Re: cvs commit: libh/lib/sysinstall Feature.cc

Aaarrghh... I struggled a lot with this class to make feature
comparison deterministic. I'm not sure this is the good path to
take.

I suggest you back this out. Some packages *do* use letters in their
packages, but I don't think we should include them, because they can't
essentially be used in comparisons. 

I mean, this all comes back to:

1.a.4c.3 ?= 1.2.3c

and such. By not allowing strings in version, you make a stricter
version format that is more foolproof.

Can you prove that this version format is deterministic?

Also please see the numerous mails I sent on the least WRT that, and
that I had no answer for.

Thanks

A.

On Thu Sep 26, 2002 at 07:05:45AM -0400, Alexander Langer wrote:
> alex        2002/09/26 07:05:45 EDT
> 
>   Modified files:
>     lib/sysinstall       Feature.cc 
>   Log:
>   Allow feature versions of the type
>   	[0-9][0-9A-Za-z]*([._][0-9A-Za-z]+)*
>   
>   - at least one digit at the beginning
>   - then digits or letters are allowed (some packages use them
>     for versioning *sigh*)
>   - dots or underscores are allowed, but not at the end
>     of the version string
>   
>   Note that we probably want to check compare serials of
>   packages to find out if a package is newer/older/equal.
>   
>   Revision  Changes    Path
>   1.24      +15 -6     libh/lib/sysinstall/Feature.cc
> 
> 
> To Unsubscribe: send mail to [email protected]
> with "unsubscribe freebsd-libh" in the body of the message
> 


-- 
From the age of uniformity, from the age of solitude, from the age of
Big Brother, from the age of doublethink - greetings!

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (FreeBSD)

iD8DBQE9kwhbttcWHAnWiGcRAlyXAJoDV/85uK7h0MwvudHSHGOGWF8HfgCggKeb
0l8+FxbrBhJ3RoyP89jDWPY=
=xoUb
-----END PGP SIGNATURE-----