[L2Ork-dev] Contributing to new GUI improvement?

[Albert Graef]

Hi Marco, and welcome on board! :)

On Mon, Dec 18, 2017 at 11:32 AM, Marco Donnarumma <
lists at marcodonnarumma.com> wrote:

> Being myself a webdesigner and thus expert in html and css, as well as
> familiar with nw, I wondered if I would be able to contribute somehow to
> the improvements to the new Purr GUI.
>

That'd be great, contributions are always welcome!

Where in the Purr source should I look first to get an understanding of how
> the new gui works and how it can be improved?
>

The JavaScript source of the GUI is in purr-data/pd/nw/. It's hard to
understand by itself, though, since there are some deep tie-ins with Pd's
real-time engine in the C part, i.e., the stuff in purr-data/pd/src/.
Basically, the GUI and the real-time engine each run in their own process
and communicate over a socket using Pd's "FUDI" protocol (or rather some
Purr-specific variation of it).

Basically, the GUI calls the C part using the pdsend() function (in
pdgui.js), and the C part calls back into the GUI using functions like
gui_vmess(), gui_start_vmess() ... gui_end_vmess(). So these are the kind
of calls you want to watch out for. Be warned that the sources are riddled
with these, and it's not always easy to figure out what interactions are
going on there. But we can always ask Jonathan, he has a really good grasp
of the big hairball that is the internals of Pd. ;-)

For instance, something basic that I'd like to achieve is a small public
> library of new animated gui sliders and gui buttons with embedded preset
> saving.
>

We've also been thinking along these lines. See, e.g., Section 4.3 in our
LAC paper: https://musinf.univ-st-etienne.fr/lac2017/pdfs/19_C_C_137956.pdf

I'm aware this seems not to be among your priorities (at least from the
> read me on Albert github), nevertheless would like to hear your opinion.
>

I actually consider this a high priority, it would be just *amazing* to
have this capability! My understanding is that the main issue there is the
way that Pd does its gop scaling, so we'd need a way to turn that off when
rendering data structures inside a gop area. Jonathan has given this some
thought already, if I'm not mistaken. Jonathan?

Albert

-- 
Dr. Albert Gr"af
Computer Music Research Group, JGU Mainz, Germany
Email:  aggraef at gmail.com
WWW:    https://plus.google.com/+AlbertGraef
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://disis.music.vt.edu/pipermail/l2ork-dev/attachments/20171218/b19af0a2/attachment.html>