MAGIC Magic Mailing List
 
 

From: R. Timothy Edwards (tim AT stravinsky DOT jhuapl.edu)
Date: Tue Feb 26 2002 - 12:14:19 EST

  • Next message: erik peterson: "Re: magic stability"

    Dear Erik,
    
    > the "windows" version and the wrap-magic-in-a-gui deal
    > is making us all burn the current magic to cd-roms and encasing
    > it in glass.
    
    Don't worry; I'm not going to do anything to degrade the stability
    of the magic source.  Actually, the Windows version has very little
    effect on the source;  it's just another set of routines in the
    "graphics" directory.  The only other place it touches the code is
    that Windows demands that "main()" be renamed "WinMain()" or else
    you can't get hold of a handle to the display.  So that won't break
    any code.  I am still working out how the wrapper works, but
    essentially all of the additional code goes into a single ".i" file
    used by SWIG, so that doesn't break any of the code, either.  At
    some point, I will probably be advocating the "try it you might like
    it" approach, but believe me, I really hate "feature creep" and
    want to keep it to a minimum.  Feature creep has rendered many a
    tool virtually useless, and I don't want that to happen to magic.
    The nice thing about the interpreter-wrapped version of magic is
    that it encourages people who want new features to add them as
    scripts.  That gives users the ability to twiddle with magic's
    functionality without touching the C code itself, and users who
    don't want the features aren't forced to accept them along with
    the package.  There are quite a lot of minimally-implemented and
    partially-implemented features which got added to magic over the
    years, and I'm hoping that they can be expunged from the core code,
    making magic "leaner and meaner".
    
    					Regards,
    					Tim
    


  •  
     
    Questions? Contact Rajit Manohar
    cornell logo