to think that we are ignoring you. We have a lot going on at the
moment. I'll leave it in my inbox for a week or two and we'll see if
somebody gets to it. If not, please try again!
> Bump!
>
> After a small amount of reworking and one or two new features I have
> another patch. You don't need the previous patch to use this patch, as
> it is against CVS HEAD. Also a lot of this email is copied from the last
> one so ignore that one and just read this one. It is of a prototype of
> non-destructive effects stacks using the "render-to-another-disk-buffer"
> method. See attached.
>
> Here's what to do to see it in action, after you've applied the patch
> and built it:
>
> 1) Create a new project.
> 2) Record a short sample.
> 3) Save the project. Open up the .aup and observe that the XML looks normal.
> 4) Click on the track name and choose the "Render effects" option. A
> placeholder fade-in effect will be applied to the recorded sample.
> 5) Save the project. Open up the .aup and observe that the original data
> is still there, but a new "wet" version has been added to the XML.
> 6) Play the block files on your hard drive and hear how one of them
> fades in, and the other is original.
> 7) Click "Render effects" as much as you want, and observe that the
> original data is never modified, and the "wet" version never gets the
> fade-in applied more than once.
> 8) Export the project and observe how the exported data has the fade-in
> effect applied.
> 9) (New in this patch) There is a new button on the toolbar on the top
> right. It has the undo graphic, but it toggles the view between the
> original "dry" tracks and the "wet" tracks with the fadein effect applied.
>
> The WaveClip class previously had a Sequence* mSequence; pointer as a
> member. I added Sequence* mWetSequence; to the class, and modified
> WaveClip so that it uses mWetSequence at the appropriate times instead
> of mSequence. mWetSequence behaves exactly the same as mSequence, except
> that it exports slightly different XML into the project file. This one's
> implemented a tiny bit differently than the last patch, which had a
> WetSequence class which is gone now.
>
> Studying the structure of the XML, I decided the most canonical
> structure to duplicate was the "sequence" structure. I didn't want to
> duplicate "waveclip" because it contains envelope information that you
> don't want to duplicate. (In other words, one envelope data set should
> affect both the wet and the dry.) However, "waveblock" seemed too much
> of a low-level, implementation issue, and because of its many-to-one
> relationship with sequences. I would like some input from the core
> Audacity developers on whether or not this was a good choice or whether
> I should have done it a different way.
>
> Hopefully this time I can actually put this down and not work on it
> until we get someone else looking at it. I'm waiting for input because I
> don't want to put a lot of work into this thing for it not to be
> accepted by the core developers. I've worked on a few open source
> projects where this has happened. Some things I'd like to know: Is the
> way I chose to implement this a good one? Should I go back and redo some
> things, or should we take a step back and think of another way to
> approach this? (Using the undo stacks?) Eventually I will have to
> serialize effects stacks into XML; what's the best way to go about doing
> this? My recent code changes include making the wet sequence part of the
> Sequence class instead of subclassing it, a good structural improvement.
> Other than that, what do does the Audacity team think?
>
> If all goes well, then the way forward will be to create a small GUI
> that allows the user to edit the effects stack for any particular track,
> but as I said, I'll hold off on that until I get some input on this.
>
> --
> Jorge Rodriguez
> Matrieya Studios, LLC
> Email:
jrodriguez@... <mailto:
jrodriguez@...>
> Phone: (919) 757-3066
>
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Don't miss this year's exciting event. There's still time to save $100.
> Use priority code J8TL2D2.
>
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Audacity-devel mailing list
>
Audacity-devel@...
>
https://lists.sourceforge.net/lists/listinfo/audacity-develDefy all challenges. Microsoft(R) Visual Studio 2008.