Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 21 Nov 1997 17:22:42 -0700
From:      Nate Williams <nate@mt.sri.com>
To:        Richard Wackerbarth <rkw@dataplex.net>
Cc:        Don Lewis <Don.Lewis@tsc.tdk.com>, Nate Williams <nate@mt.sri.com>, Brian Somers <brian@awfulhak.org>, freebsd-stable@freebsd.org
Subject:   Re: Version Resolution?
Message-ID:  <199711220022.RAA16923@mt.sri.com>
In-Reply-To: <l03110701b09bcac5add4@[208.2.87.4]>
References:  <199711212307.PAA15838@salsa.gv.tsc.tdk.com> <rkw@dataplex.net> <199711212310.QAA16353@mt.sri.com> <l03110701b09bcac5add4@[208.2.87.4]>

next in thread | previous in thread | raw e-mail | index | archive | help
> >> } >And if you don't do this, how do you get the right token if you export
> >> } >a version from CVS using -r <branch_tag>:<date> to get something other
> >> } >than the head of a branch?
> >> }
> >> } You don't. The purpose of this hack is to distinguish between two
> >> } different source trees which are extracted using exactly the same
> >> } (head of a branch) tag, but at different times.
> >> }
> >> } We should also be using a CVS/RCS macro to tell us what tag we used for
> >> } the extraction. This information should also get included in the system
> >> } identification.
> >>
> >> Couldn't this solve the original problem as well?  If you are extracting
> >> at the head of a branch, we've already got the branch tag info, so we
> >> just need to set the date macro to the time of the extraction.
> 
> No! The time of cvs extraction is not what matters at all.

Right, but the state of the CVS tree at the time of extraction does
matter.  So, if CVS generates the information based on the source
information and not the time of day we have a solution.  (Except for the
non-CVS bit).

> >What about the folks that don't use CVS to get the bits?  The CVS macro
> >doesn't exist for them. ;(
> 
> That doesn't matter. The people who don't get their bits from a CVS tree
> get them from SOMETHING that got them from a CVS tree.

Except in the case of CVSup, which wouldn't (shouldn't) know how to
generate this RCS/CVS macro to modify files and essentially 're-create'
the same information that CVS does. :(

> However, this ignores the real problem which is that you don't care when the
> stuff is EXTRACTED. You care when it was last updated.
> 
> By analogy --
> 
> "Why is this milk sour? I just opened the bottle 5 minutes ago."
> 
> "Could it have anything to do with this September date on the bottle?"

Aka, could it have something to do with the contents of the bottle, not
when you opened it up?  Just because you left it out overnight doesn't
mean it's fresh. :)


Nate



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199711220022.RAA16923>