Bart Smaalders wrote:
> We've worked very hard to make this work for ON; getting everyone to
> remove __DATE__ and __TIME__ macros from their source code everywhere
> is hard.  Also, compilers love to leave notes about stuff as well.

Yes, I'm painfully aware of the issues.  (Those compiler notes... 
speaking of sections randomly added to the file and delivered to 
customers...)  Particularly bad are the C++ compilers that apparently 
generate some of their internal symbols randomly.  (Note that __DATE__, 
__TIME__, those C++ symbols, and jar timestamps will all still look like 
"real" differences to you.)

> One of the things we said we were NOT going to do is mandate a build system
> for everyone.  Getting all file generation to be deterministic seems 
> like a tall order.

Indeed.  The question is whether breaking the assumption that two 
installations of the same version of a package will have bit-for-bit 
identical files is acceptable.  I'd personally say "no", that it's a 
reasonable and valuable assumption and that breaking it is likely to 
cost more than any savings.  (Remember that service calls start at about 
$25 each just to say "hello", and you can move an awful lot of bits
for $25.)

If you really must optimize out the "unimportant" changes, I'd recommend 
pushing it as far "upstream" as possible, to minimize the number of 
people who see two packages with different files in them.  The stuff I'm 
saying about comparing systems isn't a thought experiment; people do 
that kind of thing all the time.
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to