Re: [emms-help] Re: EMMS is great stuff, thank you!

[Nix]

On 18 Aug 2007, Michael Olson stated:

> Nix <address@hidden> writes:
>
>> I've got some patches if anyone wants them: a one-liner fixing an
>> XEmacs compatibility bug,
>
> Feel free to send that.  I try to keep the code compatible with XEmacs.

Coming right up.

>> and a rather larger patch adding support for the performer and
>> composer tags. (I'm wondering if it would be possible to rejig things
>> so that tags could be added more easily, but it's sort of hard because
>> tag-related things are bound to keys in things like the tag browser.)
>>
>> I can send a couple of darcs bundles emms-patches-wards if anyone
>> thinks they'll be interested.
>
> Nifty!

I'm not sure if something this small could be considered nifty.

I thought it would be really annoying to add (it's been anything from
annoying to impossible to add to other music players I've tried: they
all have this assumption of a tiny fixed list of tags): but it was
trivial. The code's really rather nice; the only part opaque for a
while was the bdata stuff in emms-browser.el, but after an episode of
infinite recursion I think I got it :)

What's the protocol here? Split bundles up and put a single set of
dependent patches in each?

>         Would you be willing to sign a copyright assignment to the FSF
> for these changes to EMMS, so that we can use your code in future
> releases of EMMS?  I will send you the assignment template in a separate
> off-list message.

(Received, sent off.)

>> Features I can't add yet but will in time:
>>
>>  - proper, permanently accurate synchronization of playlists and
>>    playing time with MPD. (This requires the implementation of the
>>    WIP MPD listener protocol first.)
>
> Interesting.  I didn't know that the MPD developers were working on
> something like that.

Well, it's on the Wiki:
<http://mpd.wikia.com/wiki/MusicPlayerDaemonListenerProtocol>. I don't
think it counts as `working' or even `designed', just `acknowledged as
being nice'.

(I envisage something like SMTP's ETRN, i.e. the client connects and
then flips the connection round, not least because that makes it much
easier to cater for Emacs's lack of server sockets :) )

>                       I made and maintain emms-player-mpd.el, and would
> be happy to apply such a change once this feature is in an MPD release
> (or alternatively, to make the change configurable so that things work
> for both stable MPD releases and the latest bleeding edge MPD).

This is pre-bleeding-edge blue-sky stuff still, but when it's ready...