Rosegarden

Chris Cannam wrote:

>  -- The mute buttons don't seem to show the correct state when
>     you load a file

Ah, bugger, I thought I'd fixed that.  Actually yeah I can't get
this to fail - I've added a slotSynchroniseWithComposition() to
TrackButtons which is probably quite a good thing to have anyway.
Wire it up when you can replicate the problem.

>  -- Document isn't marked as modified when you change muting,
>     rename tracks, etc. 

Ok, fixed those.

>     I think the only thing that changes file state but that
>     shouldn't change the document's modified-ness is moving the
>     playback pointer.

Actually the pointer position is stored in the document when saving 
so this is behaviour is correct.

>  -- I don't have audacity (yet) on my home machine

Ok I've changed this so that it we don't report when we can't find the
default external editor - we just silently blank the field.  You might
want to change this behaviour (rosegardenconfiguredialog.cpp) as I can't
think of a good way of doing this right now.

>  -- There are Configure Rosegarden options on all the windows'
>     Settings menus, but only the one on the main window works.
>     (At least the one on the notation view certainly doesn't.)
>     It'd be nice if the one on the notation view took you
>     straight to the notation configuration pane, etc.: any idea
>     if that's feasible?

I should certainly think so.  I'd like to do this kind of thing the
first time we run up too so that we can force users to examine their
audio latency settings for example.

The "configure shortcuts" and "configure toolbars" aren't wired up
in clients either.  Perhaps a job for Mr L?

> I couldn't get iiwusynth to work last night either.  No errors
> reported, just no sound.  Rosegarden didn't pick up an
> instrument for it, and I couldn't get iiwuplay to make any
> sound either.  I had it working once and I'm pretty sure I'm
> using the same soundfont now as then, but I've completely
> reinstalled the system since.  Maybe some stupid mistake, or
> maybe the audio bits of my ALSA setup really are broken.

It won't be a soundfont thing - are you running is in alsa_seq mode?
Sounds like something more fundamental with your sound setup.

Now you've fixed your JACK problem I think I've worked out what it was
- of course your AWE32 is so old it probably doesn't do full duplex
audio (i.e. recording whilst playing) and this is JACK's default mode.
Hence your success with the -P option - shame the JACK error messages
aren't more useful - it might be worth submitting a patch to this effect
actually.

B

Richard Bown wrote:
> Chris Cannam wrote:
>>    I think the only thing that changes file state but that
>>    shouldn't change the document's modified-ness is moving the
>>    playback pointer. 
> 
> Actually the pointer position is stored in the document when saving 
> so this is behaviour is correct.

Which behaviour?  The behaviour always used to be that the
document's modified-ness was not changed when the playback
pointer was moved; I'm saying that's a good thing, because
(despite the fact that the pointer position is stored in
the saved document) I don't think you should get the
document marked as modified just because you've played it.
In fact that would be very bad indeed.

> Ok I've changed this so that it we don't report when we can't find the
> default external editor - we just silently blank the field.

Shurely better just to leave it (in case I'm about to install
audacity, or install it between two invocations of RG) and
then to complain when I try to run it?

>>I couldn't get iiwusynth to work last night either.
> 
> It won't be a soundfont thing - are you running is in alsa_seq mode?

I tried iiwusynth -m alsa_seq mysoundfont.sf2 but it complained
about alsa_seq.  So I checked the help text and then tried -m alsa
which it didn't complain about (but silence ensued).  The fact
that I failed to get sound from iiwuplay either suggested that
the problem wasn't just the mode I was running iiwusynth in though.


Chris

Chris Cannam wrote:

> I don't think you should get the
> document marked as modified just because you've played it.
>
> In fact that would be very bad indeed.

Rubbish, if the playback pointer has moved then you've edited the
document - we're an editing program not a player.  Either we store
the position pointer and movement of it has modified the document
or we don't store the position pointer at all - and if you
extrapolate this behaviour to more complete session managment we
should also eventually note what windows are open at the time of
saving and what segments or notes we have selected etc. etc.
The whole point being that we can restore the editing state of the
document when we restore the data.

We could make this a feature in itself - being able to turn this
session management on and off.

> Shurely better just to leave it (in case I'm about to install
> audacity, or install it between two invocations of RG) and
> then to complain when I try to run it?

Whatever - as I said please fiddle with it yourself.

> I tried iiwusynth -m alsa_seq mysoundfont.sf2 but it complained
> about alsa_seq.

"You're screwed" etc.

B

On Wednesday 24 July 2002 13:27, Richard Bown wrote:
> Rubbish, if the playback pointer has moved then you've edited the
> document - we're an editing program not a player.

Does that mean that merely playing the document would mark it as modified=
 ?

> Either we store
> the position pointer and movement of it has modified the document
> or we don't store the position pointer at all

There's a middle ground, see vim. It silently stores the cursor position =
for=20
each files the user opens in ~/.vimrc.

--=20
=09=09=09=09Guillaume
=09=09=09=09http://www.telegraph-road.org

On Wednesday 24 July 2002 12:48, Richard Bown wrote:
> The "configure shortcuts" and "configure toolbars" aren't wired up
> in clients either.  Perhaps a job for Mr L?

Will add that to my ever-growing todo list.

--=20
=09=09=09=09Guillaume
=09=09=09=09http://www.telegraph-road.org

Guillaume Laurent wrote:

> There's a middle ground, see vim. It silently stores the cursor position for 
> each files the user opens in ~/.vimrc.

That's a good middle ground actually - I like it.

And you said the magic word "vim" too which helps quite a bit.

B

Richard Bown wrote:
> And you said the magic word "vim" too which helps quite a bit.

Indeed, if you like I can support Guillaume's argument by
pointing out that Emacs doesn't do this.


Chris

On Wednesday 24 July 2002 15:03, Chris Cannam wrote:
> Richard Bown wrote:
> > And you said the magic word "vim" too which helps quite a bit.
>
> Indeed, if you like I can support Guillaume's argument by
> pointing out that Emacs doesn't do this.

I'm pretty sure there's an elisp package which does, though. Anyway.

--=20
=09=09=09=09Guillaume
=09=09=09=09http://www.telegraph-road.org

Richard Bown wrote:
> Chris Cannam wrote:
>>I tried iiwusynth -m alsa_seq mysoundfont.sf2 but it complained
>>about alsa_seq.
> 
> "You're screwed" etc.

Well, I think perhaps I see some of the problem.

  http://www.hanappe.com/iiwusynth.html
  "Current version is 0.0.3 ... Download: iiwusynth-0.0.3.tar.gz"

whereas at http://savannah.gnu.org/download/iiwusynth/ you can also
download 0.0.4, 0.0.5, 0.0.6, 0.1, 0.2, and 0.2.1.

Maybe I'll be giving that another go later then.


Chris

Chris Cannam wrote:

>   "Current version is 0.0.3 ... Download: iiwusynth-0.0.3.tar.gz"

Oh that old chestnut - yes I don't think 0.0.3 works with the ALSA 0.9.
This has happened before - it could be FAQ worthy.  I definitely shouted
at someone else about this.

B