[GRASS-user] v.rast.stats for just a single statistic, in a python loop

GRASS GRASS user
1

Hello,

I would like to use the v.rast.stats inside a loop in a python script.
The thing is I just need to get one of the nine columns it generates (sum).
As it runs many times in the loop, it takes a lot of (unnecessary, for this script) time and also creates a huge table with too many (unnecessary, for this script) columns.
I was wondering then if there was a way to specify the calculation for just that single statistic.

Also, I wanted to ask you if there is a way to access the source code of a specific module (in this case, v.rast.stats) via web (without downloading the whole source code), just to have a quick look at it (and try to adapt it to my needs, eventually contribute to it if possible, … I don’t know, … just wondering what’s the right way to do it …).
This might be a stupid question, but please take into account that i’m still quite new in all this (fascinating) stuff.

Thank you very much in advance.
Regards

–
Ismael G.
Ingeniero de Caminos (UPM). Ingénieur des Ponts et Chaussées (ENPC).
PhD researcher @ Urban & Land Planning Department @ ETSICCP UPM www.caminos.upm.es +34 91 336 67 83

2

On 18/04/12 20:54, Ismael Gómez wrote:

Hello,

I would like to use the v.rast.stats inside a loop in a python script.
The thing is I just need to get one of the nine columns it generates (sum).
As it runs many times in the loop, it takes a lot of (unnecessary, for
this script) time and also creates a huge table with too many
(unnecessary, for this script) columns.
I was wondering then if there was a way to specify the calculation for
just that single statistic.

v.rast.stats is a script, so fairly easy to modify (see below for accessing it).

Also, I wanted to ask you if there is a way to access the source code of
a specific module (in this case, v.rast.stats) via web (without
downloading the whole source code), just to have a quick look at it (and
try to adapt it to my needs, eventually contribute to it if possible,

http://trac.osgeo.org/grass/browser/grass

For the development version (aka grass7) go to trunk, for other version to branches.

For v.rast.stats in the 6.4 release branch:

http://trac.osgeo.org/grass/browser/grass/branches/releasebranch_6_4/scripts/v.rast.stats/v.rast.stats.

See "Download in other formats" at the bottom for downloading it.

A quick and dirty guess would be that modifying

line 247: BASECOLS="n min max range mean stddev variance cf_var sum"

and

line 368 sed -e '1d' "$STATSTMP" | awk -F "|" '{printf "\nUPDATE '${TABLE}' SET '${col1}' = %i , '${col2}' = %.2f , '${col3}' = %.2f , '${col4}' = %.2f , '${col5}' = %.2f , '${col6}' = %.2f , '${col7}' = %.2f , '${col8}' = %.2f , '${col9}' = %.2f WHERE '${KEYCOL}' = %i;", $2,$3,$4,$5,$6,$7,$8,$9,$10,$1}' > "$SQLTMP"

could actually be enough.

Moritz

3

Thanks a million, Moritz.
Yours,
Ismael

2012/4/18 Moritz Lennert <mlennert@club.worldonline.be>

On 18/04/12 20:54, Ismael Gómez wrote:

Hello,

I would like to use the v.rast.stats inside a loop in a python script.
The thing is I just need to get one of the nine columns it generates (sum).
As it runs many times in the loop, it takes a lot of (unnecessary, for
this script) time and also creates a huge table with too many
(unnecessary, for this script) columns.
I was wondering then if there was a way to specify the calculation for
just that single statistic.

v.rast.stats is a script, so fairly easy to modify (see below for accessing it).

Also, I wanted to ask you if there is a way to access the source code of
a specific module (in this case, v.rast.stats) via web (without
downloading the whole source code), just to have a quick look at it (and
try to adapt it to my needs, eventually contribute to it if possible,

http://trac.osgeo.org/grass/browser/grass

For the development version (aka grass7) go to trunk, for other version to branches.

For v.rast.stats in the 6.4 release branch:

http://trac.osgeo.org/grass/browser/grass/branches/releasebranch_6_4/scripts/v.rast.stats/v.rast.stats.

See “Download in other formats” at the bottom for downloading it.

A quick and dirty guess would be that modifying

line 247: BASECOLS=“n min max range mean stddev variance cf_var sum”

and

line 368 sed -e ‘1d’ “$STATSTMP” | awk -F “|” ‘{printf “\nUPDATE ‘${TABLE}’ SET ‘${col1}’ = %i , ‘${col2}’ = %.2f , ‘${col3}’ = %.2f , ‘${col4}’ = %.2f , ‘${col5}’ = %.2f , ‘${col6}’ = %.2f , ‘${col7}’ = %.2f , ‘${col8}’ = %.2f , ‘${col9}’ = %.2f WHERE ‘${KEYCOL}’ = %i;”, $2,$3,$4,$5,$6,$7,$8,$9,$10,$1}’ > “$SQLTMP”

could actually be enough.

Moritz

–

Ismael G.
Ingeniero de Caminos (UPM). Ingénieur des Ponts et Chaussées (ENPC).
PhD researcher @ Urban & Land Planning Department @ ETSICCP UPM www.caminos.upm.es +34 91 336 67 83