Moritz Lennert wrote:
Markus Neteler said:
> On Fri, Apr 11, 2003 at 12:47:03PM +0200, Moritz Lennert wrote:
>> Markus Neteler said:
>>
>> > On Fri, Apr 11, 2003 at 11:25:33AM +0200, Markus Neteler wrote:
Moritz, when you are at it, can you also check the r.mapcalc to make
sure that it has all fixes (e.g. it writes the expression to the history
file?) and run nviz just to view some surface, to make sure that nothing
fundamental is broken there?
thank you,
Helena
It installs without a glitch and rapid testing (d.*, r.thin, r.line, d.dm,
nviz, g.select.pg, d.legend, d.what.vect (although I don't see any
"flashing"...)) does not show any problems.
Only found bugs:
- when I launch v.to.db without any arguments, I get:
PROGRAMMER ERROR: first item in gisprompt is <vector>
Must be either new, old, mapset, or any
- the announced v.in.dgn module does no seem to be available
Moritz
Moritz
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
Hi,
Testing results for GRASS 5.0.2 on SuSe
1) SuSe 8.0:
GRASS 5.0.2 on SuSe 8.0 works fine.
- r.mapcalc writes the expression to the history file
- for NVIZ I needed 2 links to make it work
ln -s /usr/lib/libtk8.3.so /usr/lib/libtk.so.0
ln -s /usr/lib/libtcl8.3.so /usr/lib/libtcl.so.0
- PNG Treiber is OK
- Postgres works
- all other modules I was testing do not make any problems
2) SuSe 8.1:
Same as 8.0 but NVIZ does not work at all. I guess it is again
the tcl/tk8.4 problem, since you compiled with 8.3 ...
- everything else seems to be fine.
bye
Otto
Helena Mitasova said:
Moritz Lennert wrote:
Markus Neteler said:
> On Fri, Apr 11, 2003 at 12:47:03PM +0200, Moritz Lennert wrote:
>> Markus Neteler said:
>>
>> > On Fri, Apr 11, 2003 at 11:25:33AM +0200, Markus Neteler wrote:
Moritz, when you are at it, can you also check the r.mapcalc to make
sure that it has all fixes (e.g. it writes the expression to the history
file?) and run nviz just to view some surface, to make sure that nothing
fundamental is broken there?
I'd already tested nviz, and everything seems to work fine for basic
operation.
r.mapcalc writes into the history file (as Otto already mentioned).
However, I can't use arrow keys in the interactive session. Is this normal
? (I don't know r.mapcalc that well).
Moritz
thank you,
Helena
It installs without a glitch and rapid testing (d.*, r.thin, r.line,
d.dm,
nviz, g.select.pg, d.legend, d.what.vect (although I don't see any
"flashing"...)) does not show any problems.
Only found bugs:
- when I launch v.to.db without any arguments, I get:
PROGRAMMER ERROR: first item in gisprompt is <vector>
Must be either new, old, mapset, or any
- the announced v.in.dgn module does no seem to be available
Moritz
Moritz
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
Moritz Lennert wrote:
Helena Mitasova said:
> Moritz Lennert wrote:
>>
>> Markus Neteler said:
>>
>> > On Fri, Apr 11, 2003 at 12:47:03PM +0200, Moritz Lennert wrote:
>> >> Markus Neteler said:
>> >>
>> >> > On Fri, Apr 11, 2003 at 11:25:33AM +0200, Markus Neteler wrote:
> Moritz, when you are at it, can you also check the r.mapcalc to make
> sure that it has all fixes (e.g. it writes the expression to the history
> file?) and run nviz just to view some surface, to make sure that nothing
> fundamental is broken there?
I'd already tested nviz, and everything seems to work fine for basic
operation.
r.mapcalc writes into the history file (as Otto already mentioned).
However, I can't use arrow keys in the interactive session. Is this normal
? (I don't know r.mapcalc that well).
I would say that is a feature as it does not work in the old version of
r.mapcalc (pre3).
I don't think there is need to do anything about it as most people
use r.mapcalc on command line anyway,
thank you for testing
Helena
Moritz
>
> thank you,
>
> Helena
>>
>> It installs without a glitch and rapid testing (d.*, r.thin, r.line,
>> d.dm,
>> nviz, g.select.pg, d.legend, d.what.vect (although I don't see any
>> "flashing"...)) does not show any problems.
>>
>> Only found bugs:
>>
>> - when I launch v.to.db without any arguments, I get:
>>
>> PROGRAMMER ERROR: first item in gisprompt is <vector>
>> Must be either new, old, mapset, or any
>>
>> - the announced v.in.dgn module does no seem to be available
>
>>
>> Moritz
>>
>> Moritz
>>
>> _______________________________________________
>> grass5 mailing list
>> grass5@grass.itc.it
>> http://grass.itc.it/mailman/listinfo/grass5
>
On Fri, Apr 11, 2003 at 07:12:38PM +0200, Moritz Lennert wrote:
[...]
However, I can't use arrow keys in the interactive session. Is this normal
? (I don't know r.mapcalc that well).
AFAIK this only works if GRASS was compiled with readline support
(which I didn't specify, should I?).
Markus
Moritz Lennert wrote:
> Moritz, when you are at it, can you also check the r.mapcalc to make
> sure that it has all fixes (e.g. it writes the expression to the history
> file?) and run nviz just to view some surface, to make sure that nothing
> fundamental is broken there?
I'd already tested nviz, and everything seems to work fine for basic
operation.
r.mapcalc writes into the history file (as Otto already mentioned).
However, I can't use arrow keys in the interactive session. Is this normal
? (I don't know r.mapcalc that well).
It's normal; you have to use --with-readline to get history support,
but --with-readline has problems of its own. Readline support was
added in 5.0.0, but isn't enabled by default.
--
Glynn Clements <glynn.clements@virgin.net>
Markus Neteler wrote:
> However, I can't use arrow keys in the interactive session. Is this normal
> ? (I don't know r.mapcalc that well).
AFAIK this only works if GRASS was compiled with readline support
(which I didn't specify, should I?).
No; that would result in an r.mapcalc binary that only works if you
have the correct version of libreadline.
--
Glynn Clements <glynn.clements@virgin.net>
The Baylor University website still shows 5.0.1 as the current version on
the main page and in the source link. In fact the last news update was in
January announcing 5.0.1.
This is a long term problem, we are aware of it and in regular contact
with Bruce trying to get it automated.
Bernhard, did you make any progress with Bruce on his rsync problem?
thanks,
Helena
cheg01@attbi.com wrote:
The Baylor University website still shows 5.0.1 as the current version on
the main page and in the source link. In fact the last news update was in
January announcing 5.0.1.
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
Related to the recent 5.0.2 release we need volunteers
to provide binaries for
- SGI/Irix
- winGRASS
- X11 display driver
- generic display driver
- Mac OSX
Some RPMs (Linux/x86 and PPC) are probably generated for
the next FreeGIS release?
Markus
On Sat, Apr 12, 2003 at 07:26:45PM -0400, Helena wrote:
This is a long term problem, we are aware of it and in regular contact
with Bruce trying to get it automated.
Bernhard, did you make any progress with Bruce on his rsync problem?
No.
He still wanted to get back to me.
cheg01@attbi.com wrote:
>
> The Baylor University website still shows 5.0.1 as the current version on
> the main page and in the source link. In fact the last news update was in
> January announcing 5.0.1.
On Mon, Apr 14, 2003 at 02:17:12PM +0200, Markus Neteler wrote:
Related to the recent 5.0.2 release we need volunteers
to provide binaries for
- SGI/Irix
- winGRASS
- X11 display driver
- generic display driver
- Mac OSX
Some RPMs (Linux/x86 and PPC) are probably generated for
the next FreeGIS release?
The next FreeGIS-CD (v1.2.3) is in the test period
and will contain GRASS 5.0.1 to make sure we release stable packages
with basic testing.
We don't produce PPC rpms anymore, the last FreeGIS-CD for PPC was 1.0.4.
This is because of the low demand for it.
If you thing we are doing an important job and you're using the packages.
We appreciate tipping: http://freegis.org/about-paying.en.html 
Bernhard
On Mon, Apr 14, 2003 at 03:14:33PM +0200, Bernhard Reiter wrote:
On Mon, Apr 14, 2003 at 02:17:12PM +0200, Markus Neteler wrote:
> Related to the recent 5.0.2 release we need volunteers
> to provide binaries for
>
> - SGI/Irix
> - winGRASS
> - X11 display driver
> - generic display driver
> - Mac OSX
>
> Some RPMs (Linux/x86 and PPC) are probably generated for
> the next FreeGIS release?
The next FreeGIS-CD (v1.2.3) is in the test period
and will contain GRASS 5.0.1 to make sure we release stable packages
with basic testing.
well, this were a pity as 5.0.1 only contains 2 changes. So you better
include 5.0.0 ... 
But why distributing partially oldish code?
I thought the idea of the release branch is to release stable GRASS.
Now FreeGIS doesn't follow the own ideas of it's maintainers (I remember
that you are strongly recommending branches...)?
We don't produce PPC rpms anymore, the last FreeGIS-CD for PPC was 1.0.4.
This is because of the low demand for it.
Ok
Markus
On Mon, Apr 14, 2003 at 03:48:03PM +0200, Markus Neteler wrote:
On Mon, Apr 14, 2003 at 03:14:33PM +0200, Bernhard Reiter wrote:
> The next FreeGIS-CD (v1.2.3) is in the test period
> and will contain GRASS 5.0.1 to make sure we release stable packages
> with basic testing.
well, this were a pity as 5.0.1 only contains 2 changes. So you better
include 5.0.0 ... But why distributing partially oldish code?
I thought the idea of the release branch is to release stable GRASS.
Now FreeGIS doesn't follow the own ideas of it's maintainers (I remember
that you are strongly recommending branches...)?
Of course we are following it.
And the quality of GRASS releases seems to have improved.
To sell a CD like the FreeGIS-CD you need to test it
on different platforms. If at all times improvements are accepted,
testing would start again and you can almost never release.
This why all serious Free Software distributions most of the time
contain older versions. It is a tradeoff decision to make.
Anyway I'm not deeply in the CD releasing process here,
we'll see how far we are and if GRASS 5.0.2 poses problems.
In any case there is going to be a next revision and hopefully
a fast update.
Demand for the CD is not that high that we could affort
more manpower on it.
Bernhard
Markus Neteler wrote:
On Mon, Apr 14, 2003 at 03:14:33PM +0200, Bernhard Reiter wrote:
> On Mon, Apr 14, 2003 at 02:17:12PM +0200, Markus Neteler wrote:
> > Related to the recent 5.0.2 release we need volunteers
> > to provide binaries for
> >
> > - SGI/Irix
> > - winGRASS
> > - X11 display driver
> > - generic display driver
> > - Mac OSX
> >
> > Some RPMs (Linux/x86 and PPC) are probably generated for
> > the next FreeGIS release?
>
> The next FreeGIS-CD (v1.2.3) is in the test period
> and will contain GRASS 5.0.1 to make sure we release stable packages
> with basic testing.
well, this were a pity as 5.0.1 only contains 2 changes. So you better
include 5.0.0 ... But why distributing partially oldish code?
I have already written to Bernhard a suggestion to distribute 5.0.2
as 5.0.0 and 5.0.1 has some serious deficiencies and 5.0.2 is simply
much better.
Often a small modification in the code can make a big difference for
everyday work
and that is the case for 5.0.2 (compared to 5.0.0 or 5.0.1)
So I would just like to ask whoever can test 5.0.2 please do it so that
we can make 5.0.2 the version of GRASS recommended for wide distribution
as a stable and reliable code.
Helena
I thought the idea of the release branch is to release stable GRASS.
Now FreeGIS doesn't follow the own ideas of it's maintainers (I remember
that you are strongly recommending branches...)?
> We don't produce PPC rpms anymore, the last FreeGIS-CD for PPC was 1.0.4.
> This is because of the low demand for it.
Ok
Markus
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
On Mon, Apr 14, 2003 at 04:03:31PM +0200, Bernhard Reiter wrote:
On Mon, Apr 14, 2003 at 03:48:03PM +0200, Markus Neteler wrote:
> On Mon, Apr 14, 2003 at 03:14:33PM +0200, Bernhard Reiter wrote:
> > The next FreeGIS-CD (v1.2.3) is in the test period
> > and will contain GRASS 5.0.1 to make sure we release stable packages
> > with basic testing.
>
> well, this were a pity as 5.0.1 only contains 2 changes. So you better
> include 5.0.0 ... But why distributing partially oldish code?
>
> I thought the idea of the release branch is to release stable GRASS.
> Now FreeGIS doesn't follow the own ideas of it's maintainers (I remember
> that you are strongly recommending branches...)?
Of course we are following it.
And the quality of GRASS releases seems to have improved.
To sell a CD like the FreeGIS-CD you need to test it
on different platforms. If at all times improvements are accepted,
testing would start again and you can almost never release.
This why all serious Free Software distributions most of the time
contain older versions. It is a tradeoff decision to make.
Anyway I'm not deeply in the CD releasing process here,
we'll see how far we are and if GRASS 5.0.2 poses problems.
You are always invited to report problems! Using the feedback
of the development team will hopefully enable you to distribute
5.0.2 instead of 5.0.1.
In any case there is going to be a next revision and hopefully
a fast update.
Demand for the CD is not that high that we could affort
more manpower on it.
That's why we want to help the FreeGIS project
Cheers
Markus
The next FreeGIS-CD (v1.2.3) is in the test period
and will contain GRASS 5.0.1 to make sure we release stable packages
with basic testing.
Is there a standard test suite for new releases that users can run after
building on a new platform? That might shorten the testing period.
On Mon, Apr 14, 2003 at 07:44:38AM -0700, cheg01@attbi.com wrote:
>
> The next FreeGIS-CD (v1.2.3) is in the test period
> and will contain GRASS 5.0.1 to make sure we release stable packages
> with basic testing.
>
Is there a standard test suite for new releases that users can run after
building on a new platform? That might shorten the testing period.
A bit is described in
documents/pretesting_grass.txt
http://freegis.org/cgi-bin/viewcvs.cgi/~checkout~/grass/documents/pretesting_grass.txt
and something more in
documents/release_rules.txt
http://freegis.org/cgi-bin/viewcvs.cgi/~checkout~/grass/documents/release_rules.txt
However, these document should be clarified, shortened etc.
Markus
On Mon, Apr 14, 2003 at 04:15:28PM +0200, Markus Neteler wrote:
On Mon, Apr 14, 2003 at 04:03:31PM +0200, Bernhard Reiter wrote:
> On Mon, Apr 14, 2003 at 03:48:03PM +0200, Markus Neteler wrote:
> > On Mon, Apr 14, 2003 at 03:14:33PM +0200, Bernhard Reiter wrote:
>
> > > The next FreeGIS-CD (v1.2.3) is in the test period
> > > and will contain GRASS 5.0.1 to make sure we release stable packages
> > > with basic testing.
>
> Anyway I'm not deeply in the CD releasing process here,
> we'll see how far we are and if GRASS 5.0.2 poses problems.
You are always invited to report problems! Using the feedback
of the development team will hopefully enable you to distribute
5.0.2 instead of 5.0.1.
The problems are more with the way we try to build packages
that run on many GNU/Linux distributions.
Hi Markus.
- winGRASS
- X11 display driver
- generic display driver
I'll do the generic one after the Easter weekend.
Sorry for the delay,
Mike Thomas