021-05-19 14:17:47 | <[exa]> | ah what a protocolar mistake I made |
| 2021-05-19 14:17:48 | → | merijn joins (~merijn@83-160-49-249.ip.xs4all.nl) |
| 2021-05-19 14:18:02 | <davean> | I mean telnet works. I've never seen anything notice the problem. |
| 2021-05-19 14:20:45 | <[exa]> | also, no one ever complaned that telnet breaks for them on irc, right? |
| 2021-05-19 14:21:05 | → | biglama joins (~alex@static-176-165-167-17.ftth.abo.bbox.fr) |
| 2021-05-19 14:21:34 | → | puke joins (~vroom@217.138.252.58) |
| 2021-05-19 14:22:13 | × | star_cloud quits (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) (Ping timeout: 240 seconds) |
| 2021-05-19 14:23:23 | × | aib42 quits (~aib@unaffiliated/aib42) (Quit: ZNC - http://znc.in) |
| 2021-05-19 14:23:57 | → | d1rges joins (~d1rges@2001-b011-8011-4077-fd30-d0dc-11ee-27dc.dynamic-ip6.hinet.net) |
| 2021-05-19 14:24:17 | × | coddinkn_ quits (~coddinkn@octayn.net) (Remote host closed the connection) |
| 2021-05-19 14:24:21 | → | luke joins (~luke@bitnomial/staff/luke) |
| 2021-05-19 14:24:54 | → | star_cloud joins (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) |
| 2021-05-19 14:25:08 | → | kritzefitz_ joins (~kritzefit@2003:5b:203b:200::10:49) |
| 2021-05-19 14:26:53 | × | d1rges quits (~d1rges@2001-b011-8011-4077-fd30-d0dc-11ee-27dc.dynamic-ip6.hinet.net) (Client Quit) |
| 2021-05-19 14:27:02 | × | usr25 quits (~usr25@unaffiliated/usr25) (Quit: Leaving) |
| 2021-05-19 14:27:30 | → | usr25 joins (~usr25@4.red-83-63-129.staticip.rima-tde.net) |
| 2021-05-19 14:28:19 | → | timCF joins (~i.tkachuk@254-149-20-81.sta.estpak.ee) |
| 2021-05-19 14:28:48 | × | jiribenes quits (~jiribenes@rosa.jiribenes.com) (Quit: jiribenes) |
| 2021-05-19 14:28:53 | × | kritzefitz_ quits (~kritzefit@2003:5b:203b:200::10:49) (Client Quit) |
| 2021-05-19 14:29:01 | × | kritzefitz quits (~kritzefit@2003:5b:203b:200::10:49) (Quit: Leaving) |
| 2021-05-19 14:29:18 | → | kritzefitz joins (~kritzefit@2003:5b:203b:200::10:49) |
| 2021-05-19 14:31:25 | × | usr25 quits (~usr25@4.red-83-63-129.staticip.rima-tde.net) (Client Quit) |
| 2021-05-19 14:32:07 | → | jiribenes joins (~jiribenes@rosa.jiribenes.com) |
| 2021-05-19 14:32:37 | → | cr3 joins (~cr3@192-222-143-195.qc.cable.ebox.net) |
| 2021-05-19 14:32:53 | × | malumore_ quits (~malumore@151.62.117.153) (Ping timeout: 240 seconds) |
| 2021-05-19 14:32:54 | × | peutri quits (~peutri@ns317027.ip-94-23-46.eu) (Quit: Changing server) |
| 2021-05-19 14:33:15 | → | usr25 joins (~usr25@4.red-83-63-129.staticip.rima-tde.net) |
| 2021-05-19 14:33:37 | → | Aquazi joins (uid312403@gateway/web/irccloud.com/x-llrwtwjbkwdbgfbp) |
| 2021-05-19 14:34:10 | × | jai9 quits (~textual@cpe-74-72-114-231.nyc.res.rr.com) (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 2021-05-19 14:34:46 | × | star_cloud quits (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) (Excess Flood) |
| 2021-05-19 14:35:52 | → | Sgeo joins (~Sgeo@ool-18b9875e.dyn.optonline.net) |
| 2021-05-19 14:36:16 | <timCF> | Hello guys! Unpopular question there. I'm contributing to figlet-style ascii library. Before it was able to parse figlet font files only in runtime (IO) with possible failure. I did added TH support to actually embed Haskell terms representing fonts directly into sources https://github.com/tkachuk-labs/haskii/blob/9c2e89065d82e102f7cbc53e4f338ce470b556af/Haskii/Figlet/TH.hs#L72-L84 It works relatively ok, |
| 2021-05-19 14:36:22 | <timCF> | but TH for whatever reason is very slow and eats a lot of RAM. I'm thinking rewriting TH fonts embed to source code generation using ghc-source-gen library. But there is one thing which I actually like in TH - it's how easy to lift runtime terms to source code representation of terms using class Lift which is easily derivable basically for free for the most sum and product types. I'm just thinking, is |
| 2021-05-19 14:36:28 | <timCF> | there such easy way to lift runtime representation of FLF fonts to ghc-source-gen AST. I'm looking to something like Lift class in TH. |
| 2021-05-19 14:37:47 | × | kritzefitz quits (~kritzefit@2003:5b:203b:200::10:49) (Remote host closed the connection) |
| 2021-05-19 14:38:03 | → | kritzefitz joins (~kritzefit@2003:5b:203b:200::10:49) |
| 2021-05-19 14:38:21 | → | ddellac__ joins (~ddellacos@ool-44c73afa.dyn.optonline.net) |
| 2021-05-19 14:38:22 | → | star_cloud joins (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) |
| 2021-05-19 14:38:28 | → | JanBessai joins (~JanB@85-22-27-105.ip.dokom21.de) |
| 2021-05-19 14:39:11 | × | ubert quits (~Thunderbi@p200300ecdf259d35e6b318fffe838f33.dip0.t-ipconnect.de) (Remote host closed the connection) |
| 2021-05-19 14:39:28 | → | ubert joins (~Thunderbi@p200300ecdf259d35e6b318fffe838f33.dip0.t-ipconnect.de) |
| 2021-05-19 14:39:51 | × | kritzefitz quits (~kritzefit@2003:5b:203b:200::10:49) (Remote host closed the connection) |
| 2021-05-19 14:40:08 | → | kritzefitz joins (~kritzefit@2003:5b:203b:200::10:49) |
| 2021-05-19 14:40:08 | × | ddellacosta quits (~ddellacos@ool-44c73afa.dyn.optonline.net) (Ping timeout: 265 seconds) |
| 2021-05-19 14:40:18 | → | ddellacosta joins (~ddellacos@86.106.143.191) |
| 2021-05-19 14:40:44 | → | mitchellsalad joins (uid40617@gateway/web/irccloud.com/x-wojllcxnqpqyxkiy) |
| 2021-05-19 14:41:24 | × | byorgey quits (~byorgey@155.138.238.211) (Quit: leaving) |
| 2021-05-19 14:41:32 | → | byorgey joins (~byorgey@155.138.238.211) |
| 2021-05-19 14:41:52 | × | luke quits (~luke@bitnomial/staff/luke) (Quit: sleep) |
| 2021-05-19 14:41:59 | <merijn> | timCF: oof, way to make your life difficult :p |
| 2021-05-19 14:42:52 | <timCF> | merijn: you mean TH or ghc-source-gen? :) |
| 2021-05-19 14:42:56 | <merijn> | Yes |
| 2021-05-19 14:43:01 | <timCF> | Haha |
| 2021-05-19 14:43:32 | × | byorgey quits (~byorgey@155.138.238.211) (Client Quit) |
| 2021-05-19 14:43:32 | <merijn> | Why do you wanna embed the compile time? That'll probably make your compile times hell too |
| 2021-05-19 14:44:07 | → | byorgey joins (~byorgey@155.138.238.211) |
| 2021-05-19 14:44:30 | × | sedeki quits (~textual@unaffiliated/sedeki) (Quit: Textual IRC Client: www.textualapp.com) |
| 2021-05-19 14:44:32 | × | ddellacosta quits (~ddellacos@86.106.143.191) (Ping timeout: 240 seconds) |
| 2021-05-19 14:44:41 | <timCF> | merijn: just easier to use library itself this way, don't need IO at all to render text into ascii art |
| 2021-05-19 14:44:55 | <timCF> | at the moment IO is needed to load fonts |
| 2021-05-19 14:45:47 | × | acidjnk_new2 quits (~acidjnk@p200300d0c72b958680cf6a8401116b8a.dip0.t-ipconnect.de) (Ping timeout: 250 seconds) |
| 2021-05-19 14:45:58 | <merijn> | So? You need IO at some point anyway :p |
| 2021-05-19 14:47:03 | <merijn> | I mean, at some point you gotta ask yourself "is the pain I am inflicting on myself worth the gain?" :) |
| 2021-05-19 14:47:24 | × | sphalerite quits (~sphalerit@NixOS/user/lheckemann) (Quit: WeeChat 2.9) |
| 2021-05-19 14:47:37 | × | xensky quits (~xensky@xengarden.xen.prgmr.com) (Quit: i quit) |
| 2021-05-19 14:48:05 | × | star_cloud quits (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) (Excess Flood) |
| 2021-05-19 14:48:21 | → | star_cloud joins (~star_clou@ec2-52-11-151-184.us-west-2.compute.amazonaws.com) |
| 2021-05-19 14:48:28 | × | thc202 quits (~thc202@unaffiliated/thc202) (Ping timeout: 258 seconds) |
| 2021-05-19 14:49:20 | × | jamm__ quits (~jamm@unaffiliated/jamm) (Remote host closed the connection) |
| 2021-05-19 14:49:35 | <timCF> | TH is pretty usable with Lift. But slow. And eats all the RAM for bigger ascii fonts. And I heard about problems with cross-compilation (not sure it's true) |
| 2021-05-19 14:49:45 | <merijn> | Yes |
| 2021-05-19 14:49:59 | <tapas> | bye #haskell! moving over to irc.libera.chat |
| 2021-05-19 14:50:04 | <merijn> | Or rather, there aren't problems plural, just one really, really big one |
| 2021-05-19 14:50:19 | ← | tapas parts (sid467876@gateway/web/irccloud.com/x-wqbatyuvmrgyaano) () |
| 2021-05-19 14:50:27 | <merijn> | Like "no one really defined how TH is supposed to work in a cross-compilation context" |
| 2021-05-19 14:50:56 | <merijn> | So there's just no sensical interpretation of what TH means when cross compiling |
| 2021-05-19 14:50:56 | → | thc202 joins (~thc202@unaffiliated/thc202) |
| 2021-05-19 14:51:26 | <merijn> | Extremely simple example |
| 2021-05-19 14:51:34 | <merijn> | > finiteBitSize (1 :: Int) |
| 2021-05-19 14:51:35 | <lambdabot> | 64 |
| 2021-05-19 14:51:51 | → | clam joins (~malc@51.15.134.150) |
| 2021-05-19 14:51:58 | <maerwald> | merijn: just pick the largest number |
| 2021-05-19 14:52:00 | <merijn> | If I have a TH expression using that, should it use the Int size of the *build* machine or the *target*? |
| 2021-05-19 14:52:12 | ← | clam parts (~malc@51.15.134.150) () |
| 2021-05-19 14:52:20 | <timCF> | It depends |
| 2021-05-19 14:52:21 | <merijn> | maerwald: That doesn't work, because whatever you pick I can come up with a case that breaks |
| 2021-05-19 14:52:26 | <merijn> | timCF: Correct |
| 2021-05-19 14:52:39 | × | hiroaki quits (~hiroaki@2a02:8108:8c40:2bb8:989b:41dd:d1e6:c3a3) (Ping timeout: 260 seconds) |
| 2021-05-19 14:52:48 | → | mastarija_ joins (~mastarija@31.217.23.9) |
| 2021-05-19 14:52:51 | <merijn> | timCF: Ideally you wanna distinguish between target and host when cross compiling, but TH wasn't designed with that in mind |
| 2021-05-19 14:53:07 | <merijn> | timCF: So that's not fixable without a pretty significant overhaul/redesign of TH |
| 2021-05-19 14:55:02 | <merijn> | timCF: personally I'd probably load/parse fonts on startup once, then cache them |
| 2021-05-19 14:55:21 | → | Jerub joins (~sthorne@37.17.237.231) |
| 2021-05-19 14:55:35 | <merijn> | timCF: Could even consider using compact regions for faster caching/loading |
| 2021-05-19 14:55:41 | × | mastarija quits (~mastarija@31.217.6.4) (Ping timeout: 240 seconds) |
| 2021-05-19 14:55:42 | → | mouseghost joins (~draco@87-206-9-185.dynamic.chello.pl) |
| 2021-05-19 14:55:42 | × | mouseghost quits (~draco@87-206-9-185.dynamic.chello.pl) (Changing host) |
| 2021-05-19 14:55:42 | → | mouseghost joins (~draco@wikipedia/desperek) |
| 2021-05-19 14:56:28 | <timCF> | merijn: Actually for some tasks I got another very annoying limitation that TH can not generate `import` and `module` expressions which was critical for one project which was using > 500 protoc generated files, every single of them I needed to import manually to TH module haha |
All times are in UTC.