| disbot11 | <antononcube> @Coke Ok, good. I prefer to have longer desriptive names for these kind of functions. For example, zeckendorf-representation and dual-zeckendorf-representation. | 00:13 | |
| <antononcube> I can file an issue on GitHub, where that can be debated. | |||
| [Coke] | Sure. It's an easy change, and I don't mind. | 00:17 | |
|
02:13
kylese left
02:14
kylese joined
02:26
sibl joined
02:50
sibl left
03:15
kylese left,
kylese joined
03:35
dg left
03:56
dg joined
04:05
human-blip left
04:07
human-blip joined
04:22
Aedil joined
04:32
lichtkind__ joined
04:35
lichtkind_ left
04:43
sibl joined
05:13
cm left,
cm_ joined,
cm_ is now known as cm
06:47
eseyman left
06:53
manu_ joined,
manu_ is now known as eseyman
07:04
annamalai left
07:12
sibl left
07:19
sibl joined
08:17
abraxxa joined
08:30
wayland joined
08:32
Sgeo left
08:33
xinming left
08:35
xinming joined
08:36
sibl left,
sibl joined
|
|||
| wayland | Hi all (and especially lizmat who was kind enough to make suggestions yesterday). I've created a new repo ( github.com/wayland/raku-Mythod ) which contains the Slang I was thinking might make a good example. I'm trying to make something that's basically a method, but is called a "mythod". Pretty much everything in the Slang file (grammar, actions) is taken from Grammar.nqp and Actions.nqp, just with "mythod" instead of "method". Unfortunately, | 08:55 | |
| it doesn't work (it gets that error that I mentioned yesterday). | |||
|
09:08
ilogger2 left
09:09
ilogger2 joined
09:12
dakkar joined
09:25
ilogger2 left
09:26
ilogger2 joined
09:32
abraxxa1 joined
09:35
abraxxa left
10:05
abraxxa joined
10:08
abraxxa1 left
10:09
wayland76 joined,
wayland left
10:26
sibl left
10:27
sibl joined
|
|||
| librasteve_ | over on Reddit - www.reddit.com/r/rakulang/comments...are_button - I said … | 10:31 | |
| > Wonders who all the signatures are on the signed Periodic Table image (scroll to the bottom of the weekly newsletter to view)? Are there any of these folks still active in the community (sorry I should know but some of the signature are hard to decipher). | |||
| [Coke]: thanks for putting a smile on my face this morning | 10:33 | ||
|
10:47
sibl left
11:25
xinming left
11:27
xinming joined
11:36
xinming left
11:38
xinming joined
11:46
itaipu left
|
|||
| wayland76 | Thanks also because this discussion made me realise that the Raku weekly ended up in my Spam folder. I've now rectified this. | 12:10 | |
|
12:22
itaipu joined
|
|||
| lizmat | weekly: neexistujeme.bandcamp.com/track/la...ina-padr-n # Jonathan Worthington's poignant song, inspired by the Russian war in Ukraine | 12:39 | |
| notable6 | lizmat, Noted! (weekly) | ||
| disbot11 | <antononcube> @Coke Are you a "numberphile"? If you are, I think I have couple of "homework" tasks for you. (Related to the "Math::Zeckerdorf" package.) | 15:03 | |
| [Coke] | I am subscribed to the channel and watch several of them. Also follow standupmaths. stopped short of majoring in math in college. | 15:04 | |
| I gave up after trying to plow through discrete math. :) | 15:05 | ||
| Sure, happy to consider them. | |||
| disbot11 | <antononcube> Ok. So, these are my homework tasks: 1) Full documentation of zeckendorf and dual-zeckendorf 2) doing in Raku the Wolfram Language post «Companion to "Base Fibonacci - Numberphile"». | 15:07 | |
| <antononcube> Both require using of Markdown and/or Jupyter. | 15:09 | ||
| <antononcube> Here is an example of "full documentation" for random-maze : github.com/antononcube/Raku-Graph-...nths.ipynb | |||
| <antononcube> I can provided the "starting" document templates for zeckendorf and dual-zeckendorf. | 15:10 | ||
| <antononcube> I guess it is best to file issues in the GitHub repository, right? | 15:11 | ||
| tbrowder | hi, can any MacOS user please tell me if Spreadsheet::XLMS installs ok on yr PC | 15:17 | |
| disbot11 | <librasteve> sure - 2 mins | 15:18 | |
| tbrowder | arg XLSX!! | ||
| disbot11 | <librasteve> Welcome to Rakudo™ v2025.12. | 15:19 | |
| <librasteve> (ie I am on latest Raku, clean rakubrew build) | |||
| <librasteve> ===> Testing [FAIL]: Archive::Libarchive::Raw:ver<0.1.5>:auth<zef:FRITH> Aborting due to test failure: Archive::Libarchive::Raw:ver<0.1.5>:auth<zef:FRITH> (use --force-test to override) | 15:20 | ||
|
15:21
ds7832 joined
|
|||
| tbrowder | pls try --force-test if you haven't yet | 15:22 | |
| not sure what it does, but in testing on new hosts it seems to not need it after the initial install | 15:24 | ||
| disbot11 | <librasteve> it disables test | 15:25 | |
| <librasteve> lemme go zef install Spreadsheet::XLSX --deps-only --force-test then I can just test Spreadsheet::XYZU to eliminate issue with dependency modules | 15:27 | ||
| ugexe | --force-test does not disable tests, that is what --/test does. --force-test just ignores failed tests so the e.g. install process can continue when a test fails | 15:28 | |
| disbot11 | <librasteve> errr - ok | 15:30 | |
| <librasteve> tx! | |||
| <librasteve> ok the thing I did loaded the deps - now for zef install Spreadsheet::XLSX | 15:31 | ||
| <librasteve> Testing [FAIL]: Spreadsheet::XLSX:ver<0.3.5>:auth<zef:raku-community-modules> Aborting due to test failure: Spreadsheet::XLSX:ver<0.3.5>:auth<zef:raku-community-modules> (use --force-test to override) | |||
| <librasteve> hope that helps | |||
| <librasteve> remembers to always use --/test in future | 15:32 | ||
| <librasteve> ;-) | |||
| ugexe | fwiw `--/test` implies `--/test --/test-depends`, so if you still want to fetch the test depends you need to do so explicitly `--/test --test-depends` | 15:33 | |
| [Coke] | anononcube - issues work best as todos, yes. | 15:34 | |
| disbot11 | <antononcube> @Coke feel free to reject the documentation suggestions. | 15:35 | |
| <antononcube> Ideally, I want to include "Math::Zeckendorf" as a dependency in "Math::NumberTheory". | 15:38 | ||
| tbrowder | it installs on my debian hosts—weird | 15:40 | |
| librasteve_ | www.irccloud.com/pastebin/iFpD3B1S | 15:44 | |
| ^^ here’s what happens when I run with —verbose | 15:45 | ||
| [Coke] | regarding libarchive - is it just that you don't the correct version of libarchive installed? | 15:49 | |
| I really don't like skipping tests. I would NOT recommend skipping tests in your general installations, unless you are using a known good install. | 15:50 | ||
|
16:02
stanrifkin joined
16:06
stanrifkin left
|
|||
| disbot11 | <librasteve> checks libarchive situation | 16:10 | |
|
16:13
abraxxa left
|
|||
| tbrowder | ok, i think i may know at least part of the problem, it's a non-standard test setup for one thing. see the source "run-test" file. not sure how that works on macs | 16:16 | |
| also, i did make sure i manually installed the prereqs because i was reluctant to do anything xtra for that change, just trying to use the updated LibXML | 16:20 | ||
| disbot11 | <librasteve> hmmm - reinstalled libarchive with homebrew - but that still errors on zef install Libarchive ;-( ... must attend to $day-job ... let me know if you want me to try something else | ||
| tbrowder | ok, thnx | ||
| disbot11 | <librasteve> yw (sorry couldn't do more) | 16:25 | |
| tbrowder | have you installed MacOS::NativeLib (i assume most here have done that). there is a discussion on the NativeHelpers::Callback about mac problems, not sure if it was entirely resolved. | 16:47 | |
| maybe make MakeOS::Native lib a requirement? | 16:48 | ||
|
17:05
sjn left
17:06
sjn joined
|
|||
| [Coke] | I did 'port install libarchive' on the mac, reinstalled Archive::Libarchive::Raw, died same way | 17:31 | |
| Cannot locate native library 'libarchive.13.dylib' | 17:32 | ||
|
17:39
dakkar left
|
|||
| [Coke] opens github.com/frithnanth/perl6-Archiv.../issues/13 | 17:51 | ||
| lucs | I'm writing myself a File::Snippets module to handle file snippets I work with. | 17:54 | |
| Just want to show you folks the neato naming scheme I came up with for identifiers in the module. | 17:55 | ||
| With maybe a trigger warning for those of you who have a strong aversion to weird naming :-) | |||
| So here goes: | |||
| $snip - snippet | 17:56 | ||
| $snam - snippet name | |||
| $snid - snippet id | |||
| And last but certainly not least: | |||
| $snob - snippet glob :-) | |||
| timo | why does the name "Dr Seuss" come to mind ... | 17:57 | |
| lucs | Hehe! | ||
| [Coke] | ... You mean $snuees | 17:58 | |
| ... You mean $snuess | |||
| timo sneeses | |||
| [Coke] | bleesya | 17:59 | |
| disbot11 | <aruniecrisps> I may have asked this before, but is there a way to do with with multiple values? | 18:31 | |
| <aruniecrisps> I can think of with all(...) but I was wondering if there were other ways of doing it | 18:35 | ||
|
18:36
annamalai joined
18:46
abraxxa joined
19:07
ds7832 left,
ds7832 joined
|
|||
| tbrowder | note Spreadsheet::XLSX did not have github workflows tests for MacOS or Windows. i'm adding them to see clues on my end. | 19:13 | |
|
19:13
abraxxa left
19:14
abraxxa joined
|
|||
| ugexe | is there any reason to believe these issues are just OS/system specific around being able to find those libraries? | 19:23 | |
| for [Coke]'s example i wonder if the .13.dylib variant isn't being found | 19:24 | ||
| given he has shown he installed libarchive | |||
| ah i see in the issue he created that the path it is installed to is not one of the paths that the error says it searched in | 19:28 | ||
| that being said, [Coke] i think you need to add /opt/local/lib to DYLD_LIBRARY_PATH, i.e. export DYLD_LIBRARY_PATH="/opt/local/lib:${DYLD_LIBRARY_PATH}" | 19:30 | ||
| of course that is a system dependent solution, so i don't suggest anyone else try that unless you are using macos and installing your libraries via macports | 19:31 | ||
| tbrowder | i hope lizmat's NativeLib will do the trick | ||
| ugexe | eh, i'm not thrilled with what is does | 19:34 | |
| running brew the brew command specifically | |||
| [Coke]'s example for instance uses macports, not homebrew | 19:36 | ||
| and those are just two of many ways it could be installed | |||
| i would argue if MacOS::NativeLib is a dependency, then homebrew is also a required transitive dependency | 19:37 | ||
| and thus i'd try to avoid that dependency | |||
| disbot11 | <librasteve> I have had similar concerns with libpq.5.dylib on macos | 19:38 | |
| <librasteve> a suitable simlink resiolved that | |||
| <librasteve> I have tried your export DYLD_LIBRARY_PATH="/opt/homebrew/opt/libarchive/bin:${DYLD_LIBRARY_PATH}" (note brew variant in my case), but no luck | 19:41 | ||
| ugexe | does /opt/homebrew/opt/libarchive/bin contain libraries? | ||
| disbot11 | <librasteve> bsdcat bsdcpio bsdtar bsdunzip cpio tar | 19:42 | |
| ugexe | why are you passing that path to DYLD_LIBARY_PATH then? | ||
| disbot11 | <librasteve> also tried export DYLD_LIBRARY_PATH="/opt/homebrew/opt:${DYLD_LIBRARY_PATH}" | ||
| ugexe | where does the libarchive dylib file actually exist? | 19:43 | |
| disbot11 | <librasteve> lemme see | ||
| ugexe | how did you even come up with those paths you tried if you don't know? lol | ||
| disbot11 | <librasteve> (answer is that homebrew puts `export PATH="/opt/homebrew/opt/libarchive/bin:$PATH"`` in the .zshrc) | 19:44 | |
|
19:44
abraxxa left
|
|||
| disbot11 | <librasteve> sorry my braindead application | 19:44 | |
| <librasteve> so the answer is /opt/homebrew/opt/libarchive/lib | 19:47 | ||
| <librasteve> contains: libarchive.13.dylib libarchive.a libarchive.dylib pkgconfig | |||
|
19:47
ds7832 left
|
|||
| disbot11 | <librasteve> but export DYLD_LIBRARY_PATH="/opt/homebrew/opt/libarchive/lib:${DYLD_LIBRARY_PATH}" | 19:47 | |
|
19:47
ds7832 joined
|
|||
| disbot11 | <librasteve> still fails on zef install Libarchive | 19:48 | |
| ugexe | what is your actual failure? run with --debug | ||
| disbot11 | <librasteve> ok | ||
| librasteve_ | www.irccloud.com/pastebin/4LRsDT0A | 19:50 | |
| ugexe | yeah i think you have a different issue | 19:51 | |
| run `zef look Libarchive` (or just cd into the repo) and then `raku -I. t/02-simple-tar-memory.t6` | |||
| maybe that will show more information | |||
| [Coke] | this works for me: | 19:54 | |
| export DYLD_FALLBACK_LIBRARY_PATH=/opt/local/lib | |||
| I think the fallback is the preferred way these days? ISTR it changed a few releases ago, and I only have to look it up when something breaks annually. | |||
| disbot11 | <librasteve> Cannot locate native library 'libarchive.13.dylib': | ||
| librasteve_ | www.irccloud.com/pastebin/vJIkC1RK | 19:55 | |
| disbot11 | <librasteve> then i did export DYLD_LIBRARY_PATH="/opt/homebrew/opt/libarchive/lib:${DYLD_LIBRARY_PATH}" in that terminal | 19:56 | |
| <librasteve> worked ok | |||
| ugexe | 👍 | ||
| disbot11 | <librasteve> must be failing on a different test ... | 19:57 | |
|
19:58
ds7832 left,
ds7832 joined
|
|||
| librasteve_ | www.irccloud.com/pastebin/YiUC7uvj | 20:05 | |
| disbot11 | <librasteve> hmm - I am a confused bunny - maybe ^^ will shed some light for you? (sorry back in ~1 hr) | 20:06 | |
| tbrowder | well i'm sorry i can't fix the package for macos or windows. my pr version is now broken on linux :-\ | 20:20 | |
| kudos and maybe a prize for anyone who can fix it for macos | 20:26 | ||
| ugexe | the error is showing you the paths it is looking in | 20:27 | |
| none of those paths are the one you said you previously set in dyld_fallback_library_path | |||
| you should verify your environment | |||
| [Coke] | ... I can't seem to get that env var set in my .zshrc... or .zshenv... | ||
| weird. works live in the shell, but I can't get it to stick for new ones. | 20:28 | ||
| ... I will figure this out later. | |||
| tbrowder | it's really a cool pkg. lizmat could probably fix it but not too interested in xlsx | 20:32 | |
| ugexe | i doubt there is anything wrong with it | ||
| everything i've read from everyone suggests people are just unfamiliar with setting up external libraries on a given system | 20:33 | ||
| librasteve_ | www.irccloud.com/pastebin/hQPt93DH | ||
| ugexe | librasteve_: maybe try deleting .precomp | ||
| disbot11 | <librasteve> yes and no - here I have rerun the export DYLIB and then the prove6 runs into the same errors I shared previously | ||
| ugexe | otherwise the error is still suggesting that the path provided to the fallback env isn't getting seen | 20:34 | |
| disbot11 | <librasteve> no joy with rm -rf .precomp | 20:35 | |
| ugexe | run the tests without prove6 | 20:36 | |
| i don't know what that is doing | |||
| disbot11 | <librasteve> yet raku -I. t/02-simple-tar-memory.t6 works fine afaict | ||
| ugexe | so your issue is related to prove6 in some way. i wonder if it is adding -Ilib, in which case you may need to delete lib/.precomp | ||
| disbot11 | <librasteve> how to run test without prove6? | 20:37 | |
| ugexe | `raku -I. t/02-simple-tar-memory.t6` | ||
| which you already did | |||
| disbot11 | <librasteve> there is no /perl6-libarchive/lib/.precomp | 20:39 | |
| ugexe | i dunno what the issue is with prove6 then | 20:40 | |
| disbot11 | <librasteve> hmmm - ok - thanks for all your great help! | 20:41 | |
| ugexe | you can make zef not use prove6 by passing --/tap-harness fwiw | 20:43 | |
| [Coke] | ugexe++ | ||
| disbot11 | <librasteve> sudo ln -s /opt/homebrew/opt/libarchive/lib/libarchive.13.dylib /usr/local/lib/libarchive.13.dylib works ok | 20:44 | |
| <librasteve> \o/ | 20:45 | ||
| <librasteve> I suspect there is some macos security at play here | |||
| <librasteve> [Coke]: pls can you try the symlink approach for macports? not sure how to make a general solution ... maybe a script to run in the CI (and somehow during install ... even manual would be better than just failure) | 20:52 | ||
| <librasteve> tbrowder: happy to report that (once the Libarchive shenanigans is fixed), then Spreadsheet::XLSX installed fine | 20:55 | ||
| <librasteve> fwiw I tried reverting the symlink and using --tap-harness - no joy | 20:57 | ||
| <librasteve> chatgpt.com/share/6966b251-59b4-80...bc82d071e6 | 21:00 | ||
| ugexe | i said --/tap-harness | 21:11 | |
| disbot11 | <librasteve> tried both -- and --/ both fail | 21:16 | |
|
21:16
itaipu left
|
|||
| disbot11 | <librasteve> (sorry) | 21:16 | |
| <librasteve> (all work with the symlink btw) | 21:17 | ||
| <librasteve> DYLD_LIBRARY_PATH, DYLD_INSERT_LIBRARIES are ignored per the ^^ | 21:18 | ||
| ugexe | and earlier you said | ||
| 11:56:20 <disbot11><librasteve> then i did export DYLD_LIBRARY_PATH="/opt/homebrew/opt/libarchive/lib:${DYLD_LIBRARY_PATH}" in that terminal | |||
| 11:56:33 <disbot11><librasteve> worked ok | |||
| so you're saying running raku $the-test-file works, but prove6 $the-test-file or zef test don't? | 21:19 | ||
|
21:19
itaipu joined
|
|||
| [Coke] ended up doing the softlink also. | 21:39 | ||
|
21:53
jjido joined
21:57
wayland76 left
|
|||
| disbot11 | <librasteve> the export made the raku ... work, but not the prove6... | 21:58 | |
| librasteve_ | weekly: medium.com/@brianstorms/a-1980-tee...f8a5fdbf6c | 22:21 | |
| notable6 | librasteve_, Noted! (weekly) | ||
| tbrowder | we should use zef test instead of prove6 | 22:53 | |
| i do | 22:55 | ||
|
23:04
Sgeo joined
|
|||
| leont is puzzled wht prove6 would not work | 23:08 | ||
| Though I'd use -b instead of -I. anyway, -I. assumes a META6.json. | 23:11 | ||
| ugexe | leont: it doesn't work with zef either from what I understand, which makes me think how zef/prove6 spawns processes somehow is losing their environment setting | 23:29 | |
| i guess i shouldn't imply zef or prove6 are losing it and instead because its almost certainly something with their system/shell. they just exercise their underlying issue | 23:31 | ||
| leont | Yeah, now that you gave me that description the problem is obvious | 23:33 | |
| And probably a one-line fix. | |||
| github.com/Raku/tap-harness6/blob/...kumod#L909 | 23:35 | ||
|
23:35
ds7832 left
|
|||
| leont | I wanted to override $RAKULIB, but end up nuking all other environmental variables. | 23:35 | |
| No wait, I do start with a copy of %*ENV, so it should be all-right | 23:36 | ||
| ugexe | yea it looks fine to me | ||
|
23:48
jjido left
|
|||