Re: [gentoo-dev] [PATCH] eclass/vim-plugin.eclass: delete if has_version condition.

[Sam James]

> On 21 Jul 2021, at 17:47, Patrice Clement <monsieurp@g.o> wrote:
>
> ... and replace it with a test against the REPLACING_VERSIONS variable.
>
> See https://projects.gentoo.org/pms/8/pms.html#x1-10900011.1.
>
> Patch courtesy of Arfrever.
>

Ideally, we'd use Arfrever's sign off, but this is kind of a grey
area, so it's not a blocker at all right now.

> Closes: https://github.com/gentoo/gentoo/pull/21733
> Signed-off-by: Patrice Clement <monsieurp@g.o>
> ---
> eclass/vim-plugin.eclass | 12 ++++++------
> 1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/eclass/vim-plugin.eclass b/eclass/vim-plugin.eclass
> index 6b72d66111d..460edcce3eb 100644
> --- a/eclass/vim-plugin.eclass
> +++ b/eclass/vim-plugin.eclass
> [snip]

The patch itself seems fine, but I have some suggestions while we're
working on the eclass:

- Add EAPI 8 support

- Please drop obsolete EAPI compatibility hacks like:
>has "${EAPI:-0}" 0 1 2 && ! use prefix && ED="${D}"

This is unnecessary because we're only declaring support for EAPI 6 and 7 right now.

- Could you please convert POSIX tests to Bash tests? [0]

- In update_vim_afterscripts, the einfo seems (at first glance) to be misleading. Should
it be printing ${afterdir} instead?

- Ideally, we'd add the standard inherit guard if possible (see e.g. gnuconfig.eclass for a simple example).

[0] https://devmanual.gentoo.org/tools-reference/bash/#single-versus-double-brackets-in-bash

Let me know if you need any assistance, etc.

thanks,
sam