[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: MacOS X standalone lilypond app revisited

From: Matthias Neeracher
Subject: Re: MacOS X standalone lilypond app revisited
Date: Fri, 3 Jun 2005 09:24:03 -0700

On Jun 3, 2005, at 2:38 AM, Han-Wen Nienhuys wrote:

Matthias Neeracher wrote:

I used the extended weekend for some Cocoa hacking and came up with  the following prototype:
This is an application that combines the functionality of,, and in one application. By  default, it will just run lilypond out of your fink installation, but  it can also be run as a fully bundled standalone if you copy  {bin,etc,lib,share} from the official into this app and  drop in the following slightly modified script into Contents/Resources :
Let me know whether you think this is an approach worth pursuing further

after some further thought, here are my ideas:

 - I would like to drop the current Platypus based approach, integrating and

 - Integrating is nice, but it should be optional for those who are not on 10.4 (me, for instance :-)

Yes, that should be feasible, especially since PDFKit is used from a separate nib.

Am I correct in assuming that you handle textedit:// urls inside the app itself (without using Launch services).

Yes. Again, this could be split so on pre-10.4, this goes through LaunchServices, on 10.4 internally.

 - For reasons of maintainability, I prefer python over Objective C. Python's dynamic nature makes it a very natural match with Objective C.

I've never written any Python Cocoa code (I'm more of a Perl, Ruby, and Tcl programmer), but I've had a look at the ClickEdit source and it seems quite a natural match indeed. The drawback over Objective C, I assume, is that integration with InterfaceBuilder is not as tight.

Given cocoa's nice MVC structure, I expect that it would be easy to replace the ObjC code with Python. Python with pyobjc comes as a standard install with 10.3 and later, I believe.

One of the other nice things of python is that we'd be able to run the support scripts (both the wrapper for the standalone binary and convert-ly) inside

What do you think?

This all sounds quite plausible. I'll take a stab at implementing it over the weekend.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]