Logs: liberachat/#haskell
| 2021-07-21 16:03:20 | <ph88_> | you install all your dependencies that you need on ubuntu anyway and can just use dynamic linking |
| 2021-07-21 16:03:27 | → | o1lo01ol1o joins (~o1lo01ol1@bl11-109-140.dsl.telepac.pt) |
| 2021-07-21 16:03:45 | <maerwald> | you asked how you can link with musl |
| 2021-07-21 16:03:56 | <ph88_> | ah ok i get it now .. you were mainly refering to the alpine part of the build ^^ |
| 2021-07-21 16:03:57 | <maerwald> | there it is... if you want to link dynamically with musl, you do it the same way |
| 2021-07-21 16:04:00 | <maerwald> | your question is answered |
| 2021-07-21 16:04:04 | <ph88_> | ok thanks |
| 2021-07-21 16:04:38 | → | pesada joins (~agua@2804:14c:8793:8e2f:6091:e445:c3bf:d39e) |
| 2021-07-21 16:04:54 | <ph88_> | is that your username hasufell ? |
| 2021-07-21 16:05:59 | <ph88_> | i'm still gonna try installing stack directly on alpine without ghcup .. just to see what happens ^^ |
| 2021-07-21 16:06:16 | <maerwald> | I remember that it simply doesn't install GHC |
| 2021-07-21 16:06:46 | <maerwald> | I also had to build several GHCs from scratch for alpine: https://downloads.haskell.org/~ghcup/unofficial-bindists/ghc/ |
| 2021-07-21 16:06:50 | <maerwald> | these are all included in ghcup |
| 2021-07-21 16:07:06 | × | Vajb quits (~Vajb@hag-jnsbng11-58c3a1-224.dhcp.inet.fi) (Read error: Connection reset by peer) |
| 2021-07-21 16:07:12 | × | agua quits (~agua@2804:18:4a:c35:1:0:3be5:8cb2) (Ping timeout: 255 seconds) |
| 2021-07-21 16:07:23 | <maerwald> | recent GHCs have improved alpine support, olders not |
| 2021-07-21 16:07:37 | × | glassofethanol quits (~ethanmorg@cpc78471-cwma11-2-0-cust81.7-3.cable.virginm.net) (Quit: leaving) |
| 2021-07-21 16:07:53 | → | Vajb joins (~Vajb@2001:999:62:1d53:26b1:6c9b:c1ed:9c01) |
| 2021-07-21 16:11:54 | × | jlamothe quits (~jlamothe@104.158.48.100) (Quit: leaving) |
| 2021-07-21 16:12:46 | → | tremon joins (~tremon@217-63-61-89.cable.dynamic.v4.ziggo.nl) |
| 2021-07-21 16:13:16 | → | h98 joins (~h98@187.83.249.216.dyn.smithville.net) |
| 2021-07-21 16:16:22 | → | dunkeln joins (~dunkeln@188.70.26.84) |
| 2021-07-21 16:18:14 | → | xkuru joins (~xkuru@user/xkuru) |
| 2021-07-21 16:21:44 | → | abhixec joins (~abhixec@c-67-169-139-16.hsd1.ca.comcast.net) |
| 2021-07-21 16:22:34 | × | juhp quits (~juhp@128.106.188.220) (Quit: juhp) |
| 2021-07-21 16:22:47 | → | juhp joins (~juhp@128.106.188.220) |
| 2021-07-21 16:23:20 | × | merijn quits (~merijn@83-160-49-249.ip.xs4all.nl) (Ping timeout: 265 seconds) |
| 2021-07-21 16:25:04 | <dminuoso> | ph88_: Disclaimer: If you want to do static builds, keep in mind the licensing implication if you dont also switch out libgmp. Namely, if you stick to libgmp (GHCs default), then that forces GPL onto your build product. |
| 2021-07-21 16:25:21 | → | mr-red joins (~drd@2001:b07:a70:9f1f:1562:34de:f50f:77d4) |
| 2021-07-21 16:25:35 | <dminuoso> | If this matters, be sure to also switch to integer-simple. |
| 2021-07-21 16:25:56 | <ph88_> | can i use it for proprietary binaries that are not distributed ? |
| 2021-07-21 16:26:13 | × | drd quits (~drd@2001:b07:a70:9f1f:1562:34de:f50f:77d4) (Ping timeout: 246 seconds) |
| 2021-07-21 16:26:30 | <maerwald> | it's not AGPL, so the answer is probably yes |
| 2021-07-21 16:26:35 | <geekosaur> | licensing applies to distribution |
| 2021-07-21 16:27:47 | × | ukari quits (~ukari@user/ukari) (Remote host closed the connection) |
| 2021-07-21 16:28:47 | → | ukari joins (~ukari@user/ukari) |
| 2021-07-21 16:29:05 | <dminuoso> | ph88_: For GPL, it only matters if you distribute code or programs. |
| 2021-07-21 16:31:41 | → | Gurkenglas joins (~Gurkengla@dslb-002-203-144-156.002.203.pools.vodafone-ip.de) |
| 2021-07-21 16:32:43 | × | chele quits (~chele@user/chele) (Remote host closed the connection) |
| 2021-07-21 16:32:57 | × | DNH quits (~DNH@2a02:8108:1100:16d8:808b:9f4a:c222:7def) (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 2021-07-21 16:33:35 | × | chomwitt quits (~chomwitt@ppp-94-67-220-135.home.otenet.gr) (Ping timeout: 258 seconds) |
| 2021-07-21 16:35:08 | <ph88_> | is it possible to remove Integer support alltogether ? |
| 2021-07-21 16:35:26 | <maerwald> | oh dear... I had a bug that caused recursive directory creation on windows (probably nested a few million times)... now windows is unable to remove any of that recursively due to path issues xD |
| 2021-07-21 16:35:32 | <dminuoso> | ph88_: Not reasonably, no. |
| 2021-07-21 16:36:05 | <ph88_> | maerwald, try powershell |
| 2021-07-21 16:36:09 | <maerwald> | I did |
| 2021-07-21 16:36:39 | <geekosaur> | isn't there a magic path prefix that evades the old path limits? |
| 2021-07-21 16:36:53 | <maerwald> | tried that too :D |
| 2021-07-21 16:37:41 | <dminuoso> | ph88_: You'd have to remove anything that transitively depends on it. base alone has several hundred references to Integer, you'd have to remove them all. If you have any library dependencies, you will reasonably need Integer. |
| 2021-07-21 16:37:56 | <maerwald> | this could be the most annoying malware on windows... just create insanely deeply nested directories everywhere and the user will go nuts trying to remove them |
| 2021-07-21 16:38:02 | <yushyin> | also gmp is dual licensed LGPLv3/GPLv2 and with lgpl you only need to provide linkable object files ( https://www.gnu.org/licenses/gpl-faq.html#LGPLStaticVsDynamic ) |
| 2021-07-21 16:38:28 | <ph88_> | dminuoso, ya that sounds painful .. thought maybe could just swap it out on the very low level that all Integer are converted to i64 or what not |
| 2021-07-21 16:38:58 | <maerwald> | ph88_: would you really expect that to work? |
| 2021-07-21 16:39:00 | <dminuoso> | ph88_: integer-simple is a drop in replacement for libgmp :) |
| 2021-07-21 16:39:05 | <h98> | https://stackoverflow.com/questions/8460268/delete-a-very-very-deep-tree-of-subdirectories-on-windows do any of these help? |
| 2021-07-21 16:39:16 | × | eggplantade quits (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) (Remote host closed the connection) |
| 2021-07-21 16:39:58 | × | mikoto-chan quits (~mikoto-ch@ip-213-49-189-31.dsl.scarlet.be) (Read error: No route to host) |
| 2021-07-21 16:39:58 | <maerwald> | h98: git bash also fails :D |
| 2021-07-21 16:40:01 | <ph88_> | thanks yushyin |
| 2021-07-21 16:40:09 | <maerwald> | "File name too long" |
| 2021-07-21 16:40:11 | <dminuoso> | yushyin: But you are right, I falsely thought it was GPL. But LGPL still might not be what they want. |
| 2021-07-21 16:40:17 | <dminuoso> | Besides, you can just use integer-simple and be done with it |
| 2021-07-21 16:40:23 | → | mikoto-chan joins (~mikoto-ch@ip-213-49-189-31.dsl.scarlet.be) |
| 2021-07-21 16:40:32 | <geekosaur> | \\?\fullPath |
| 2021-07-21 16:40:35 | <dminuoso> | As some have pointed out, often there's not a real performance difference, and sometimes integer-simple is even faster. |
| 2021-07-21 16:40:44 | <maerwald> | geekosaur: tried that already |
| 2021-07-21 16:40:46 | <dminuoso> | Most code isn't Integer crunching heavy. |
| 2021-07-21 16:40:46 | <c_wraith> | unless you're in one of those few cases where gmp is actually faster |
| 2021-07-21 16:41:03 | <geekosaur> | you managed to make it over 32k deep? wow |
| 2021-07-21 16:41:08 | <maerwald> | hehehe |
| 2021-07-21 16:41:25 | <yushyin> | dminuoso: all true, just wanted to mention this little thing, for the sake of clarity. |
| 2021-07-21 16:41:58 | <dminuoso> | Yes absolutely, it was good of you to correct me here |
| 2021-07-21 16:42:19 | <h98> | https://superuser.com/questions/416351/how-to-remove-an-infinitely-recurring-directory-tree/425666#425666 what about something like this? |
| 2021-07-21 16:43:10 | → | eggplantade joins (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) |
| 2021-07-21 16:45:45 | <ph88_> | yushyin, so i would have to compile my app code into a shared object and distribute that along ? |
| 2021-07-21 16:46:44 | <dminuoso> | ph88_: Why not just switch to integer-simple? |
| 2021-07-21 16:47:20 | <ph88_> | dminuoso, i'm not even in this scenario that i have to choose. I am writing this in a little documentation about what the options are just in case we do get into that scenario |
| 2021-07-21 16:47:36 | <dminuoso> | Alright. |
| 2021-07-21 16:48:20 | <yushyin> | ph88_: that would be a valid option under the LGPLv3, how feasible it is with the haskell toolchains? I don't know. Never done that. |
| 2021-07-21 16:49:23 | <geekosaur> | that's more or less what a package is, no? |
| 2021-07-21 16:49:27 | <ph88_> | ya |
| 2021-07-21 16:49:48 | <geekosaur> | so all you'fd need is a stub main that calls into the package, much like the default Setup.hs |
| 2021-07-21 16:50:23 | × | azeem quits (~azeem@dynamic-adsl-84-220-213-65.clienti.tiscali.it) (Ping timeout: 252 seconds) |
| 2021-07-21 16:51:18 | <ph88_> | that doesn't seem that complicated |
| 2021-07-21 16:51:35 | <ph88_> | might be a good way to do it for the cases that integer-simple is a lot slower |
| 2021-07-21 16:52:01 | → | azeem joins (~azeem@176.201.2.247) |
| 2021-07-21 16:52:45 | × | nate1 quits (~nate@108-233-125-227.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 258 seconds) |
| 2021-07-21 16:53:06 | → | chris_ joins (~chris@81.96.113.213) |
| 2021-07-21 16:54:10 | × | azeem quits (~azeem@176.201.2.247) (Read error: Connection reset by peer) |
| 2021-07-21 16:54:24 | → | azeem joins (~azeem@dynamic-adsl-84-220-213-65.clienti.tiscali.it) |
| 2021-07-21 16:59:29 | × | h98 quits (~h98@187.83.249.216.dyn.smithville.net) (Quit: Client closed) |
| 2021-07-21 17:03:14 | → | nvmd joins (~nvmd@user/nvmd) |
| 2021-07-21 17:06:50 | × | slowButPresent quits (~slowButPr@user/slowbutpresent) (Ping timeout: 265 seconds) |
| 2021-07-21 17:06:56 | → | lbseale joins (~lbseale@user/ep1ctetus) |
| 2021-07-21 17:07:45 | → | DNH joins (~DNH@2a02:8108:1100:16d8:808b:9f4a:c222:7def) |
| 2021-07-21 17:07:54 | × | juhp quits (~juhp@128.106.188.220) (Quit: juhp) |
| 2021-07-21 17:08:08 | → | juhp joins (~juhp@128.106.188.220) |
| 2021-07-21 17:08:21 | → | slowButPresent joins (~slowButPr@user/slowbutpresent) |
| 2021-07-21 17:10:30 | × | fef quits (~thedawn@user/thedawn) (Remote host closed the connection) |
| 2021-07-21 17:10:58 | → | fef joins (~thedawn@user/thedawn) |
All times are in UTC.