Opened 12 years ago
Closed 11 years ago
#1619 closed defect (fixed)
v.krige won't load: ImportError: No module named globalvar
| Reported by: | momsen | Owned by: | |
|---|---|---|---|
| Priority: | normal | Milestone: | 7.0.0 |
| Component: | Python | Version: | svn-trunk |
| Keywords: | v.krige | Cc: | aghisla |
| CPU: | x86-64 | Platform: | Linux |
Description
Hello, I have installed GRASS 7 this week, along with gdal/geos/spgrass6/etc. I'm on a Debian system. R and many of the dependencies came from Debian packages. But gdal and GRASS I installed from source.
v.krige gives the following errors. R/spgrass6 does work, I used readVECT6 and readRAST6 to load data into R.
Is there something else I should install/change on my side, or is this a bug?
v.krige
Loading dependencies, please wait...
rgeos: (SVN revision 326)
GEOS runtime version: 3.3.1-CAPI-1.7.1
Polygon checking: TRUE
WARNING! if you turn polygon checking off, and polygons are
not valid in GEOS, you risk losing data as your R session may crash!
Loading required package: zoo
Attaching package: ‘zoo’
The following object(s) are masked from ‘package:base’:
as.Date, as.Date.numeric
Loading required package: xts
GRASS GIS interface loaded with GRASS version: GRASS 7.0.svn (2012)
and location: spearfish70
Checking rgeos availability: TRUE
Traceback (most recent call last):
File "/usr/local/grass-7.0.svn/scripts/v.krige", line 411,
in <module>
main()
File "/usr/local/grass-7.0.svn/scripts/v.krige", line 299,
in main
import globalvar
ImportError: No module named globalvar
(Thu Mar 15 14:55:17 2012) Command finished (1 sec)
(Thu Mar 15 15:43:42 2012)
g.gisenv -n
LOCATION_NAME=spearfish70
GISDBASE=/home/emomsen/GRASSDATA
MAPSET=user1
GUI=wxpython
I tried loading the R package dependencies I knew about in R, that seemed to be a common theme in previous v.krige errors. I think I have everything installed.
> sessionInfo() R version 2.14.1 (2011-12-22) Platform: x86_64-pc-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 [7] LC_PAPER=C LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] maptools_0.8-14 lattice_0.20-0 foreign_0.8-49 automap_1.0-12 [5] gstat_1.0-10 spacetime_0.6-0 rgeos_0.2-3 stringr_0.6 [9] xts_0.8-2 zoo_1.7-7 spgrass6_0.7-9 XML_3.9-4 [13] sp_0.9-97 loaded via a namespace (and not attached): [1] grid_2.14.1 plyr_1.7.1
Other dependencies:
emomsen@gis-dev-test:/usr/local/grass-7.0.svn/scripts$ aptitude search rpy2 i python-rpy2 - Python interface to the GNU R language and environment (ver v python2.6-rpy2 - v python2.7-rpy2 -
Anything I should normally be reporting? I checked the subversion number:
root@gis-dev-test:/opt/grass_trunk# svn info Path: . URL: https://svn.osgeo.org/grass/grass/trunk Repository Root: https://svn.osgeo.org/grass Repository UUID: 15284696-431f-4ddb-bdfa-cd5b030d7da7 Revision: 51063 Node Kind: directory Schedule: normal Last Changed Author: martinl Last Changed Rev: 51060 Last Changed Date: 2012-03-14 17:13:32 -0500 (Wed, 14 Mar 2012)
Thanks for any help!
Change History (15)
comment:1 by , 12 years ago
comment:2 by , 12 years ago
It is nice to hear I've finally gotten to the point where my installation wasn't at fault. I have not yet updated/written code for any open source projects, but want to learn. I have a presentation Tuesday, so if someone can provide guidance I could look at it then. If you find time to make the update before that, I'm sure there will be other things I can help with later.
I'm late for leaving for the weekend, I'll be able to check more on zoo and xts later. Both can be loaded directly in R as well, from memory zoo was a dependency for one of the spgrass6 related packages. At least it shows up each time I installed it!
-Eric
comment:3 by , 12 years ago
I am able to load rgeos, gstat, spgrass6, and maptools in a R session.
I found where zoo and xts came from:
automap is optional (provides automatic variogram fit) for v.krige. automap depends on gstat, which depends on spacetime, which depends on zoo and xts.
-Eric
comment:4 by , 12 years ago
| Cc: | added |
|---|
Startup problems fixed in r53127.
Remaining issue: creating the variogram leads to
Traceback (most recent call last):
File "/home/neteler/grass70/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 444, in OnPlotButton
if globals()["InputData"] is None:
KeyError: 'InputData'
Traceback (most recent call last):
File "/home/neteler/grass70/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 444, in OnPlotButton
if globals()["InputData"] is None:
KeyError: 'InputData'
Traceback (most recent call last):
File "/home/neteler/grass70/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 256, in OnRunButton
self.goutput.RunCmd(command, switchPage = True)
File "/home/neteler/grass70/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/gui_core/goutput.py", line 604, in RunCmd
self._notebook.SetSelectionByName('output')
AttributeError: 'FlatNotebook' object has no attribute 'SetSelectionByName'
follow-up: 6 comment:5 by , 12 years ago
I can fix the FlatNotebook part of error (probably tomorrow, I must first download some packages to be able to test it).
Anna
comment:6 by , 12 years ago
Replying to annakrat:
I can fix the FlatNotebook part of error (probably tomorrow, I must first download some packages to be able to test it).
After installation of needed packages I get this error:
GRASS 7.0.svn (spearfish60):/usr/local/src/trunk > v.krige
Loading dependencies, please wait...
rgeos: (SVN revision 348)
GEOS runtime version: 3.3.3-CAPI-1.7.4
Polygon checking: TRUE
Loading required package: zoo
Attaching package: ‘zoo’
The following object(s) are masked from ‘package:base’:
as.Date, as.Date.numeric
Loading required package: xts
GRASS GIS interface loaded with GRASS version: GRASS 7.0.svn (2012)
and location: spearfish60
Checking rgeos availability: TRUE
Loading required package: automap
Traceback (most recent call last):
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/scripts/v.krige", line 414, in <module>
main()
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/scripts/v.krige", line 312, in main
controller = controller)
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 269, in __init__
self.Panel = KrigingPanel(self, Rinstance, controller)
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 98, in __init__
self.CreatePage(package = Rpackage, Rinstance = Rinstance, controller = controller)
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 175, in CreatePage
controller = controller)))
File "/usr/local/src/trunk/dist.i686-pc-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 394, in __init__
ModelFactor = robjects.r.vgm().rx('long')
AttributeError: 'RDataFrame' object has no attribute 'rx'
I have no idea what rx is or how too fix this.
Anna
follow-up: 11 comment:10 by , 11 years ago
| Component: | Default → Python |
|---|---|
| Keywords: | v.krige added; krige removed |
Hi,
current status in trunk, starting from the command line:
(loading the R stuff works for me)
...
Traceback (most recent call last):
File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/scripts/v.krige", line 412, in <module>
main()
File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/scripts/v.krige", line 310, in main
controller = controller)
File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 289, in __init__
self.Panel = KrigingPanel(self, Rinstance, controller)
File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 86, in __init__
self.InputDataColumn.SetSelection(0)
AttributeError: 'ColumnSelect' object has no attribute 'SetSelection'
not sure how to fix that; the same works in devbr6. (v.krige now updated in devbr6 to reflect recent wxgui dir layout changes & starts up ok there)
If I comment out the self.InputDataColumn.SetSelection(0) line the v.krige GUI starts in trunk, so it isn't too badly broken.
Hamish
follow-up: 12 comment:11 by , 11 years ago
Replying to hamish:
Hi,
current status in trunk, starting from the command line:
(loading the R stuff works for me) ... Traceback (most recent call last): File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/scripts/v.krige", line 412, in <module> main() File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/scripts/v.krige", line 310, in main controller = controller) File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 289, in __init__ self.Panel = KrigingPanel(self, Rinstance, controller) File "/home/hamish/src/grass/svn/trunk/dist.x86_64-unknown-linux-gnu/etc/gui/wxpython/scripts/vkrige.py", line 86, in __init__ self.InputDataColumn.SetSelection(0) AttributeError: 'ColumnSelect' object has no attribute 'SetSelection'not sure how to fix that; the same works in devbr6. (v.krige now updated in devbr6 to reflect recent wxgui dir layout changes & starts up ok there)
If I comment out the
self.InputDataColumn.SetSelection(0)line the v.krige GUI starts in trunk, so it isn't too badly broken.
right, I removed it in r56920.
Hamish
follow-up: 13 comment:12 by , 11 years ago
Replying to annakrat:
Replying to hamish:
not sure how to fix that; the same works in devbr6.
...
If I comment out the
self.InputDataColumn.SetSelection(0)line the v.krige GUI starts in trunk, so it isn't too badly broken.right, I removed it in r56920.
I guess that's one way of fixing it :) but there are a couple more like it. what do they do?
wxpython/scripts/vkrige.py", line 212, in OnInputDataChanged
self.InputDataColumn.SetSelection(0)
AttributeError: 'ColumnSelect' object has no attribute 'SetSelection'
and after commenting that out ...
wxpython/scripts/vkrige.py", line 213, in OnInputDataChanged
self.RunButton.Enable(self.InputDataColumn.GetSelection() is not -1)
AttributeError: 'ColumnSelect' object has no attribute 'GetSelection'
but then the [Plot/refresh variogram] button never lights up, nor the [Run] button, and the suggested output map names don't get filled in.
(v.krige now updated in devbr6 to reflect recent wxgui dir layout changes & starts up ok there)
still a couple tracebacks there re. 'FlatNotebook' object has no attribute 'SetSelectionByName' and if globals()["InputData"] is None: ... KeyError: 'InputData'.
see yesterday's comments in #1196.
thanks, Hamish
follow-up: 14 comment:13 by , 11 years ago
Replying to hamish:
but then the [Plot/refresh variogram] button never lights up, nor the [Run] button, and the suggested output map names don't get filled in.
another fix r56924
(v.krige now updated in devbr6 to reflect recent wxgui dir layout changes & starts up ok there)
still a couple tracebacks there re.
'FlatNotebook' object has no attribute 'SetSelectionByName'andif globals()["InputData"] is None: ... KeyError: 'InputData'. see yesterday's comments in #1196.thanks, Hamish
comment:14 by , 11 years ago
Replying to annakrat:
another fix r56924
great, that fixes the two greyed out buttons and the output map name suggestions are back.
when I press [Plot/refresh variogram] I get the same KeyError: 'InputData' traceback as mentioned for devbr6, I tried adding it as "global InputData" the the outer scripts/v.krige.py function which calls the controller, but it still wasn't being exported to the list of globals read by wxpython/scripts/vkrige.py. (maybe I did it wrong though)
when I press [Run] I got another traceback, but fixed that in r56925. Now it runs in trunk from the GUI with reasonable looking results! (default params, if I start messing with the fitting I run into other problems, and some of the progress output is a bit garbled, but one step at a time.)
I notice it didn't pick up 'cat' as a numeric column since the OnInputMapChanged() function in wx/vkrige.py has excludeKey = True. I'm not sure how often the key column will be also used as a data column, but if nothing else it's useful for testing, so for now I've re-enabled it in r56926.
(* in testing spearfish's arch points using the cat column as the input I notice one stray bad cell value (-2.27374e-13) in the results)
thanks, Hamish
comment:15 by , 11 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
Hello all from GRASS Community Sprint in Prague!
v.krige is now fixed in trunk. I moved the global variables into Controller class, and store them as attributes.
Testing is welcome!
I close this ticket and welcome you to open new ones with a proper title - this one is getting a bit too crowded.
Thanks! Anne

Hello momsen,
thanks for the well detailed report.
v.krige is not aware of last GUI reorganisations, so it can't load properly, as stated by the traceback:
So far I didn't find the time to update the module. If you are willing to help, let me know.
Moreover, the module should not require zoo and xts packages. I can't see them as dependencies of the modules needed by v.krige, that are:
Can you load these packages in a R session?
Thanks for the feedback, Anne