A couple of things were fixed:
* gis.m fixes for geo rectifier (Michael)
* gis.m zoom fixes (Michael)
* install of include files problem on Solaris (Glynn)
* Handle non-standard ETRS_1989 datum name (Paul)
* d.histogram: clear just the current frame, not the full screen (Hamish)
* i.group: fix subgroup listing (Hamish)
* ps.map: broken for named paper sizes (Hamish, Glynn)
* gis.m: meaningful error messages on startup (Michael)
* v.db.select: fix SQL where= option error (Hamish)
* Add module r.to.rast3elev from Soeren
* crash v.in.ogr with gcc4.1.x and non-C locale (Andrey Kiselev)
* v.in.db: with where="condition" it copies the entire attribute table (Martin)
On Thu, Dec 07, 2006 at 05:39:14PM +1300, Hamish wrote:
Markus Neteler wrote:
> a first release candidate of the new 6.2.1 has been published
> today:
..
> See also:
> http://grass.gdf-hannover.de/wiki/GRASS_6.2_Feature_Plan#6.2.1RC1
>
> Please test this release candidate as much as possible.
> Thanks to all contributors and testers.
great.
If there are no problems, will we skip more RCs and release 6.2.1 soon?
That's fine.
But so far we had zero feedback if 6.2.1RC1 works.
Especially the gis.m needs a test.
A couple of things were fixed:
* gis.m fixes for geo rectifier (Michael)
* gis.m zoom fixes (Michael)
* install of include files problem on Solaris (Glynn)
* Handle non-standard ETRS_1989 datum name (Paul)
* d.histogram: clear just the current frame, not the full screen (Hamish)
* i.group: fix subgroup listing (Hamish)
* ps.map: broken for named paper sizes (Hamish, Glynn)
* gis.m: meaningful error messages on startup (Michael)
* v.db.select: fix SQL where= option error (Hamish)
* Add module r.to.rast3elev from Soeren
* crash v.in.ogr with gcc4.1.x and non-C locale (Andrey Kiselev)
* v.in.db: with where="condition" it copies the entire attribute table (Martin)
On Sat, 2006-12-09 at 16:06 +0100, Markus Neteler wrote:
On Sat, Dec 09, 2006 at 06:26:11AM -0800, Brad Douglas wrote:
> On Sat, 2006-12-09 at 11:42 +0100, Sören Gebbert wrote:
> > Hi Markus,
> > the lates grass-test-suite results for grass-6.2.1RC1 are available here:
> > http://www-pool.math.tu-berlin.de/~soeren/grass/GRASS_TestSuite/html_grass-6.2/
> >
> > A test for r.to.rast3elev was added.
> > No errors found.
>
> I cannot speak highly enough of the test suite. It lends credibility
> and saves time.
>
> Markus,
>
> I also built successfully. You may no longer make claims of lacking
> feedback.
Excellent!
Now we only need to receive a quick test of gis.m.
I spent 5-10 minutes testing gis.m. It appears in working order, but I
only tested basics.
--
Brad Douglas <rez touchofmadness com> KB8UYR/6
Address: 37.493,-121.924 / WGS84 National Map Corps #TNMC-3785
But so far we had zero feedback if 6.2.1RC1 works.
Especially the gis.m needs a test.
I am travelling, making detailed testing awkward, but so far:
- on MacOS X 10.4.8 (Intel):
- built cleanly
- gis.m starts up fine, displays raster and vector maps as expected, quick test of thematic maps fine, legends find, sketchy random tests of running miscellaneous modules fine, random help links OK too
- nviz passes light testing
- on MacOSX Server 10.4.8 (PPC):
- builds fine - no further testing yet
- on Fedora Core 5 (Intel):
- builds fine, gis.m starts up fine, but slow speed of connection between hotel and lab preventing proper testing of GUI stuff
- on Debian (stable) (Intel):
- still building (old, slow machine)
I could do more targeted testing on the machine I have with me tonight/tomorrow, but as a mostly command line user, hints as to what is most useful to test in gis.m would help.
* gis.m fails with a nice message if g.region or libgdal is missing.
- not familiar enough with Zooming issues to test that.
One bug remainning in gis.m AFAIK is that in the constrained zoom mode
the region aspect ratio is not preserved during single-click zoom
in/out (and the aspect ratio must be preserved, because it is the
*constrained zoom mode*).
Other zooming issues I used to complain about seem to have been fixed
by Michael. Great.
* `nviz vol=` gives a warning that it won't work
There is still a bug in NVIZ, that usually after adding a vector point
layer, removing it, and adding a new one, artifacts are displayed,
including:
1. different icons then the chosen one, for some points
2. some points are displayed with random colors, not the one specified
3. line artifact happen to pop up
Hamish wrote:
> - not familiar enough with Zooming issues to test that.
One bug remainning in gis.m AFAIK is that in the constrained zoom mode
the region aspect ratio is not preserved during single-click zoom
in/out (and the aspect ratio must be preserved, because it is the
*constrained zoom mode*).
Other zooming issues I used to complain about seem to have been fixed
by Michael. Great.
> * `nviz vol=` gives a warning that it won't work
There is still a bug in NVIZ, that usually after adding a vector point
layer, removing it, and adding a new one, artifacts are displayed,
including:
1. different icons then the chosen one, for some points
2. some points are displayed with random colors, not the one specified
3. line artifact happen to pop up
- not familiar enough with Zooming issues to test that.
One bug remainning in gis.m AFAIK is that in the constrained zoom mode
the region aspect ratio is not preserved during single-click zoom
in/out (and the aspect ratio must be preserved, because it is the
*constrained zoom mode*).
Other zooming issues I used to complain about seem to have been fixed
by Michael. Great.
* `nviz vol=` gives a warning that it won't work
There is still a bug in NVIZ, that usually after adding a vector point
layer, removing it, and adding a new one, artifacts are displayed,
including:
1. different icons then the chosen one, for some points
2. some points are displayed with random colors, not the one specified
3. line artifact happen to pop up
do both of these issues exist in 6.2 and 6.3, or have they been fixed in
6.3 but not backported?
I did not test it with spearfish, but I can confirm that there is a problem still in 6.3
My problem was not the same - icons, colors worked fine, but when I chose x as
a symbol all points were displayed under the surface, using other symbols they were
draped properly: here is the image of what I tested it with (I sneak preview
of the new data set we plan to replace the spearfish with)
It would be useful if others tested this with different types of data and sequence
of operations - this happened after I deleted and reloaded the data as Maciek has indicated.
On 12/11/06 5:41 PM, "Helena Mitasova" <hmitaso@unity.ncsu.edu> wrote:
I did not test it with spearfish, but I can confirm that there is a
problem still in 6.3
My problem was not the same - icons, colors worked fine, but when I
chose x as
a symbol all points were displayed under the surface, using other
symbols they were
draped properly: here is the image of what I tested it with (I sneak
preview
of the new data set we plan to replace the spearfish with)
It would be useful if others tested this with different types of data
and sequence
of operations - this happened after I deleted and reloaded the data
as Maciek has indicated.
Helena,
I checked the x icon in a current (as of last Friday) version of NVIZ. On my
system (Mac OSX with x11 version of GRASS), I have the opposite problem from
you. The x's float above the landscape. I think the problem is that they are
trying to conform to the landscape rather than be a 3D icon like the rest of
them. This has been a long-standing problem AFAICT. This is one of the
reasons I switched the default icon from X to sphere (the other is that the
x's are difficult to see). You can see a screenshot of the floating x's at
<http://www.public.asu.edu/~cmbarton/files/nviz_screenshots/floating_x.jpg>
BTW, the new landscapes look very cool. Where are they from? Switching the
demo set from Spearfish to someplace else after over 20 years is BIG news
for GRASS. Is the city of Spearfish protesting?
Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution & Social Change
Center for Social Dynamics and Complexity
Arizona State University
There is still a bug in NVIZ, that usually after adding a vector point
layer, removing it, and adding a new one, artifacts are displayed,
including:
1. different icons then the chosen one, for some points
2. some points are displayed with random colors, not the one specified
3. line artifact happen to pop up
On Tue, 2006-12-12 at 17:56 +0100, Maciej Sieczka wrote:
Hamish wrote:
> Maciej Sieczka wrote:
>> There is still a bug in NVIZ, that usually after adding a vector point
>> layer, removing it, and adding a new one, artifacts are displayed,
>> including:
>>
>> 1. different icons then the chosen one, for some points
>> 2. some points are displayed with random colors, not the one specified
>> 3. line artifact happen to pop up
>>
>> See the attached screendump (spearfish: elevation.10m + archsites) and
>> latest remarks in the http://intevation.de/rt/webrt?serial_num=4492.
> do both of these issues exist in 6.2 and 6.3, or have they been fixed in
> 6.3 but not backported?
Both issues exist in 6.2 and 6.3.
This appears to be a bug in ogsf (Gp3.c) where it uses the adapted sites
library to load the vector points. I have been testing a modified
version where I load the points with direct calls to the vector library
(not sites). This seems to clear up the reported problem.
Once I have finished testing I will commit to cvs. This will also get
rid of the ogsf dependance on the temporary sites library.
On Tue, 2006-12-12 at 17:56 +0100, Maciej Sieczka wrote:
Both issues exist in 6.2 and 6.3.
This appears to be a bug in ogsf (Gp3.c) where it uses the adapted sites
library to load the vector points. I have been testing a modified
version where I load the points with direct calls to the vector library
(not sites). This seems to clear up the reported problem.
Once I have finished testing I will commit to cvs. This will also get
rid of the ogsf dependance on the temporary sites library.
On 12/11/06 5:41 PM, "Helena Mitasova" <hmitaso@unity.ncsu.edu> wrote:
I did not test it with spearfish, but I can confirm that there is a
problem still in 6.3
My problem was not the same - icons, colors worked fine, but when I
chose x as
a symbol all points were displayed under the surface, using other
symbols they were
draped properly: here is the image of what I tested it with (I sneak
preview
of the new data set we plan to replace the spearfish with)
It would be useful if others tested this with different types of data
and sequence
of operations - this happened after I deleted and reloaded the data
as Maciek has indicated.
Helena,
I checked the x icon in a current (as of last Friday) version of NVIZ. On my
system (Mac OSX with x11 version of GRASS), I have the opposite problem from
you. The x's float above the landscape. I think the problem is that they are
trying to conform to the landscape rather than be a 3D icon like the rest of
them. This has been a long-standing problem AFAICT. This is one of the
reasons I switched the default icon from X to sphere (the other is that the
x's are difficult to see). You can see a screenshot of the floating x's at
<http://www.public.asu.edu/~cmbarton/files/nviz_screenshots/floating_x.jpg>
BTW, the new landscapes look very cool. Where are they from? Switching the
demo set from Spearfish to someplace else after over 20 years is BIG news
for GRASS. Is the city of Spearfish protesting?
I am not releasing it yet as I am working with it to find out what is the minimum data set
needed to cover the material in the book and tutorials - the amount of
data for this area is huge and we want to keep it easy to handle.
I plan for additional mapsets covering various topics (e.g. terrain data
and analysis from different sources, coastal, urban, etc.) - others will be probably able to
create their own using the public data available.
I think that we will have Spearfish around for a while until the new data set
gets well tested and established.
If anybody is working on any new tutorials or educational materials based on spearfish,
and would rather use the new data set, let me know, I will post what I have now.
Helena
Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution & Social Change
Center for Social Dynamics and Complexity
Arizona State University
--
Helena Mitasova
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
1125 Jordan Hall
NCSU Box 8208
Raleigh, NC 27695-8208 http://skagit.meas.ncsu.edu/~helena/
email: hmitaso@unity.ncsu.edu
ph: 919-513-1327 (no voicemail)
fax 919 515-7802
