Tom Breton (Tehom | 1 Feb 02:07
Picon

Re: Trouble building logical instrument branch

> I think that considering all the circumstances here, the simplest thing
> would
> probably be to just pick one file, say logical-segments-4.rg or whatever
> the
> current version is, and just keep switching it out from there, rather than
> working through a succession of logical-segments-n++.rg files.  It seems
> like
> that would simplify life for everyone.

I will do that in the future.  I mistakenly thought it would be easier all
around to give the new demo a new name.

Update: I have made a work-alike script that does the same as build-qrc. 
It's almost the same.  The only difference between output is that the
header now says "created automatically by" its own name.

I can do whatever makes sense to you.  Leave it local, add it so it's
available but not otherwise include it, or make some attempt to divine
when to use it.

	Tom Breton (Tehom)

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
Julie S | 1 Feb 05:02
Picon
Favicon

Re: Trouble building logical instrument branch

Hello Tom,

May I suggest the radical solution of having both make-qrc-svn and a make-qrc-git scripts.

You are not the only person who has used git for development in RG.  If I'm ever in the position to develop on RG
again, I would probably have a local git repository and sync with svn like you are doing and that script
would make development using git one step easier.

We currently have a make target of 'qrc' as well, as mentioned in README.DEVELOPER.

So a few moments of your time to keep it all tidy would be a nice bonus.

Sincerely,
Julie S.
--- On Tue, 1/31/12, Tom Breton (Tehom) <tehom@...> wrote:

> From: Tom Breton (Tehom) <tehom@...>
> Subject: Re: [Rosegarden-devel] Trouble building logical instrument branch
> To: "D. Michael McIntyre" <michael.mcintyre@...>
> Cc: rosegarden-devel@...
> Date: Tuesday, January 31, 2012, 8:07 PM
> > I think that considering all the
> circumstances here, the simplest thing
> > would
> > probably be to just pick one file, say
> logical-segments-4.rg or whatever
> > the
> > current version is, and just keep switching it out from
> there, rather than
> > working through a succession of logical-segments-n++.rg
(Continue reading)

Julie S | 1 Feb 05:05
Picon
Favicon

Re: Trouble building logical instrument branch

Hello Tom,

I got your new build and it compiled without a hitch here.  Unfortunately, I don't have the time tonight to
give it a good run and provide feedback.

I look forward to trying it real soon.

Sincerely,
Julie S.

--- On Tue, 1/31/12, Julie S <msjulie_s@...> wrote:

> From: Julie S <msjulie_s@...>
> Subject: Re: [Rosegarden-devel] Trouble building logical instrument branch
> To: "D. Michael McIntyre"
<michael.mcintyre@...>, "Tom Breton (Tehom)" <tehom@...>
> Cc: rosegarden-devel@...
> Date: Tuesday, January 31, 2012, 11:02 PM
> Hello Tom,
> 
> May I suggest the radical solution of having both
> make-qrc-svn and a make-qrc-git scripts.
> 
> You are not the only person who has used git for development
> in RG.  If I'm ever in the position to develop on RG
> again, I would probably have a local git repository and sync
> with svn like you are doing and that script would make
> development using git one step easier.
> 
> We currently have a make target of 'qrc' as well, as
(Continue reading)

Favicon
Gravatar

Re: Trouble building logical instrument branch

On Tuesday, January 31, 2012, Chris Cannam wrote:

> [Is the problem] the fact that two different ones (any two) were in use? Or
> is the root problem just that we didn't automate rebuilding the data bundle?

Both, really.

If we add or remove data files, it's apparently necessary to run that script 
manually to pick up the changes.  I probably left it that way on purpose, 
since adding and removing data files is done very infrequently, and there 
isn't much point in adding some constant source of overhead to try to make it 
work automatically.

The other problem is the whole business with SVN, because the script checks to 
make sure files are in SVN before acknowledging they exist.  That prevents you 
from adding foo.rg to your local tree, committing a version of data.cpp that 
references the file, then committing that version without the accompanying 
file.

It's a sound principal, in my opinion, but it doesn't play nice with the Git 
to SVN bridge thing.  If somebody works out a solution to that, I have no 
objection.
--

-- 
D. Michael McIntyre

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
(Continue reading)

Favicon
Gravatar

Re: [Rosegarden-bugs] SF.net SVN: rosegarden:[12798] trunk/rosegarden/src/gui/editors/segment

On Wednesday, February 01, 2012, tedfelix@... wrote:

> No functional changes.  Just comments.  Trying to understand
> TrackLabel::setAlternativeLabel().  Getting closer...

It's all just stupidly complicated, isn't it Ted?

I got into that mess while working on some project related to translation, and 
I don't think I ever fully understood what did what why.  I really can't 
remember what I found so confusing about it, but I think it had something to 
do with misleading uses of "name" and "label" in the code.

If you think there's something to simplify or improve here, you're almost 
certainly right.  If translations still work afterwards, then so much the 
better!

I'd also like to take a moment to apologize for not having more to say about 
all the work you've been doing.  You've been taking on one nasty drudge of a 
project after another, and you've really gotten a lot of tedious and thankless 
work accomplished.  You've managed to nail down some impressively elusive bugs 
along the way too.

Rosegarden's code is in much better shape for your efforts, and I'm personally 
grateful that you wandered by and stepped up to tackle some of this stuff.
--

-- 
D. Michael McIntyre

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
(Continue reading)

Joshua O'Leary | 2 Feb 21:59

Fixed PSR-E403 Rgd File

Hi,
I am using the Yamaha PSR-E403 RGD file, but have noticed that the 
sustain controller and GM key mapping are missing, which could be needed 
for some users of Rosegarden. I have added in those things and exported 
it in the attached file if you are interested.

Joshua

Attachment (Yamaha PSR-E403_fixed.rgd): application/octet-stream, 5606 bytes
------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Rosegarden-devel mailing list
Rosegarden-devel@... - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
Ted Felix | 4 Feb 19:23
Favicon
Gravatar

Re: [Rosegarden-bugs] SF.net SVN: rosegarden:[12798] trunk/rosegarden/src/gui/editors/segment

On 2/2/2012 4:23 AM, D. Michael McIntyre wrote:
> It's all just stupidly complicated, isn't it Ted?

   I've spent many days on TrackLabel, and it is really tough.  However, 
it looks like what it is trying to do is actually really simple.  I'm on 
the verge of doing the actual rewrite and it looks doable.  It will be a 
little more change than I like, but I think it's pretty safe.  I have a 
detailed test plan that I'm running through after each round of 
changes.  So far, so good.

> I got into that mess while working on some project related to translation

   Since you mentioned that, I did notice something about converting 
QString's to C-strings to QString's to C-strings.  There's other stuff 
in there too.  I don't think any of my changes will actually touch the 
translation stuff in any way.  So I think we're safe.

> and you've really gotten a lot of tedious and thankless work accomplished.

   This is my dream-job, so abuse me.  I'm enjoying it.  And thanks 
right back at you.  Without your tireless effort who knows where rg 
would be now.  Nowhere, most likely.

Ted.

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
(Continue reading)

Favicon
Gravatar

Re: [Rosegarden-bugs] SF.net SVN: rosegarden:[12798] trunk/rosegarden/src/gui/editors/segment

On Saturday, February 04, 2012, Ted Felix wrote:

> It will be a little more change than I like, but I think it's pretty safe.

I have no doubt it will come out well.  If translations do get mangled in the 
process, I'll deal with it.  That stuff is fickle and delicate.

>    This is my dream-job, so abuse me.  I'm enjoying it.

Well carry on then!

>    And thanks
> right back at you.  Without your tireless effort who knows where rg
> would be now.  Nowhere, most likely.

I'm quite tired, in truth, but I remain committed to supporting you guys who 
are actually doing work in any way I can.

If I have another big project left in me, I'm probably good for the Qt 5 slog 
that will be coming along before we know it.  I'm really hoping it turns out 
to be a much cleaner process than the last one, and I'm cautiously optimistic 
that it will be.
--

-- 
D. Michael McIntyre

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
(Continue reading)

Tom Breton (Tehom | 5 Feb 04:21
Picon

Logical instruments is ready for a look

I just committed the last changes on my todo list for logical-instrument
and it runs my demo stuff right, so it is now at least semi-stable. 
Please tell me if there are any problems.

On the MidiFile front, it is now exporting repeats, triggers, etc because
MidiFile runs thru metaiterator just like playing sound does.

I macro-guarded a fair amount of debug-printing, mostly in
RosegardenSequencer.cpp.  I needed a debug-print, and a way to turn it
off, and it was easiest to make a new macro-guard.

	Tom Breton (Tehom)

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
Tom Breton (Tehom | 5 Feb 04:24
Picon

Subtle bug squished

In the course of coding this, I fixed a subtle bug that I thought I'd
explain.  At first I thought my code had caused it, so I tracked it down,
but I found that it wasn't, though they interacted.

Almost seven screenfuls of notes made short: When user changes device
connections, we were not resetting the device's or its instruments'
controllers, and that could cause there to be no sound (for me, by sending
zeroes for expression controller).  This could happen if you loaded a
document that guesses the wrong device-name for your setup and you
manually changed to the right output device.  (For me, bogus-surf-jam.rg)

I made setConnection call a new virtual function in Device,
refreshForConnection, that tries to set Device up right again.  Not
perfectly, but ISTM better than just using the old Device's controller
info.  It assumes that zero values are leftover from the old device, and
only removes excess controllers, doesn't add new ones as needed.  Which is
a bit of a cheat, but I was quite tired.

	Tom Breton (Tehom)

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2

Gmane