Re: [elpa] externals/notmuch-indicator f563869b6a 2/4: Tweak how the keymap is implemented

[Protesilaos Stavrou <info@protesilaos.com>]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Date: Wed, 28 Sep 2022 14:30:01 -0400
>
>> branch: externals/notmuch-indicator
>> commit f563869b6a5e5e12fd791bc109232ed45064f7a7
>> Author: Protesilaos Stavrou <info@protesilaos.com>
>> Commit: Protesilaos Stavrou <info@protesilaos.com>
>>
>>     Tweak how the keymap is implemented
>>     
>>     This continues the work done by Henrik Kjerringvåg in commit f5c9b2e.
>>     Henrik does not require to assign copyright to the Free Software
>>     Foundation for that commit, as it is within the ~15 line threshold.

> [... 37 lines elided]

> Hmm... really?  The new `define-key` will overwrite the binding added in
> the previous call to `notmuch-indicator--format-label`, so if you have
> several elements in `notmuch-indicator-args`, they'll all use the last
> element's `terms` when you click `mouse-1` on them.

Thanks for pointing it out!  I reverted to the previous approach.

> While I'm here:
>
>> +   (if (and face label)
>> +       (format "%s%s " (propertize label 'face face) count)
>> +     (format "%s%s " (or label "") count))
>
> can be rewritten to
>
>     
>         (format "%s%s "
>                 (if (and face label) (propertize label 'face face)
>                   (or label "")
>                 count)

Done.  Thank you!

> Oh, and I see elsewhere that you call `mapconcat` with only 2 args,
> which is a new feature in Emacs-29 whereas the package claims to want to
> support Emacs-27.

Thanks!  On a more general note, how can we catch cases like this one?
The compiler did not warn about it and the doc string of mapconcat does
not explain as much.  I know there is an entry in the NEWS, but having
to check that manually is not easy.

-- 
Protesilaos Stavrou
https://protesilaos.com