[GRASS5] (no subject)

GRASS GIS GRASS GIS developer
1

Dear list,
Markus, Stephan, Jachym and I are developing a small GRASS Testframework.
This framework allows to implement tests for single grass modules, or for
a combination of serveral grass modules. It also allows to check the output
from raster, vector, raster3 and file-export-modules via md5 checksums.

The html result of a simple testrun with some sample testscripts is
available here:

http://www-pool.math.tu-berlin.de/~soeren/grass/GRASS_TestSuite/GRASS.html

The framework is still a protype and we have some problems. So i would like
to ask
for some help in a md5 checksum issue.

We are testing the sample testscripts within the spearfish60 location.
While testing the vector md5 validation, we have the problem that the
md5 checksums of the coor and topo files are different between two
computers.
The two computers are running 32Bit linux, but one of them has a 64Bit AMD
cpu.

So my question:
Depends the file format of the vector files (coor, topo) from the cpu or os?
Or are there environment variables which creates different vector files
on different systems?

Any help or suggestions are welcome,
Best regards
Soeren

btw.:
what are you thinking about the html output?

--
10 GB Mailbox, 100 FreeSMS/Monat http://www.gmx.net/de/go/topmail
+++ GMX - die erste Adresse für Mail, Message, More +++

2

On 1/24/06, "Sören Gebbert" <soerengebbert@gmx.de> wrote:

We are testing the sample testscripts within the spearfish60 location.
While testing the vector md5 validation, we have the problem that the
md5 checksums of the coor and topo files are different between two
computers.
The two computers are running 32Bit linux, but one of them has a 64Bit AMD
cpu.

So my question:
Depends the file format of the vector files (coor, topo) from the cpu or os?
Or are there environment variables which creates different vector files
on different systems?

coor and topo files should only differ if they were written on machines
with different byte order. Can you find how the files differ (od, diff)?

Radim

3

Markus, Stephan, Jachym and I are developing a small GRASS Testframework.
This framework allows to implement tests for single grass modules, or for
a combination of serveral grass modules. It also allows to check the output
from raster, vector, raster3 and file-export-modules via md5 checksums.

..

btw.:
what are you thinking about the html output?

very nice!

Perhaps use the PNG driver to test some of the display modules?
(md5sum on the .png output)
Perhaps not.

Hamish

4

Hello Sören,

On Tue, 24 Jan 2006 16:30:42 +0100 (MET) "Sören Gebbert"
<soerengebbert@gmx.de> wrote:

We are testing the sample testscripts within the spearfish60 location.
While testing the vector md5 validation, we have the problem that the
md5 checksums of the coor and topo files are different between two
computers.
The two computers are running 32Bit linux, but one of them has a
64Bit AMD cpu.

So my question:
Depends the file format of the vector files (coor, topo) from the cpu
or os? Or are there environment variables which creates different
vector files on different systems?

I have tested with your provided files and MD5-summ from your maschine.

The differ from mine. Here are some results, which I cannot clearly
interpret:

My maschine: AMD Sempron(tm) 2600+

The command (taken from spearfish60 in user1-mapset):
g.region res=150 res3=150 t=80 b=0 tbres=10
v.buffer input=bugsites@PERMANENT output=bugsites_buff buffer=200

The md5sum from Soeren calculated at my maschine:
md5sum /home/holl/soerens_files/bugsites_buff/*
3c3b6801aa755149eb22954b925f44da /home/holl/soerens_files/bugsites_buff/cidx
99c750d7ae0868f2cf2cee868ca5a3ae /home/holl/soerens_files/bugsites_buff/coor
dcd038557184668b58c37a7b04492306 /home/holl/soerens_files/bugsites_buff/topo

the md5-sums calculated at soerens maschine:
3c3b6801aa755149eb22954b925f44da cidx
99c750d7ae0868f2cf2cee868ca5a3ae coor
dcd038557184668b58c37a7b04492306 topo

My calculations and my md5-sums:
md5sum /home/holl/grassdata/spearfish60/user1/vector/bugsites_buff/*
7511344bb7b329a264aa04c27075a8af /home/holl/[...]/bugsites_buff/cidx
0f55a9ae0dd50a124b220e47e6d9b995 /home/holl/[...]/bugsites_buff/coor
6dd48f3002358e57b388bf7568f8d27f /home/holl/[...]/bugsites_buff/topo

They all do not match!

Filesize of coor differs!
-rw-r--r-- 1 holl users 2833 2006-01-25 11:07 cidx
-rw-r--r-- 1 holl users 115001 2006-01-25 11:07 coor
-rw-r--r-- 1 holl users 19508 2006-01-25 11:07 topo

Mine:
-rw-r--r-- 1 holl users 2833 2006-01-25 11:51 bugsites_buff/cidx
-rw-r--r-- 1 holl users 114971 2006-01-25 11:51 bugsites_buff/coor
-rw-r--r-- 1 holl users 19508 2006-01-25 11:51 bugsites_buff/topo

File-output: Soerens output
cidx: Hitachi SH big-endian COFF object, not stripped
coor: PDP-11 old overlay
topo: Hitachi SH big-endian COFF object, not stripped

Mine:
file /home/holl/[...]/bugsites_buff/*
cidx: Hitachi SH big-endian COFF object, not stripped
coor: PDP-11 old overlay
topo: Hitachi SH big-endian COFF object, not stripped

Any help or suggestions are welcome,

What else could be the problem?

Thanks

  Stephan

--
GDF Hannover - Solutions for spatial data analysis and remote sensing
Hannover Office - Mengendamm 16d - D-30177 Hannover
Internet: www.gdf-hannover.de - Email: holl@gdf-hannover.de
Phone : ++49-(0)511.39088507 - Fax: ++49-(0)511.39088508