[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: master 2e090da: Revert commit 1f44a776729adf9c6468a76f8310616fde62ee
From: |
Stefan Kangas |
Subject: |
Re: master 2e090da: Revert commit 1f44a776729adf9c6468a76f8310616fde62eeaa for Flymake. |
Date: |
Mon, 26 Oct 2020 17:58:12 -0700 |
haj@posteo.de (Harald Jörg) writes:
>>> Could such an ert-backport.el be a candidate for ELPA? It should be
>>> useful for all dual-life (Emacs + ELPA) packages.
>>
>> How about just making ert itself into a GNU ELPA core package?
>
> I'd be happy with that.
>
> ert.el would then need to implement some compatibility features itself:
> It uses the new format-prompt, for example, which can be easily covered.
> It also uses backtrace.el, which doesn't exist in Emacs 26.
> backtrace.el, in turn, needs lisp-el-font-lock-keywords-for-backtraces,
> which has been added to lisp-mode.el (in the same commit e09120d6 where
> backtrace.el was created). I didn't dig any further, but I guess
> there's some effort to unravel all of this.
>
> Maybe it would make sense to _start_ this as a long-term goal right now,
> working for versions 28 and beyond? That would make ert-backport.el a
> temporary workaround, to be eliminated as soon as versions below 28 go
> out of scope.
Given the amount of work involved (full disclosure: more than I
expected), it sounds like your compatibility package is probably the
better choice for now. We can think about making ert into a package in
its own right in the future, if we need or want to.
Making ert-backport.el into a GNU ELPA package sounds good to me. But
let's see if anyone else has any comments.
Thanks for working on this.