Fire is step THREE! | github.com/perl6/toolchain-bikeshed | Channel logs: irclog.perlgeek.de/perl6-toolchain/today | useful prior art: metacpan.org/pod/CPAN::Meta::Spec
Set by moderator on 21 January 2016.
01:50 ribasushi joined
jdv79 i think we could structure that better, no? 03:04
ugexe everything else is already structured. it leaves 1 spot for an arbitrary structure of anything external tools might look at 03:45
07:29 domidumont joined 07:32 FROGGS joined 07:36 domidumont joined 11:42 ilbot3 joined
moderator Fire is step THREE! | github.com/perl6/toolchain-bikeshed | Channel logs: irclog.perlgeek.de/perl6-toolchain/today | useful prior art: metacpan.org/pod/CPAN::Meta::Spec
11:59 leont joined
ranguard hi 12:01
raw.githubusercontent.com/perl6/ec.../META.list lists p6-Foo-v1.0.0 and p6-Foo-v1.2.0 - but modules.perl6.org/repo/Foo doesn't offer disambiguation - discuss :) 12:04
ranguard was actually looking for a senario of 2 authors having the same name'd module, rather than same author with 2 versions 12:05
flussence has a package on there with two versions of the same module inside, and afaik one just vanishes on install... 12:16
seems like CURLI is the only part of the stack that takes module versioning seriously right now
ranguard and 2 'auth' releasing the same module name? 12:18
ranguard still thinks it's madness 12:19
nine It's really hard to say in advance how this feature will play out in reality 13:49
jdv79 ranguard: what are you using modules.perl6 for? 14:11
ugexe i dont see any problems with auth/ver etc and multiple packages. if you consider `use Foo::Barr;` bar form and insist people be explicit ala `use Foo::Barr:auth<cpan:xxx>` its easier to digest 15:02
s/bar form/bad form/
lizmat and a new Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2016/01/25/...waiting-4/ 15:15
for some reason, some "f"'s are not rendering, although they *are* in the source 15:16
I blame wordpress JS madness
commute to AmsterdamX.pm meeting 15:17
&
oops ww :-( 15:18
ranguard jdv79: an example of 2 authors releasing a module of the same name
nine: yea, that's part of my issue 15:19
ugexe: so I release Foo::Barr:authLLAP and everyone hard codes that, then I want to pass it on to someone else to maintain, now every has to update to Foo::Barr:authOTHER... OR I release everything with a unique author that I can then pass on to someone else 15:21
ugexe see: s22 superedes, superceded_by, augments, excludes 15:22
design.perl6.org/S22.html#emulates (and the couple headings under it) 15:23
ranguard ugexe: so I run a bit of code on one box that has Foo::Barr:authLLAP.. it works, I deploy to a box with Foo::Barr:authOTHER that superedes Foo::Barr:authLLAP and shit breaks but I have to look through meta files to work out what's acutally being used? 15:26
ugexe the thing is they should also be putting the :ver<xxx> as well (not just the auth). so in most cases you would be hard coding *something*
ranguard or am I misunderstanding?
ugexe you can see which one is being used with RAKUDO_MODULE_DEBUG=1
but again, i think ver and auth should both be explicit if you want to avoid any type of auto-pick madness 15:27
i will say i dont think anyone has any idea how that aliasing will end up working 15:28
ranguard heh, all feels a bit too clever to me 15:29
ugexe the more time that passes, the more the things in s22 make sense to me 15:30
ranguard hasn't got a big enough coffee mug to really process it, I guess I need to set some time asside to really review 15:31
ugexe so it might be overly clever for a reason (or stockholm syndrome has set in after a few years)
nine my first reaction was thinking over engineering, too, but then I came across use cases at work where a those power tools really would have been nice 15:37
S22 does come from experience with real world code bases. 15:38
ugexe yeah. s22 is over-engineered but i think its one area that may have to be 15:40
jdv79 i'm not convinced its clean enough but still working out why 16:20
16:46 leont joined 17:26 Kassandry joined
b2gills .oO( Sometimes you have to over-engineer something to find the right level of engineering ) 18:04
jdv79 grow it then trim it 18:07
18:11 FROGGS joined 18:47 domidumont joined
[Coke] b2gills: it's very hard to do only the right amount of work! 18:49
18:54 ugexe joined 18:59 domidumont1 joined 21:36 autarch joined 22:52 leont joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!