[Clam-devel] Re: update

David García Garzón dgarcia at iua.upf.edu
Sat Jun 2 11:28:24 PDT 2007


On Dissabte 02 Juny 2007, bennett kolasinski wrote:
> Hi David,
>
> Not sure where you went earlier-- I remember you said your connection @
> home was reliant on your neighbors' wi-fi :)  

Yes you are right. I lost my connection and it is a shame because i had a 
stepped guide for you. But, anyway i am glad to say you did pretty well by 
your own :-)

> I have been trying to make 
> small changes to things but everything depends on one another-- so changes
> I made to Tonnetz.hxx affected KeySpace.hxx, which led to changes in
> FloatArrayDataSource.hxx and now PoolArrayDataSource.  I've attached a
> patch of what I've done so far if you'd like to have a look.  KeySpace.cxx
> and Tonnetz.cxx are compiling but InstantViewPlugin gives me an error
> because CLAM::VM::FloatArrayDataSource::frameData() is pure virtual and I'm 
> not sure where to provide an implementation for it.

The problem is that on NetworkEditor we needed to remove the 'const'ness of 
the function. PoolADS still has the const and the compiler thinks that you 
have not such method. I removed it and also changed some explicit 'double' by 
TData that were needed to have it compiled.


> Anyhow, I'm off for the day- I may be back a bit in the evening but that's
> very late your time-- and I can be online tomorrow morning again around a
> similar time as today.  My internet access will be sporadic from Monday -
> Wednesday this week, but I'll have my laptop and the source code with me.

Ok, lets try again then.

> Also, two questions I came up with that maybe you could clear up:
> - why is NetworkEditor's Tonnetz a QDESIGNER_WIDGET_EXPORT but Annotator's
> isn't? (what is QDESIGNER_WIDGET_EXPORT?)

Because in NetworkEditor the widget is used as a Qt plugin and we need to add 
such macro. So if we want shared implementation Annotator needs so.


> - what is Clam Registrator?  NetworkEditor::Keyspace .. should I be
> registering the Annotator widgets with that?  And if so, should their names
> be distinct from the NetworkEditor registered names so as to prevent any
> conflicts?  (i.e. 'AnnotatorKeyspace' or something like that)

The registrator thing is just for registering the monitors (processings which 
are also data sources) on the processing factory so that the NetworkEditor 
can create this kind of processings. Annotator doesn't have such need, as its 
data sources are not processings. I just removed it of your patch.

About the question on the irc i think remained unanswered: "without a timer, 
how did tonnetz update its display?  i.e. who was controlling the updates"
That's how it works now: As the reproduction goes on the host calls the method 
PoolFloatArrayDataSource::setCurrentTime. The return value means that the 
frame changed and then the host updates the view. That's how it works now. Of 
course it is not 'the way' (tm), so, if as results of our refactoring we may 
get a better way.

I am about commiting your changes but i have some problems with the fonts and 
the keyspace colors. Soon i hope.

-- 
David García Garzón
(Work) dgarcia at iua dot upf anotherdot es
http://www.iua.upf.edu/~dgarcia
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.clam-project.org/pipermail/clam-devel-clam-project.org/attachments/20070602/88f46a79/attachment-0003.pgp>


More information about the clam-devel mailing list