Hi Paul,

thanks for the reply. To be honest I'm a bit confused about your suggestion as I don't think I declared any new associative array. The patch only add new value in the string with delimeter. So I'm not sure what you expect me to change there :)

MYT


On Wednesday, 27 November 2024 at 16:14, Paul Zander <negril.nx+gentoo@gmail.com> wrote:

Hi,

associative arrays declared in functions are implicitly local in bash.

That means any associative array declared in global ebuild scope needs
to be declare global explicitly, as it's not guaranteed not to be
sourced within a function by a PM ( e.g. paludis does this ).

https://lists.gnu.org/archive/html/bug-bash/2011-08/msg00274.html https://github.com/pkgcore/pkgcheck/issues/628

Paul

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, Nov 29, 2024 at 01:42:40AM +0000, MYT wrote:

Hi Paul,

thanks for the reply. To be honest I'm a bit confused about your suggestion as I don't think I declared any new associative array. The patch only add new value in the string with delimeter. So I'm not sure what you expect me to change there :)

Probably(?) seen the new documentation bit with declare -A GIT_CRATES,
however it matches the other example and that's fine.

Also, -g should not really matter for that specific associative array
(should make no real difference if it's local or global) and we haven't
been using -g for that array across the tree in ebuilds.

If there is a problem, then it would be beyond the scope of the changes
you're proposing and it should be handled elsewhere (it'd be a
tree-wide issue).

tl;dr nothing for you to change there

MYT

On Wednesday, 27 November 2024 at 16:14, Paul Zander <negril.nx+gentoo@gmail.com> wrote:

Hi,

associative arrays declared in functions are implicitly local in bash.

That means any associative array declared in global ebuild scope needs
to be declare global explicitly, as it's not guaranteed not to be
sourced within a function by a PM ( e.g. paludis does this ).

https://lists.gnu.org/archive/html/bug-bash/2011-08/msg00274.html https://github.com/pkgcore/pkgcheck/issues/628

Paul

--
ionen

-----BEGIN PGP SIGNATURE-----

iQEzBAABCAAdFiEEx3SLh1HBoPy/yLVYskQGsLCsQzQFAmdJhSsACgkQskQGsLCs QzTNXAgApfhhpeGR+Gs4778LjzoK4nHyFJYjjNoLERh/Ze4c1q1BhLJnYdB0eDMs z7JKHecy5Jn6GaSco1VPx4WAr8AlOpj7pj2msA+yrSmSGnKMfAxwjTOumpdy8pvI UkNO8vKX4OmGter2A3/nqPkkz3MFw85YOHA1xvaAIYGCi19WGfzRENB2xbq8jaDJ oEX3PfNqmhUEiJfl4TvLMJIV2pKrUo8dC2DXgWwznhPCHh/L85sLLDcankGRKvbt O43FRbIM4LgY2mMWyOp1OTBUS10BLMV9PCfJSSIMoxijYhehlHaOar9GXBRQk5Rk IPTnx08KBoxGtbkL0q+YzgQw51zo8Q==
=4O3h
-----END PGP SIGNATURE-----

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Hi MYT,

This looks good. I was going to merge however I encountered an issue;
there's no GCO signoff for this commit:

* https://www.gentoo.org/glep/glep-0076.html#certificate-of-origin

Please add the signoff (your pseudonym is fine) and I'm happy to merge.

Aside from that, since I already have you amending the commit message,
it would be nice if the commit message was a touch more descriptive, and
if possible had a bit of an explanation in the body.

If you're confused at all please reach out or open a Pull Request on
GitHub and we can work on it there.

Thanks for your effort on this!

Cheers,

Matt

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)