This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Pending packages status

Christopher Faylor wrote:

I'm sure that everyone here gets this point.  There is a *potential*
that the same need that we have for incrementing our standard releases
from -1 to -2, etc.  might be an issue for cygwin-apps.  As is said,
above "I don't remember any confusion caused by the current practice."
I suspect that's because the class of user here (at least for those
doing the review) is a few thousand steps above the standard cygwin
user and can manage with things like "date and time" rather than
-1, -2, -3.


So, when you upload your changes, adjust the version to the next
released version.

This causes problems for me, which I won't bore you with. Mainly having to do with the extreme slowness of my only remaining Windows machine, and rebuilding time, and having to babysit the libtool 'make check'.

However, I don't really care.  If you think this is the only way for
you to manage your libtool issues, then use whatever works.


The cygwin user base != the cygwin-apps maintainers.  We are supposed to
be a breed apart.  That's why I don't like the idea of adding, IMO,
silly rules that every "How does it look now" release prior to the
official release needs to be incremented on the off chance that someone
here will be terminally confused and not realize that they might not
have the most up-to-date version.  I can easily imagine the "I can't
review this because you didn't bump the number from -1 to -2" cropping
up.  That just delays the process of getting a package released.

Granted. IMO, if a potential maintainer is happy to keep prerelease versions at a constant REL, that's his lookout. If I bump REL everytime I make a minor pretest/prerelease change, that's MY lookout.

No silly rules either way. In Pavel's original message, he was chiding the packager, who HAD bumped REL, to keep it constant. *THAT's* what I'm against -- forcing the maintainer/packager to do it one way, or the other. BOTH "rules" have problems, and it's up to the maintainer to decide which set of problems she'd rather deal with -- because it's her job to do so; not mine or Pavel's or cgf's.

So, as usual, I opt for flexibility (anyone want to guess at my
political leanings?).  I don't think anyone should be forced to use this
method.  If it makes people more comfortable to bump their versions,
then have at it.

Yeah, what he said (I swear, I wrote my "No silly rules" paragraph before reading cgf's "So, as usual" one). Wanna guess my leanings?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]