Binding Space to select-entry gives Key is not bound.

raf mutt at raf.org
Tue Sep 21 10:07:10 UTC 2021


On Mon, Sep 20, 2021 at 08:31:07PM -0700, "Kevin J. McCarthy" <kevin at 8t8.us> wrote:

> Thanks raf, I certainly can't say your opinion is due to a lack of
> experience with Mutt!  However, just to be clear to everyone, my opposition
> to the "tag" interpretation is because of Mutt's current usage of
> <select-entry>.
> 
> Menus that allow both single and multiple selection (e.g. alias list, file
> browser, query menu) use tagging for multiple selection and <select-entry>
> for single selection.  Menus that allow single selection (e.g.
> background-edit list, key selection, postponed list) respond to
> <select-entry> by immediately using/processing the entry (and usually,
> though not always, exiting the menu afterwards).

That convinces me that <display-message> makes much more sense.
There's still so much I don't know about mutt. :-)
Great programs are like that.

> I understand that in general the term "selection" can be interpreted either
> way.  But consistent usage within Mutt is important to me, and I think is a
> good design principle.

Definitely agree.

> The default bindings for <select-entry> are overshadowed by the default
> bindings for <display-message> in the index.  But for those who re-bind it,
> I don't think there's a good reason to drastically change its behavior in
> the index, compared to everywhere else in Mutt.  As you noted, the way to
> operate on or select multiple entries in Mutt is already well established:
> via tagging.
> 
> Now, there *is* one case where the index is used as a selection menu:
> attaching messages.  Currently that interface forces selection (either one
> or multiple message) via tagging.  One could make an argument <select-entry>
> could be used in that case for single selection and immediately returning.
> But it would entail some technical changes to an already complicated menu,
> as well as requiring a *new* keybinding in the index.  Since that operation
> isn't all that common I don't know that it would be worth it, but I'd be
> open for that discussion (on mutt-dev).

I think it's usually a bad idea to implement something that solves
a problem that isn't actually affecting anyone yet. This sounds
like a case of that. There will definitely be better ways to spend
your finite remaining keystrokes. :-)

> -- 
> Kevin J. McCarthy
> GPG Fingerprint: 8975 A9B3 3AA3 7910 385C  5308 ADEF 7684 8031 6BDA

cheers,
raf



More information about the Mutt-users mailing list