[GRASS-user] GRASS7: problems with WMS

Hermann_Peifer · September 18, 2014, 6:41am

Hi All,

I am trying to load a WMS using GRASS7, which I compiled from svn/release branch.

Connecting to the WMServer, selecting and adding a layer works fine, but I do not get any image into the map display.

The GUI's Map Console reports:
Unable to determine region, m.proj failed

The console, from which I started the GUI says:
ERROR 4: `/path/to/mapset/.tmp/hostname/3199.12' does not exist in the file system, and is not recognised as a supported dataset name.

After noting the error related to region/m.proj, I tried both: to load a WMS in EPSG:3035 into my LAEA location and to load another WMS in EPSG:4326 into my WGS84 location, whithout any luck.

Any hints?

Thanks in advance, Hermann

Hermann_Peifer · September 18, 2014, 5:24pm

Just to add that I continue trying to load the WMS, without success, though. My current error message is:

ERROR 4: `/home/peifer/grassdata/laea/test/.tmp/whitefish/17189.4' does not exist in the file system, and is not recognised as a supported dataset name.

When looking into .tmp/whitefish, I see a series of 17189.* files, but obviously not 17189.4:

[GRASS:test]> ls -l .tmp/whitefish/
total 1184
-rw-r--r-- 1 peifer mtr 1144239 Sep 18 19:18 17189.0.ppm
-rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.10
-rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.11
-rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.5
-rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.6
-rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.7
-rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.8
-rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.9

On 2014-09-18 8:41, Hermann Peifer wrote:

Hi All,

I am trying to load a WMS using GRASS7, which I compiled from
svn/release branch.

Connecting to the WMServer, selecting and adding a layer works fine, but
I do not get any image into the map display.

The GUI's Map Console reports:
Unable to determine region, m.proj failed

The console, from which I started the GUI says:
ERROR 4: `/path/to/mapset/.tmp/hostname/3199.12' does not exist in the
file system, and is not recognised as a supported dataset name.

After noting the error related to region/m.proj, I tried both: to load a
WMS in EPSG:3035 into my LAEA location and to load another WMS in
EPSG:4326 into my WGS84 location, whithout any luck.

Any hints?

Thanks in advance, Hermann

stepan.turek · September 18, 2014, 8:31pm

Hi Hermann,

would be possible to send me the WMS service URL you are trying to connect?

Best
Stepan

---------- Původní zpráva ----------
Od: Hermann Peifer peifer@gmx.eu
Komu: GRASS user list grass-user@lists.osgeo.org
Datum: 18. 9. 2014 19:24:44
Předmět: Re: [GRASS-user] GRASS7: problems with WMS

Just to add that I continue trying to load the WMS, without success,
though. My current error message is:

ERROR 4: `/home/peifer/grassdata/laea/test/.tmp/whitefish/17189.4’ does
not exist in the file system, and is not recognised as a supported
dataset name.

When looking into .tmp/whitefish, I see a series of 17189.* files, but
obviously not 17189.4:

[GRASS:test]> ls -l .tmp/whitefish/
total 1184
-rw-r–r-- 1 peifer mtr 1144239 Sep 18 19:18 17189.0.ppm
-rw-r–r-- 1 peifer mtr 9125 Sep 18 19:10 17189.10
-rw-r–r-- 1 peifer mtr 8903 Sep 18 19:10 17189.11
-rw-r–r-- 1 peifer mtr 574 Sep 18 19:10 17189.5
-rw-r–r-- 1 peifer mtr 9125 Sep 18 19:10 17189.6
-rw-r–r-- 1 peifer mtr 8903 Sep 18 19:10 17189.7
-rw-r–r-- 1 peifer mtr 574 Sep 18 19:10 17189.8
-rw-r–r-- 1 peifer mtr 574 Sep 18 19:10 17189.9

On 2014-09-18 8:41, Hermann Peifer wrote:

Hi All,

I am trying to load a WMS using GRASS7, which I compiled from
svn/release branch.

Connecting to the WMServer, selecting and adding a layer works fine, but
I do not get any image into the map display.

The GUI’s Map Console reports:
Unable to determine region, m.proj failed

The console, from which I started the GUI says:
ERROR 4: `/path/to/mapset/.tmp/hostname/3199.12’ does not exist in the
file system, and is not recognised as a supported dataset name.

After noting the error related to region/m.proj, I tried both: to load a
WMS in EPSG:3035 into my LAEA location and to load another WMS in
EPSG:4326 into my WGS84 location, whithout any luck.

Any hints?

Thanks in advance, Hermann

grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

=

Hermann_Peifer · September 19, 2014, 5:12am

As far as I tried out: the error occurs for any WMS in any projection, including all the built-in servers. It looks like no other GRASS7 user ever had the same issue, which is surprising. From the error message, I assume that soemthing is going systematically wrong with the numbering of the files in the .tmp directory.

I am currently trying with layer: "WMS Global Mosaic, pan sharpened" from the built-in server list: http://onearth.jpl.nasa.gov/wms.cgi

The result is:

ERROR 4: `/home/peifer/grassdata/wgs84/PERMANENT/.tmp/whitefish/32754.4' does not exist in the file system, and is not recognised as a supported dataset name.

File 32754.4 does indeed not exist. See below.

Hermann

[GRASS:PERMANENT]> file .tmp/whitefish/*
.tmp/whitefish/32754.0.ppm: Netpbm PPM "rawbits" image data
.tmp/whitefish/32754.10: XML document text
.tmp/whitefish/32754.11: XML document text
.tmp/whitefish/32754.5: XML document text
.tmp/whitefish/32754.6: XML document text
.tmp/whitefish/32754.7: XML document text
.tmp/whitefish/32754.8: HTML document text
.tmp/whitefish/32754.9: XML document text

On 2014-09-18 22:31, Štěpán Turek wrote:

Hi Hermann,

would be possible to send me the WMS service URL you are trying to connect?

Best
Stepan

---------- Původní zpráva ----------
Od: Hermann Peifer <peifer@gmx.eu>
Komu: GRASS user list <grass-user@lists.osgeo.org>
Datum: 18. 9. 2014 19:24:44
Předmět: Re: [GRASS-user] GRASS7: problems with WMS

    Just to add that I continue trying to load the WMS, without success,
    though. My current error message is:

    ERROR 4: `/home/peifer/grassdata/laea/test/.tmp/whitefish/17189.4' does
    not exist in the file system, and is not recognised as a supported
    dataset name.

    When looking into .tmp/whitefish, I see a series of 17189.* files, but
    obviously not 17189.4:

    [GRASS:test]> ls -l .tmp/whitefish/
    total 1184
    -rw-r--r-- 1 peifer mtr 1144239 Sep 18 19:18 17189.0.ppm
    -rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.10
    -rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.11
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.5
    -rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.6
    -rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.7
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.8
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.9

    On 2014-09-18 8:41, Hermann Peifer wrote:
     > Hi All,
     >
     > I am trying to load a WMS using GRASS7, which I compiled from
     > svn/release branch.
     >
     > Connecting to the WMServer, selecting and adding a layer works
    fine, but
     > I do not get any image into the map display.
     >
     > The GUI's Map Console reports:
     > Unable to determine region, m.proj failed
     >
     > The console, from which I started the GUI says:
     > ERROR 4: `/path/to/mapset/.tmp/hostname/3199.12' does not exist
    in the
     > file system, and is not recognised as a supported dataset name.
     >
     > After noting the error related to region/m.proj, I tried both: to
    load a
     > WMS in EPSG:3035 into my LAEA location and to load another WMS in
     > EPSG:4326 into my WGS84 location, whithout any luck.
     >
     > Any hints?
     >
     > Thanks in advance, Hermann

    _______________________________________________
    grass-user mailing list
    grass-user@lists.osgeo.org
    http://lists.osgeo.org/mailman/listinfo/grass-user

Hermann_Peifer · September 19, 2014, 11:49am

Maybe this helps to locate the problem. If not: I simply have to give up
loading a WMS with GRASS7. Hermann

[GRASS6.4.5svn]> r.in.wms map=http://wms.cuzk.cz/wms.asp out=wms layers=kn
srs=EPSG:3035 format=png
Calculating tiles
Requesting 30 tiles.
Downloading tiles
...

[GRASS7.0.0svn]> r.in.wms url=http://wms.cuzk.cz/wms.asp out=wms layers=kn
srs=3035 format=png

ERROR: Sorry, <input> is not a valid parameter
ERROR: Unable to determine region, m.proj failed

--
View this message in context: http://osgeo-org.1560.x6.nabble.com/GRASS7-problems-with-WMS-tp5162533p5162883.html
Sent from the Grass - Users mailing list archive at Nabble.com.

annakrat · September 19, 2014, 1:33pm

On Fri, Sep 19, 2014 at 1:12 AM, Hermann Peifer <peifer@gmx.eu> wrote:

As far as I tried out: the error occurs for any WMS in any projection,
including all the built-in servers. It looks like no other GRASS7 user ever
had the same issue, which is surprising. From the error message, I assume
that soemthing is going systematically wrong with the numbering of the
files in the .tmp directory.

I am currently trying with layer: "WMS Global Mosaic, pan sharpened" from
the built-in server list: http://onearth.jpl.nasa.gov/wms.cgi

The result is:

ERROR 4: `/home/peifer/grassdata/wgs84/PERMANENT/.tmp/whitefish/32754.4'
does not exist in the file system, and is not recognised as a supported
dataset name.

I get the same, however in the gui console I also get:
WMS server error: <?xml version='1.0' encoding="UTF-8"
standalone="no" ?>
<!DOCTYPE ServiceExceptionReport SYSTEM
"http://wms.jpl.nasa.gov/exception_1_1_1.dtd ">
<ServiceExceptionReport version="1.1.1">
<ServiceException>
This server no longer provides full WMS services!
</ServiceException>
</ServiceExceptionReport>

so this service doesn't work any more.

Anna

File 32754.4 does indeed not exist. See below.

Hermann

[GRASS:PERMANENT]> file .tmp/whitefish/*
.tmp/whitefish/32754.0.ppm: Netpbm PPM "rawbits" image data
.tmp/whitefish/32754.10: XML document text
.tmp/whitefish/32754.11: XML document text
.tmp/whitefish/32754.5: XML document text
.tmp/whitefish/32754.6: XML document text
.tmp/whitefish/32754.7: XML document text
.tmp/whitefish/32754.8: HTML document text
.tmp/whitefish/32754.9: XML document text

On 2014-09-18 22:31, Štěpán Turek wrote:

Hi Hermann,

would be possible to send me the WMS service URL you are trying to
connect?

Best
Stepan

---------- Původní zpráva ----------
Od: Hermann Peifer <peifer@gmx.eu>
Komu: GRASS user list <grass-user@lists.osgeo.org>
Datum: 18. 9. 2014 19:24:44
Předmět: Re: [GRASS-user] GRASS7: problems with WMS

    Just to add that I continue trying to load the WMS, without success,
    though. My current error message is:

    ERROR 4: `/home/peifer/grassdata/laea/test/.tmp/whitefish/17189.4'
does
    not exist in the file system, and is not recognised as a supported
    dataset name.

    When looking into .tmp/whitefish, I see a series of 17189.* files, but
    obviously not 17189.4:

    [GRASS:test]> ls -l .tmp/whitefish/
    total 1184
    -rw-r--r-- 1 peifer mtr 1144239 Sep 18 19:18 17189.0.ppm
    -rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.10
    -rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.11
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.5
    -rw-r--r-- 1 peifer mtr 9125 Sep 18 19:10 17189.6
    -rw-r--r-- 1 peifer mtr 8903 Sep 18 19:10 17189.7
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.8
    -rw-r--r-- 1 peifer mtr 574 Sep 18 19:10 17189.9

    On 2014-09-18 8:41, Hermann Peifer wrote:
     > Hi All,
     >
     > I am trying to load a WMS using GRASS7, which I compiled from
     > svn/release branch.
     >
     > Connecting to the WMServer, selecting and adding a layer works
    fine, but
     > I do not get any image into the map display.
     >
     > The GUI's Map Console reports:
     > Unable to determine region, m.proj failed
     >
     > The console, from which I started the GUI says:
     > ERROR 4: `/path/to/mapset/.tmp/hostname/3199.12' does not exist
    in the
     > file system, and is not recognised as a supported dataset name.
     >
     > After noting the error related to region/m.proj, I tried both: to
    load a
     > WMS in EPSG:3035 into my LAEA location and to load another WMS in
     > EPSG:4326 into my WGS84 location, whithout any luck.
     >
     > Any hints?
     >
     > Thanks in advance, Hermann

    _______________________________________________
    grass-user mailing list
    grass-user@lists.osgeo.org
    http://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

annakrat · September 19, 2014, 2:02pm

On Fri, Sep 19, 2014 at 7:49 AM, peifer <peifer@gmx.eu> wrote:

Maybe this helps to locate the problem. If not: I simply have to give up
loading a WMS with GRASS7. Hermann

[GRASS6.4.5svn]> r.in.wms map=http://wms.cuzk.cz/wms.asp out=wms layers=kn
srs=EPSG:3035 format=png
Calculating tiles
Requesting 30 tiles.
Downloading tiles
...

[GRASS7.0.0svn]> r.in.wms url=http://wms.cuzk.cz/wms.asp out=wms layers=kn
srs=3035 format=png

ERROR: Sorry, <input> is not a valid parameter
ERROR: Unable to determine region, m.proj failed

I have no idea, the <input> probably refers to m.proj but m.proj has input
parameter. Could you try to run only m.proj (take some example from
manual), if it works?

Anna

--
View this message in context:
http://osgeo-org.1560.x6.nabble.com/GRASS7-problems-with-WMS-tp5162533p5162883.html
Sent from the Grass - Users mailing list archive at Nabble.com.
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Hermann_Peifer · September 19, 2014, 5:18pm

On 2014-09-19 16:02, Anna Petrášová wrote:

I have no idea, the <input> probably refers to m.proj but m.proj has
input parameter. Could you try to run only m.proj (take some example
from manual), if it works?

Either I am doing something terribly stupid, or m.proj is *really* broken in the release branch. I wonder what m.proj or cs2cs is expected to do with colors: [color=style] and [color=name]. Furthermore: no coordinate input seems to be needed, see [1]. This parameter list looks like a joke, to say it in a friendly way.

I "fixed" all of my earlier reported issues related to m.proj and loading a WMS by moving from the GRASS7 release branch to svn/trunk. In trunk, m.proj has a reasonable parameter list [2] and loading the WMS works as expected.

Hermann

[1] Welcome to GRASS 7.0.0svn (r61824)
...

[GRASS:test]> m.proj --h

Description:
Converts coordinates from one projection to another (cs2cs frontend).

Keywords:
miscellaneous, projection

Usage:
  m.proj [-iodec] [color=style] [output=name] [separator=character]
    [color=name] [proj_in=string] [proj_out=string] [--overwrite] [--help]
    [--verbose] [--quiet]

[2] Welcome to GRASS 7.1.svn (r62036)
...

[GRASS:PERMANENT]> m.proj --help

Description:
Converts coordinates from one projection to another (cs2cs frontend).

Keywords:
miscellaneous, projection

Usage:
  m.proj [-iodec] [coordinates=east,north] [input=name] [output=name]
    [separator=character] [proj_in=string] [proj_out=string] [--overwrite]
    [--help] [--verbose] [--quiet]

annakrat · September 19, 2014, 5:29pm

On Fri, Sep 19, 2014 at 1:18 PM, Hermann Peifer <peifer@gmx.eu> wrote:

On 2014-09-19 16:02, Anna Petrášová wrote:

I have no idea, the <input> probably refers to m.proj but m.proj has
input parameter. Could you try to run only m.proj (take some example
from manual), if it works?

Either I am doing something terribly stupid, or m.proj is *really* broken
in the release branch. I wonder what m.proj or cs2cs is expected to do with
colors: [color=style] and [color=name]. Furthermore: no coordinate input
seems to be needed, see [1]. This parameter list looks like a joke, to say
it in a friendly way.

yes, that's pretty funny. Do you have any local changes (svn diff)? I
looked in the recent changes but there is nothing suggesting this error.

I "fixed" all of my earlier reported issues related to m.proj and loading
a WMS by moving from the GRASS7 release branch to svn/trunk. In trunk,
m.proj has a reasonable parameter list [2] and loading the WMS works as
expected.

Hermann

[1] Welcome to GRASS 7.0.0svn (r61824)
...

[GRASS:test]> m.proj --h

Description:
Converts coordinates from one projection to another (cs2cs frontend).

Keywords:
miscellaneous, projection

Usage:
m.proj [-iodec] [color=style] [output=name] [separator=character]
   [color=name] [proj_in=string] [proj_out=string] [--overwrite] [--help]
   [--verbose] [--quiet]

[2] Welcome to GRASS 7.1.svn (r62036)
...

[GRASS:PERMANENT]> m.proj --help

Description:
Converts coordinates from one projection to another (cs2cs frontend).

Keywords:
miscellaneous, projection

Usage:
m.proj [-iodec] [coordinates=east,north] [input=name] [output=name]
   [separator=character] [proj_in=string] [proj_out=string] [--overwrite]
   [--help] [--verbose] [--quiet]

Hermann_Peifer · September 19, 2014, 5:44pm

On 2014-09-19 19:29, Anna Petrášová wrote:

yes, that's pretty funny. Do you have any local changes (svn diff)? I
looked in the recent changes but there is nothing suggesting this error.

No diff at all. And until yesterday, I did not even know that m.proj existed.

Hermann

Hermann_Peifer · September 19, 2014, 5:59pm

On 2014-09-19 19:44, Hermann Peifer wrote:

On 2014-09-19 19:29, Anna Petrášová wrote:

yes, that's pretty funny. Do you have any local changes (svn diff)? I
looked in the recent changes but there is nothing suggesting this error.

No diff at all. And until yesterday, I did not even know that m.proj
existed.

This might help the initiated to find some bug:

diff grass/scripts/m.proj/m.proj.py grass70_release/scripts/m.proj/m.proj.py

38c38
< #% description: '-' for standard input
---
> #% description: '-' to read from standard input

96d95
< from grass.script.utils import separator, parse_key_val

162,163c161,162
< ifs = separator(ifs)
< ofs = separator(ofs)
---
> ifs = grass.separator(ifs)
> ofs = grass.separator(ofs)

167c166
< kv = parse_key_val(s)
---
> kv = grass.parse_key_val(s)

wenzeslaus · September 19, 2014, 9:35pm

On Fri, Sep 19, 2014 at 1:59 PM, Hermann Peifer <peifer@gmx.eu> wrote:

On 2014-09-19 19:44, Hermann Peifer wrote:

On 2014-09-19 19:29, Anna Petrášová wrote:

yes, that's pretty funny. Do you have any local changes (svn diff)? I
looked in the recent changes but there is nothing suggesting this error.

No diff at all. And until yesterday, I did not even know that m.proj
existed.

This might help the initiated to find some bug:

diff grass/scripts/m.proj/m.proj.py grass70_release/scripts/m.proj/
m.proj.py

38c38
< #% description: '-' for standard input
---
> #% description: '-' to read from standard input

96d95
< from grass.script.utils import separator, parse_key_val

162,163c161,162
< ifs = separator(ifs)
< ofs = separator(ofs)
---
> ifs = grass.separator(ifs)
> ofs = grass.separator(ofs)

167c166
< kv = parse_key_val(s)
---
> kv = grass.parse_key_val(s)

This cannot cause the problem, so let's try something else.

When in GRASS command line do

which m.proj

Then open the file which it will tell and also make a diff of that file
with grass70_release/scripts/m.proj/m.proj.py

Afterwards, you can also try to recompile GRASS (make distclean &&
./configure ... && make).

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

neteler · September 20, 2014, 4:03am

On Fri, Sep 19, 2014 at 1:49 PM, peifer <peifer@gmx.eu> wrote:

Maybe this helps to locate the problem. If not: I simply have to give up
loading a WMS with GRASS7. Hermann

[GRASS6.4.5svn]> r.in.wms map=http://wms.cuzk.cz/wms.asp out=wms layers=kn
srs=EPSG:3035 format=png
Calculating tiles
Requesting 30 tiles.
Downloading tiles
...

Fine.

[GRASS7.0.0svn]> r.in.wms url=http://wms.cuzk.cz/wms.asp out=wms layers=kn srs=3035 format=png
ERROR: Sorry, <input> is not a valid parameter
ERROR: Unable to determine region, m.proj failed

Which version (g.version -g) do you use?

I have made a test:

GRASS 7.0.0svn (latlong):~ > g.region n=90 s=-90 w=-180 e=180
res=0:30:00 -pprojection: 3 (Latitude-Longitude)
zone: 0
datum: wgs84
ellipsoid: wgs84
north: 90N
south: 90S
west: 180W
east: 180E
nsres: 0:30
ewres: 0:30
rows: 360
cols: 720
cells: 259200

GRASS 7.0.0svn (latlong):~ > r.in.wms
url=http://watzmann-geog.urz.uni-heidelberg.de/cached/osm
layers=osm_auto:all output=osm format=png --o
Downloading data from WMS server...
Importing raster map into GRASS...
WARNING: Raster map <osm> already exists and will be overwritten
<osm> created.
WARNING: Raster map <osm.alpha@neteler> is a base map for <MASK@neteler>.
Remove forced.

GRASS 7.0.0svn (latlong):~ > g.version -g
version=7.0.0svn
date=2014
revision=62008
build_date=2014-09-17
build_platform=x86_64-unknown-linux-gnu

Works ok with the OSM server.

Markus

Hermann_Peifer · September 20, 2014, 5:28am

On 2014-09-19 23:35, Vaclav Petras wrote:

Afterwards, you can also try to recompile GRASS (make distclean &&
./configure ... && make).

I actually did that, without any further investigations. After make distclean, I also used: svn up, which brought me from r61824 to r62033.

When comparing the output of make to what I got 12 days ago, I see a difference related to grass70_release/dist.x86_64-unknown-linux-gnu/etc/python/grass/script/utils.py. Maybe this has to do with my problems, maybe not.

In any case: all reported issues are gone now and loading a WMS just works as expected. Thanks to all for their time.

Hermann