Roger Herikstad wrote:
I'm developing a Cocoa
application for plotting relatively large data sets (10000-100 000
points) containing waveforms recorded from electrophysiological
experiments. The program allows for efficient panning through the
data, zooming, as well as other features. Since I am doing most of my
analysis in python, the idea occurred to me to try creating an
interface of sort to my Cocoa application using python.
Options:
1. Use IPC (e.g. Apple events, Distributed Objects) to communicate
with your application from a separate Python process.
2. Embed a Python interpreter in your application and use PyObjC to
interact with Cocoa objects from Python and vice-versa. (This topic
came up last month if you check the archives.) Two variations on this:
i. embed the interpreter directly in your application (simpler)
ii. create an NSBundle-based plugin API (c.f. VoodooPad) which takes a
suitably packaged Python interpreter or anything else you might want
to add (more flexible).
3. Package your application's Model layer as a self-contained
framework (as you suggest) which a Python process can import via PyObjC.
Since you're dealing with fairly large amounts of data and speed is
presumably a requirement, that probably eliminates #1. Of #2 and #3,
#2 is probably the easier to implement as it can easily be added to an
existing application without having to do lots of refactoring to break
out the Model as an independent framework. #3 might be worth
considering if you'd have other uses for such a framework, however.
HTH
has
--
Control AppleScriptable applications from Python, Ruby and ObjC:
http://appscript.sourceforge.net
_______________________________________________
Pythonmac-SIG maillist - Pythonmac-SIG@python.org
http://mail.python.org/mailman/listinfo/pythonmac-sig
