emacs-devel
[Top][All Lists]
Advanced

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

Re: Integrating Midi into Emacs


From: David Kastrup
Subject: Re: Integrating Midi into Emacs
Date: Wed, 14 Jan 2015 11:49:34 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.50 (gnu/linux)

Niels Giesen <address@hidden> writes:

> Hi David,
>
> This looks really cool to me. I tried to use your code, but cannot for the
> life of me find out how to bind functions to those events. Could you give a
> simple example?

Well, this is work in progress.  I've been tripped up by quirks of the
raw Midi spec: to make things more challenging, it allows inserting
foreign command bytes ("realtime events") in the middle of other
commands.  And it allows leaving off command bytes altogether ("running
status byte").  When accessing Midi devices through libraries or the
Midi API, those peculiarities are taken care off, but not so with the
raw Midi stream.

I'm appending the current files I am working with.  They are _not_ a
finished application yet.  The midikbd module produces events.  The
decoding framework should now be working reliably, but currently only
the events I am interested in working with get produced.  And the
lily-midi module integrates this into LilyPond-mode, assuming that it
has already been loaded (again: work in progress, so load order,
autoloads etc are all not in place).  The notenames are hardwired to
dutch, channel changes are not currently used, timing is not recorded
and so on.  The last thing I did was implementing a recording key
signature so that pitches are produced with accidentals chosen in a
useful relation to the key signature.

The long pitch would be to use libmidiport in Emacs proper, and produce
both key up and down events (with at least the up events ignored by
default) and full timing info, similar to mouse events.  Most of this
(apart from the default binding behavior) can probably be presimulated
by extending midi-kbd.el, but the precision of the timing would likely
suffer.

-- 
David Kastrup



reply via email to

[Prev in Thread] Current Thread [Next in Thread]