Welcome the channel on the development of Cro, a set of libraries for building reactive distributed systems, lovingly crafted to take advantage of all the Raku Programming Language has to offer (cro.services). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
patrickb Maybe to clarify: The critical change is changing the depends specification in the modules from `Cro::TLS:ver<0.8.9>` to `Cro::TLS:ver<0.8.9+>:auth<zef:cro>:api<0>`. So we are not depending on one specific version anymore, but a minimum version and an upper cap via the api spec. 11:49
ab5tract So I've been poking a bit at SmokeMachine's issue in Cromponent: github.com/FCO/Cromponent/pull/9 17:15
I think I have narrowed the issue down to the $*TEMPLATE-REPOSITORY being null when this bit of the prelude is executed `'(BEGIN (((' ~ $decls ~ ') given await($*TEMPLATE-REPOSITORY.resolve-absolute` 17:17
but for instance trying to do `($*TEMPLATE-REPOSITORY //= get-template-repository` causes a dependency circularity and I haven't managed to cleanly unwind it 17:19
What's a bit confusing for me is why the existing prelude works for this code: github.com/ab5tract/cro-webapp/blo...kumod#L195 17:21
but not for github.com/FCO/Cromponent/blob/322...kumod#L179 17:23
SmokeMachine ab5tract: one odd thing is that even if I avoid adding Prelude (github.com/FCO/Cromponent/blob/min...kumod#L37) I still get this error: www.irccloud.com/pastebin/EVJCUzxx/ 19:07
ab5tract :/ 19:08
SmokeMachine I think it's breaking on %*WARNINGS now... 19:20
not only that... :( 19:23
ab5tract It’s weird.. explicitly creating the corresponding object in the prelude instead of using the dynamic variable doesn’t fix anything. I would expect it to be costlier but otherwise equivalent 19:31
However it breaks a bunch of tests in the cro web app template distri 19:32
*distro
Variants thereof likewise fail.. 19:33
SmokeMachine I think I found the minimal example! 19:36
ab5tract: github.com/FCO/Cromponent/blob/min...akumod#L37
ab5tract Nice one! 19:37
SmokeMachine if I replace `Q[sub ($_) { join "", (Q『test』) }].EVAL` with `sub ($_) { join "", (Q『test』) }` it works, and if I replace with `Q[sub ($_) { Q『test』 }].EVAL` it also works 19:38
it seems to be something related to the EVAL end the join... 19:41
is the generated sub being lost somehow?
ab5tract What about the method form of join? 19:53
(Can’t check myself atm)
SmokeMachine then method form also works 19:56
ab5tract Very interesting! 20:10
SmokeMachine .AST.EVAL also does not work… 20:20
ab5tract unfortunately switching all uses of `join "", ...` to `(...).join("")` in the relevant Cro files doesn't seem to have any impact :( 20:36
SmokeMachine Só, that might be more than one problem… 20:48
ab5tract yup :/ 20:52

Powered by Cro and the Raku® Programming Language.

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

Thank you!