GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
http://mpa.itc.it/markus/grass50/
Markus
On Wed, Oct 15, 2003 at 06:48:10PM +0200, Markus Neteler wrote:
GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
http://mpa.itc.it/markus/grass50/
Good.
It is nice to see that in got smaler in this RC cycle.
I'm wondering if we should also announce this to the user list
and other lists.
More testing means we could find showstopper bugs more easily.
Also with more annoucements like this coming,
what about a GRASS announce list?
On Fri, Oct 17, 2003 at 07:22:25PM +0200, Bernhard Reiter wrote:
On Wed, Oct 15, 2003 at 06:48:10PM +0200, Markus Neteler wrote:
> GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
> http://mpa.itc.it/markus/grass50/Good.
It is nice to see that in got smaler in this RC cycle.
I'm wondering if we should also announce this to the user list
and other lists.
More testing means we could find showstopper bugs more easily.Also with more annoucements like this coming,
what about a GRASS announce list?
Date: Tue, 21 Oct 2003 11:16:51 +0200
http://article.gmane.org/gmane.comp.gis.grass.user/2921/match=rc5
So far we had once response ("ok").
After that the r.bilinear bugfix went in (tested, ok.).
I suggest to release now.
Markus
On Wed, Oct 29, 2003 at 09:27:53AM +0100, Markus Neteler wrote:
On Fri, Oct 17, 2003 at 07:22:25PM +0200, Bernhard Reiter wrote:
> On Wed, Oct 15, 2003 at 06:48:10PM +0200, Markus Neteler wrote:
> > GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
> > http://mpa.itc.it/markus/grass50/
>
> Good.
> It is nice to see that in got smaler in this RC cycle.
> I'm wondering if we should also announce this to the user list
> and other lists.
> More testing means we could find showstopper bugs more easily.
Date: Tue, 21 Oct 2003 11:16:51 +0200
So far we had once response ("ok").
After that the r.bilinear bugfix went in (tested, ok.).
What about Hamish's tcltk comment?
> > GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
> > http://mpa.itc.it/markus/grass50/
...
I suggest to release now.
Debian/testing:
Compiles, starts
NVIZ works as long as you compile with TclTk 8.3 includes
d.text.freetype works
Redhat 7.3:
Compiles, starts
NVIZ works
d.text.freetype works
Redhat 9:
Compiles, starts
NIVZ works
d.text.freetype works
Is there a list of important things we should test?
(I know.. see ChangeLog/NEWS)
NEWS.html, insert date:
[GRASS 5.0.3 released XX October 2003]
regards,
Hamish
Markus Neteler wrote:
> > GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
> > http://mpa.itc.it/markus/grass50/
>
> Good.
> It is nice to see that in got smaler in this RC cycle.
> I'm wondering if we should also announce this to the user list
> and other lists.
> More testing means we could find showstopper bugs more easily.
>
> Also with more annoucements like this coming,
> what about a GRASS announce list?
>Date: Tue, 21 Oct 2003 11:16:51 +0200
http://article.gmane.org/gmane.comp.gis.grass.user/2921/match=rc5So far we had once response ("ok").
After that the r.bilinear bugfix went in (tested, ok.).
I suggest to release now.
I've just noticed a bug which appears to have been introduced when the
bulk of the grass5 script was moved to etc/Init.sh (and hence is
present in 5.0.2). src/general/init/Gmakefile has this:
$(SHELL) -c "sed -e \"s#GISBASE_VALUE#$(GISBASE)#\" -e \"s#PERL_COMMAND#$(PERL)#\" grass.src > $(GRASS_BIN)/$(START_UP) 2>/dev/null ; true"
However, the line which this is fixing:
GRASS_PERL=PERL_COMMAND
has moved from grass.src to init.sh.
BTW, the GRASS_WISH bug still hasn't been fixed. I'll commit a fix for
that to the HEAD, but it definitely needs testing before inclusion in
the release branch.
--
Glynn Clements <glynn.clements@virgin.net>
On Thu, Oct 30, 2003 at 04:11:50AM +0000, Glynn Clements wrote:
Markus Neteler wrote:
> > > GRASS 5.0.3 release candidate 5 (RC5) is available for testing:
> > > http://mpa.itc.it/markus/grass50/
[...]
>
> After that the r.bilinear bugfix went in (tested, ok.).
>
> I suggest to release now.I've just noticed a bug which appears to have been introduced when the
bulk of the grass5 script was moved to etc/Init.sh (and hence is
present in 5.0.2). src/general/init/Gmakefile has this:$(SHELL) -c "sed -e \"s#GISBASE_VALUE#$(GISBASE)#\" -e \"s#PERL_COMMAND#$(PERL)#\" grass.src > $(GRASS_BIN)/$(START_UP) 2>/dev/null ; true"
However, the line which this is fixing:
GRASS_PERL=PERL_COMMAND
has moved from grass.src to init.sh.
BTW, the GRASS_WISH bug still hasn't been fixed. I'll commit a fix for
that to the HEAD, but it definitely needs testing before inclusion in
the release branch.
I really tend to release today, since there is no bugfix activity any more.
In favour of our users, we should not postpone releases forever, but be a
bit more reliable (not to offend anyone here, just a general statement).
The 5.0.3 release I have started in August:
2003-08-20 10:10 markus
Now we have November and it is still not published.
IMHO we should revise our release strategy.
Markus