[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#923899: RFP: elpa-puppet-mode -- syntax highlighting for puppet manifests in emacs



On Thu, Mar 07, 2019 at 09:55:05AM -0500, Antoine Beaupré wrote:
> 
> I've opened another one to ask upstream to make a new release:
> 
> https://github.com/voxpupuli/puppet-mode/issues/117
>

Thank you Antoine :-)

> ... in the meantime, I think we shouldn't package 0.3 directly. It's
> really old (5 years!) and there's been good stuff packaged since.
> 
> So I would recommend packaging a snapshot instead, something like
> "0.3+git7dee1b5" for example. It's not ideal because git revisions don't
> increment, but hopefully the above issue will be fixed and that would be
> just a temporary hack.
> 
> Alternatively, I could just upload 0.3 as is, but I would honestly not
> use it myself since I want the latest features and bugfixes. This would
> defeat the purpose of the package in first place... ;)
>

Okie doke, that's justification enough for me.  I've uploaded
0.3+132.g7dee1b5-1 to mentors.  I use
$tag+$additional_commits_count.g$hash because it's based on git
describe output, I believe it sorts numerically using the
$additional_commits second field, and I think it would be confusing to
use the "-" separator here (instead of ".").  Arguably a colon makes
more sense than a period, but I'm already using periods in my
packages...I guess I could change this for bullseye if the difference
is significant enough.  eg:

  Such and such tag, plus however many commits is this: g$hash
    /\ colon used similarly to natural language
  vs
  using ".", which feels like a substitute for "@".
    tag+commits@g$hash
    I'm not sure that this is as evident, but I read about how other
    people prefer the "." separator in various -devel threads.
    Meh, enough delimiter philosophy!

This upstream snapshot uses an odd 'puppet-mode-0.4snapshot' as the
internal version for package.el dependency resolution...IIRC
0.4snapshot+foo-1 will sort after 0.4-1, so we can't use that.  And
MELPA unstable uses self-generated "Version: 20180813.1947".  Maybe
0.4snapshot~132.g7dee1b5-1 or 0.4snapshot~132:g7dee1b5-1?  That would
fulfill Debian sort and the elpa version will be less bogus, but I've
never used it before, and it doesn't solve the compatibility with
MELPA unstable problem...I'm definitely not interested in maintaining
a patch that harmonises with MELPA unstable's declared version.

  https://mentors.debian.net/debian/pool/main/p/puppet-mode/puppet-mode_0.3+132.g7dee1b5-1.dsc

> 
> Thanks for the extremely quick response and packaging anyways, you're
> doing an awesome job. :)

Thank you :-)

Nicholas

Attachment: signature.asc
Description: PGP signature


Reply to: