t.rast.what: issues in one_point_per_col_output()

[sbl]

There is a ; hardcoded in the header for one_point_per_col_output() in t.rast.what, which causes columns to shift. In addition, there is an indentation error.

[sbl]

In 72635:

fix header and indentation for one_point_per_col_output; see #3546

[sbl]

Please test the (tiny) fix in trunk before backporting.

[sbl]

In 72636:

use different separator for coordinates and sites; see #3546

[sbl]

Sorry, that first fix was not working properly. r72636 is more complete and works as expected here.

However, additional documentation might be needed!? Users should be probably informed that default separator for coordinates (and site) is comma, if comma is not the main column separator. In the latter case semicolon is used as spearator for coordinates and site.

[sbl]

Can be tested with example from the manual:

# Create test data
g.region s=0 n=80 w=0 e=120 b=0 t=50 res=10

# Generate data
r.mapcalc expression="a_1 = 1" -s
r.mapcalc expression="a_2 = 2" -s
r.mapcalc expression="a_3 = 3" -s
r.mapcalc expression="a_4 = 4" -s

t.create type=strds output=A title="A test" descr="A test"

t.register -i type=raster input=A maps=a_1,a_2,a_3,a_4 \
    start='1990-01-01' increment="1 month"

v.random output=points n=3 seed=1

#Query STRDS (please try this with and without r72636
t.rast.what strds=A points=points layout=col -nv

Pay attention to the changes in the separators for the coordinates and cat in the header (and the different number of resulting columns)...

Another option could be to remove coordinates from output, when cat is used (v-flag), cause coordinates would be redundant.

[veroandreo]

I tested the fix following the example in trunk (with the fix) and in release branch. Personally, I would prefer to get only cat number when using -v flag. Default could still be the coords (-n) and -v only cat values.

Moreover, testing different layouts, I found that when using layout=timerow, there's an extra empty column or line according if I use -vn or only -v (but this also happens in release branch, without the fix):

t.rast.what strds=A points=points layout=timerow -vn
cat|x|y|1990-01-01 00:00:00;1990-02-01 00:00:00|1990-02-01 00:00:00;1990-03-01 00:00:00|1990-03-01 00:00:00;1990-04-01 00:00:00|1990-04-01 00:00:00;1990-05-01 00:00:00
1|115.004358627375|36.3593955782903||1|2|3|4
2|79.681676382576|45.2391522852909||1|2|3|4
3|97.4892579600048|79.2347263950131||1|2|3|4

t.rast.what strds=A points=points layout=timerow -v

1|115.004358627375|36.3593955782903||1|2|3|4
2|79.681676382576|45.2391522852909||1|2|3|4
3|97.4892579600048|79.2347263950131||1|2|3|4

[sbl]

Fix empty column (and header) in timerow outout

[sbl]

Thanks for testing! If there is agreement on removing coordinates if cat is used, I can change that. Not sure if that can go to 7.4, though???

Please also test attached new diff that fixes the empty column in the timerow output and an empty line if header is not requested...

Looks now like this:

t.rast.what strds=A points=points layout=timerow -v
1|115.004358627375|36.3593955782903|1|2|3|4
2|79.681676382576|45.2391522852909|1|2|3|4
3|97.4892579600048|79.2347263950131|1|2|3|4

t.rast.what strds=A points=points layout=timerow -vn
cat|x|y|1990-01-01 00:00:00;1990-02-01 00:00:00|1990-02-01 00:00:00;1990-03-01 00:00:00|1990-03-01 00:00:00;1990-04-01 00:00:00|1990-04-01 00:00:00;1990-05-01 00:00:00
1|115.004358627375|36.3593955782903|1|2|3|4
2|79.681676382576|45.2391522852909|1|2|3|4
3|97.4892579600048|79.2347263950131|1|2|3|4

t.rast.what strds=A points=points layout=timerow
115.004358627375|36.3593955782903|1|2|3|4
79.681676382576|45.2391522852909|1|2|3|4
97.4892579600048|79.2347263950131|1|2|3|4

echo "115 36 Loc1                                  
79 45 Loc2" | t.rast.what -i strds=A layout=timerow -n separator=','
x,y,site,1990-01-01 00:00:00;1990-02-01 00:00:00,1990-02-01 00:00:00;1990-03-01 00:00:00,1990-03-01 00:00:00;1990-04-01 00:00:00,1990-04-01 00:00:00;1990-05-01 00:00:00
115,36,Loc1,1,2,3,4
79,45,Loc2,1,2,3,4

[veroandreo]

Replying to sbl:

Thanks for testing! If there is agreement on removing coordinates if cat is used, I can change that. Not sure if that can go to 7.4, though???

IMHO, it's a minor change in the behavior of a flag, but I don't think it can go into 7.4. I guess into 7.6, no?

Please also test attached new diff that fixes the empty column in the timerow output and an empty line if header is not requested...

Just tested the patch, looks exactly as you showed, great :)

[sbl]

In 72713:

remove empty header and column in timerow output; see #3546

[sbl]

OK, then I keep removing coordinates if cat is used separate from this ticket. If there are no objections, I would like to backport changes above, in order to get properly aligned output in 7.4.

[sbl]

BTW, 4 unit tests are failing in current release branch:

FAIL: test_col_output (__main__.TestRasterWhat)
----------------------------------------------------------------------
(...)
AssertionError: File <out_col.txt> does not have the right MD5 sum.
Expected is <8720cc237b46ddb18f11440469d0e13f>, actual is <885b6f50405b08fa9fe9ae33ed50e29b>

======================================================================
FAIL: test_col_output_cat (__main__.TestRasterWhat)
----------------------------------------------------------------------
(...)
AssertionError: File <out_col_cat.txt> does not have the right MD5 sum.
Expected is <e1d8e6651b3bff1c35e366e48d634db4>, actual is <ac0a9b14e59920c3f8b5834282a24822>

======================================================================
FAIL: test_col_output_coords (__main__.TestRasterWhat)
----------------------------------------------------------------------
(...)
AssertionError: File <out_col_coords.txt> does not have the right MD5 sum.
Expected is <ac44ebc5aa040d4ce2a5787e95f208ec>, actual is <ecdc79a6880a9e1f163cc92fa384b8a3>

======================================================================
FAIL: test_timerow_output_cat (__main__.TestRasterWhat)
----------------------------------------------------------------------
(...)
AssertionError: File <out_col_trow.txt> does not have the right MD5 sum.
Expected is <55e2ff8ddaeb731a73daca48adf2768d>, actual is <1da3350f488df9c919b4009625956b3b>

[sbl]

In 72716:

Make tests pass again; see #3546

[sbl]

In 72717:

Correct output format (trunk, r72635, r72636, r72713, r72716); fix #3546