| 6 Jan 2026 | |||
| disbot10 | <librasteve> ok, I will do that, but in crag context i don’t see the benefit of conversional variant … is there a way to programmatically get the exported functions (eg Physics::Constants exports the list as our @physics-constants is export; | 20:04 | |
| <antononcube> Interesting! Of course, something like that is already in the in CLI script -- it knows the functions/subs of "Math::NumberTheory". I have to consider having those names exported by the package itself. | 20:07 | ||
| <antononcube> See: github.com/antononcube/Raku-Math-N...-theory#L9 | 20:09 | ||
| <antononcube> I am inclined to have such internal information declared as our, though. (Not exported.) | 20:10 | ||
| itaipu | itaipu left | 20:10 | |
| disbot10 | <librasteve> OK - now fezzed as App::Crag:ver<0.0.44> should be available shortly (or you can zef the gh repo) | 20:21 | |
| <librasteve> looking at your L9, I think (i) crag is not using your number-theory CLI, (ii) since it goes use Number-Theory, then really stuff like this should be offered as a class method on the package (??) and (iii) it would be nice to be consistent - maybe I can change to Physics::Constants.list() and you can make a Math::NumberTheory.functions() [or whatever method name you like]?? | 20:26 | ||
| euandreh | euandreh left | 20:27 | |
| disbot10 | <antononcube> Right, the CLI script should be much shorter and its main functionality(/ies) should be provided by a few subs. | 20:32 | |
| <antononcube> I will refactor it at some point to be like that. | |||
| <librasteve> ok - all done ... no need to refactor the CLI on my behalf as I am just using Math::NumberTheory rakumod | 20:57 | ||
| <antononcube> Ok -- great! | 20:58 | ||
| timo | my analysis of what makes nqp segfault during building on opensuse leap 16: github.com/rakudo/rakudo/issues/60...3716399002 | 21:20 | |
| Aedil | Aedil left | 21:42 | |
| patrickb | oh wow, this is pretty tough stuff | 21:54 | |
| wayland76 | wayland76 left | 21:57 | |
| patrickb | Is there a way to specify a signature to only allow values of the following format: `:pty(:rows(123), :cols(456))`? It's not meant to be used in live code, but in the docs. It does need to compile though. | 21:59 | |
| I could just put `:@pty` there and explain which values need to be there, but it'd be nice to be a little more expressive in the signature itself... | 22:01 | ||
| disbot10 | disbot10 left | 22:06 | |
| disbot11 | disbot11 joined | 22:07 | |
| timo | is it an array with pairs in it, or a hash or an object with attributes? | 22:10 | |
| there is subsignature syntax and there is the "where" syntax | 22:11 | ||
| patrickb | that's the code extracting the values | 22:14 | |
| so it expects a list with two key/values | 22:15 | ||
| github.com/rakudo/rakudo/blob/main...kumod#L152 | 22:16 | ||