On Mon, Mar 20, 2006 at 02:07:02PM +0100, Martin Wegmann wrote:
On Monday 20 March 2006 09:33, Radim Blazek wrote:
> On 3/19/06, Markus Neteler <neteler@itc.it> wrote:
> > Hi,
> >
> > Martin Wegmann and me would like to suggest to
> > open a new "addons" CVS repository in parallel
> > to grass6/.
> > We are receiving more frequently now new code
> > submissions which may need to be worked on.
> > An example is r.li where other parties want to
> > add more fragmentation indices etc. Since r.li
> > with a single index isn't ready for the main
> > CVS in my opinion, we should help people to
> > use the CVS capabilities for distributed
> > development. However, we should also seek to
> > accumulate good stuff in a single place rather
> > than having pieces scattered over the internet.
> >
> > Suggestions:
> > - add new GRASS Extensions Manager (GEM) to main CVS
> > when available and functional
>
> Yes GEM should be added to CVS.
>
> > - create new "addons" (or however named) repo to
> > CVS
>
> I think that CVS is too rigid for this purpose. I think that to add
> a new or modified new module to the repository should be
> as easy as possible, something like Wiki where everybody
> can register and add a module.
> We cannot give CVS account to everybody who wrote a simple
> script but we want to make the script available for everybody.
I think first it has to be discussed if /addons should be just a place to
store scripts or rather to store whole modules for one task (e.g. landscape
analysis, hydrology etc.) which can be loaded afterwards via GEM if needed.
For scripts an /addons repository would be too much but for GRASS function
bundles it is appropriate.
This would make the main GRASS more slim and each additional function has to
be loaded like in R.
I think these GRASS function bundles need to have a central place where they
can be developed like in cvs or svn and not lost over time. However your
point that everybody can not granted cvs access is understandable - either
Of course CVS write access to the main repository will be handled
rather restrictive.
But suggestion below:
there does an alternative exist (separate cvs system for add-ons) or a
totally different system to manage these files (e.g. wiki or webcollab
http://webcollab.sourceforge.net/index.html but this is rather for single
file development). But the latter option would not allow a "real continuous"
development like cvs/svn.
The complexity of cvs (installation/usage) has to be kept in mind as well, but
I am not aware of any easy alternative where somebody can develop code like
in cvs.
Jan Wagner suggested offlist to me to think about SVN migration.
This will offer the possibility to grant write access on a
per-directory base.
In any case an extention manager PERSON is needed. Already today.
Does anybody know how R is managing its /contrib repositories? Does a person
exist who add new modules to cvs which he receives from contributing users?
Or does every person has access to the repository?
Maybe Roger has some insights.
Markus
regards, Martin
> When a module is mature enough and it is of general purpose
> it can be added directly to GRASS CVS.
>
> > - use this as "incubator" (Apache speech)
> > - people can use GEM to add stuff from addons into
> > their copy of GRASS
> > - move stable code, if of interest to the general
> > GRASS community, into main CVS repo
> >
> > This is very similar to the old src.contrib/ in GRASS
> > 4.x and 5.x. But in this case it would be a separate
> > CVS repository to not fill the main CVS repo with
> > less relevant or unmaintained code.
> >
> > r.lake, r.li and other goodies from the addons section
> > are good candidates for a new "addons" CVS repository.
> >
> > Opinions welcome...
> >
> > Markus
> >
> > _______________________________________________
> > grass5 mailing list
> > grass5@grass.itc.it
> > http://grass.itc.it/mailman/listinfo/grass5
>
> _______________________________________________
> grass5 mailing list
> grass5@grass.itc.it
> http://grass.itc.it/mailman/listinfo/grass5
--
Martin Wegmann
DLR - German Aerospace Center
German Remote Sensing Data Center
@
Dept.of Geography
Remote Sensing and Biodiversity Unit
&&
Dept. of Animal Ecology and Tropical Biology
University of Wuerzburg
Am Hubland
97074 Würzburg
phone: +49-(0)931 - 888 4797
mobile: +49-(0)175 2091725
fax: +49-(0)931 - 888 4961
http://www.biota-africa.org
http://www.biogis.de
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
--
Markus Neteler <neteler itc it> http://mpa.itc.it
ITC-irst - Centro per la Ricerca Scientifica e Tecnologica
MPBA - Predictive Models for Biol. & Environ. Data Analysis
Via Sommarive, 18 - 38050 Povo (Trento), Italy