Ports with version numbers going backwards: devel/ode
Oliver Eikemeier
eikemeier at fillmore-labs.com
Tue Jun 29 01:09:15 PDT 2004
Am Dienstag den, 29. Juni 2004, um 09:46, schrieb Sergey Matveychuk:
> Erik Trulsson wrote:
>
>> I understand his proposal rather as giving
>> 0.005 < 0.039 < 0.05 = 0.050 < 0.5 < 0.39 < 050 < 0.390 < 0.500
>
> Not exectly.
> 0.005 < 0.039 < 0.05 < 0.050 < 0.5 < 0.39 < 0.50 < 0.390 < 0.500
>
> Ending zoros can't be droped.
Ok, it was fun discussing this, and I admit it's a nifty idea. The
problems with that are:
- it breaks backward compatibility (and tools like portupgrade have to
be adapted to the new rules)
- it is of limited use, e.g. only when leading zeroes in a version
number are dropped *and* the resulting version number is smaller than
the previous one.
- it is another addition the the already non-trivial version number
parsing rules
So, do we expect enough benefits from this change to actually accept the
costs, or do we just bump the PORTEPOCH from time to time (or force
ports to use .500 instead of .5 when the previous version was .039)?
Btw, normally portlint should warn you of such issues when there is a
more or less up-to-date INDEX on the machine.
-Oliver
More information about the freebsd-ports
mailing list