>
>
>I guess another trick would be to have a .changes file which only 
>stored bytecodes, temp names and comments as well.  Get rid of these 
>huge .sources/.changes files taking space on my hard drive. :)  Of 
>course, a really good pretty printer becomes more important once you 
>give up source code based development.
>
>- Doug Way
>   dway at riskmetrics.com


Mmm I wonder if there are less than 64k unique temp names in the 
source, could you save space by having 2 bytes id for each temp name, 
then a master table? Ok I'll admit I'm busy on something and can't 
look (so why am I reading email..., hey you mcintosh back to 
work...), but I'd guess we do 1 byte identifiers from 255 names per 
method, so space for names + 1 byte ids compressed, versus single 
table, and 2 bytes everywhere compressed would be? Mmmm I guess it's 
a factor of how big/compressable those variable names are.


PS If I remember John Brant told me once for a packager he wrote for 
VisualWorks they folded methods that had the same byte codes together 
this saved space, seems you could do the same here, but reflecting on 
Dan note this might not be applicable to the source code if you throw 
it away.
-- 
--
===========================================================================
John M. McIntosh <johnmci at smalltalkconsulting.com> 1-800-477-2659
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
===========================================================================
Custom Macintosh programming & various Smalltalk dialects
PGP Key: DSS/Diff/46FC3BE6
Fingerprint=B22F 7D67 92B7 5D52 72D7  E94A EE69 2D21 46FC 3BE6
===========================================================================