> > In one way, yes, but then there's the overhead of loading 
> > and parsing an XML file for each core function...
> 
> True, but what about forcing everyone to load what are 
> optional searches?  I personally never use the phone number 
> lookup, or currency conversion and
> would not load them if I could avoid it..

The cool thing is, the bar just ignores these two if the autodetect_
functions are not defined. So if you want, you can just delete the curr.xml
and anywho.xml, and things will just work (except, of course, that the
currency and phone shortcuts won't fire).

> I think what was originally done was the searches were 
> split out, given a prefix and an alias was created for 
> backward compatiblity..  This could account for some of the 
> aliases in the vanilla aliases.txt file..  I think this 
> backward compatibility is the reason I was over-ruled over 
> the notion of moving all of them to standalone files...

Ahh, I see... That makes sense. Maybe we should run a poll on DQSD-Users to
see what people think about discontinuing some of the aliases (especially
the ones the programmer in me hates)...?

> Other than the handling of url detection and the calc 
> function, I was never in favor of the special handling..  I 
> guess the argument could be made that the calc function 
> should be split out as well, leaving only the URL detection..
> 
> But hey, thats just my opinion..

Theoretically, we could build similar autodetect_ stubs for this one, and
only attempt to execute it if its auto-detector is defined, but I think it
works OK for now. No need to get too stringent, IMO...

Cheers,
Kim



-------------------------------------------------------
This SF.Net email is sponsored by: GNOME Foundation
Hackers Unite!  GUADEC: The world's #1 Open Source Desktop Event.
GNOME Users and Developers European Conference, 28-30th June in Norway
http://2004/guadec.org
_______________________________________________
DQSD-Devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dqsd-devel

Reply via email to