Glynn,
thanks for updating the release branch.
What I do not fully understand:
On Sun, Jun 16, 2002 at 05:29:52PM +0200, grass@intevation.de wrote:
Author: glynn
Update of /grassrepository/grass/src/CMD/generic
In directory doto:/tmp/cvs-serv19045/src/CMD/generic
Modified Files:
Tag: releasebranch_26_april_2002_5_0_0
[ files]
Sync with CVS head
why 'releasebranch_26_april_2002_5_0_0'? We had a more recent
branch already:
GRASS 5.0.0pre4: has been tagged as "release_13_may_2002_grass5_0_0_pre4"
http://grass.itc.it/grasscvstags.html
Certainly I am missing something here... Above web page, although
written by me, confuses me at time.
Markus
Markus Neteler wrote:
thanks for updating the release branch.
What I do not fully understand:
On Sun, Jun 16, 2002 at 05:29:52PM +0200, grass@intevation.de wrote:
> Author: glynn
>
> Update of /grassrepository/grass/src/CMD/generic
> In directory doto:/tmp/cvs-serv19045/src/CMD/generic
>
> Modified Files:
> Tag: releasebranch_26_april_2002_5_0_0
[ files]
> Sync with CVS head
why 'releasebranch_26_april_2002_5_0_0'? We had a more recent
branch already:
GRASS 5.0.0pre4: has been tagged as "release_13_may_2002_grass5_0_0_pre4"
This isn't a branch tag. This is a normal (non-branch) tag which was
applied to those files which constituted the pre4 release.
If you use:
cvs update -r releasebranch_26_april_2002_5_0_0
you will get the latest versions from the release branch; more or less
the files that will constitute pre5.
If you use:
cvs update -r release_13_may_2002_grass5_0_0_pre4
you should get the files which constitute pre4.
--
Glynn Clements <glynn.clements@virgin.net>
On Mon, Jun 17, 2002 at 04:57:58PM +0100, Glynn Clements wrote:
Markus Neteler wrote:
> thanks for updating the release branch.
> What I do not fully understand:
>
> On Sun, Jun 16, 2002 at 05:29:52PM +0200, grass@intevation.de wrote:
> > Author: glynn
> >
> > Update of /grassrepository/grass/src/CMD/generic
> > In directory doto:/tmp/cvs-serv19045/src/CMD/generic
> >
> > Modified Files:
> > Tag: releasebranch_26_april_2002_5_0_0
> [ files]
> > Sync with CVS head
>
> why 'releasebranch_26_april_2002_5_0_0'? We had a more recent
> branch already:
>
> GRASS 5.0.0pre4: has been tagged as "release_13_may_2002_grass5_0_0_pre4"
This isn't a branch tag. This is a normal (non-branch) tag which was
applied to those files which constituted the pre4 release.
If you use:
cvs update -r releasebranch_26_april_2002_5_0_0
you will get the latest versions from the release branch; more or less
the files that will constitute pre5.
If you use:
cvs update -r release_13_may_2002_grass5_0_0_pre4
you should get the files which constitute pre4.
Ah yes, now I recall that (sorry). Maybe we should improve the
web page to be more clear. Suggestions welcome (or directly
modify the file in web CVS).
Markus
On Jun 17, Glynn Clements wrote:
> If you use:
>
> cvs update -r release_13_may_2002_grass5_0_0_pre4
>
> you should get the files which constitute pre4.
Why do these tags have two independent naming schemes embedded within
them:
1) a date
2) a release number
I can imagine reasons it might be nice to have tags that include dates
in them, (for example to be able to reproduce a hypothetical nightly
"snapshot" tarball). Otherwise, a tag with a date doesn't seem to
helpful since CVS can provide code for a specific date anyway.
Getting to the point, currently it's hard to deduce the tag for a
given release, (one must consult a list giving the dated tag name
above).
I would prefer a system that had simpler tag names that a mere mortal
could actually predict such as:
cvs update -r release_5_0_0_pre4
-Carl
--
Carl Worth
USC Information Sciences Institute cworth@east.isi.edu
3811 N. Fairfax Dr. #200, Arlington VA 22203 703-812-3725
Historical reasons.
The grass development team did not have enough experience with
branches then.
But it is normal that you have to consult the branch tag information
list if you participate in the development. 
On Mon, Jun 17, 2002 at 06:35:14PM +0000, Carl Worth wrote:
On Jun 17, Glynn Clements wrote:
> cvs update -r release_13_may_2002_grass5_0_0_pre4
Why do these tags have two independent naming schemes embedded within
them: