[GRASS-dev] r.median (non)functionality

GRASS GIS GRASS GIS developer
1

Trying out r.median, I discovered

GRASS 6.3.cvs (spearfish60):~/ > r.median base=fields cover=soils.Kfactor
output=K.by.farm.median --o
100%
ERROR: Reading r.stats output

I compared to r.mode: this fix seems to be necessary:

--- main.c 15 Jul 2007 12:30:37 -0000 2.9
+++ main.c 15 Jul 2007 12:48:43 -0000
@@ -79,7 +79,7 @@
     if (G_read_cats (covermap, cover_mapset, &cover_cats) < 0)
        G_fatal_error (_("Cannot read category labels of raster map <%s>"),
covermap);

- strcpy (command, "r.stats -a \"");
+ strcpy (command, "r.stats -an \"");
     strcat (command, G_fully_qualified_name (basemap, base_mapset));
     strcat (command, ",");
     strcat (command, G_fully_qualified_name (covermap, cover_mapset));

Should I submit?
Markus
--
View this message in context: http://www.nabble.com/r.median-(non)functionality-tf4082074.html#a11602182
Sent from the Grass - Dev mailing list archive at Nabble.com.

2

Markus Neteler wrote:

Trying out r.median, I discovered

GRASS 6.3.cvs (spearfish60):~/ > r.median base=fields cover=soils.Kfactor
output=K.by.farm.median --o
100%
ERROR: Reading r.stats output

I compared to r.mode: this fix seems to be necessary:

--- main.c 15 Jul 2007 12:30:37 -0000 2.9
+++ main.c 15 Jul 2007 12:48:43 -0000
@@ -79,7 +79,7 @@
     if (G_read_cats (covermap, cover_mapset, &cover_cats) < 0)
        G_fatal_error (_("Cannot read category labels of raster map <%s>"), covermap);

- strcpy (command, "r.stats -a \"");
+ strcpy (command, "r.stats -an \"");
     strcat (command, G_fully_qualified_name (basemap, base_mapset));
     strcat (command, ",");
     strcat (command, G_fully_qualified_name (covermap, cover_mapset));

Presumably r.median fails if you have any null cells?

Should I submit?

Yes.

--
Glynn Clements <glynn@gclements.plus.com>

3

Glynn Clements wrote:

Markus Neteler wrote:

Trying out r.median, I discovered

GRASS 6.3.cvs (spearfish60):~/ > r.median base=fields cover=soils.Kfactor
output=K.by.farm.median --o
100%
ERROR: Reading r.stats output

I compared to r.mode: this fix seems to be necessary:
...

Presumably r.median fails if you have any null cells?

No, it continues.

Should I submit?

Yes.

Done.

Markus
--
View this message in context: http://www.nabble.com/r.median-(non)functionality-tf4082074.html#a11603531
Sent from the Grass - Dev mailing list archive at Nabble.com.