[GRASS5] how to NULL in site lists?

GRASS GRASS developer
1

Hi all,

as working with sites at time, I face the problem that I
don't knwo how to code null values in a site_list.
(Eric, we already talk once about this weeks ago).

As we cannot simply invent a special character for
this (maybe to complex to change the entire sites lib),
what about using for GRASS 5.0 the value -0.0?

Just an idea. Or is there another way?
Using -9999 like ESRI does is of course unwanted and
currently leads to nonsense results (as it is treated
as true value).

As -0.0 is currently also treated as true value, I
would like to know if it could become a workaround unless
it is improved in 5.1?

Markus

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

2

On Tue, Jan 23, 2001 at 09:59:47PM +0000, Markus Neteler wrote:

Hi all,

as working with sites at time, I face the problem that I
don't knwo how to code null values in a site_list.
(Eric, we already talk once about this weeks ago).

As we cannot simply invent a special character for
this (maybe to complex to change the entire sites lib),
what about using for GRASS 5.0 the value -0.0?

Just an idea. Or is there another way?
Using -9999 like ESRI does is of course unwanted and
currently leads to nonsense results (as it is treated
as true value).

As -0.0 is currently also treated as true value, I
would like to know if it could become a workaround unless
it is improved in 5.1?

We've discussed this before. I don't know that modules will treat -0.0
appropriately (most likely they'll just treat it like zero). IMHO, the
whole conception of the sites format sucks from both a user's and a
programmer's perspective. When we get proper tabular data support built
into GRASS, this problem should be easy to handle. But that's going to
be a while, and I for one have yet to put my "money" where my mouth is
:wink: Hopefully, in 5.1 we can solve this. I don't think using -0.0 will
do much at this point, since most sites modules have some fundamental
assumptions about the data they are reading (which includes it not being
NULL and not being NaN). Also, this doesn't address NULL string values.

Sorry if I'm kind of a spoiler on the quick fix, but I think fundamental
changes are required.

--
Eric G. Miller <egm2@jps.net>

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

3

On Tue, 23 Jan 2001, Eric G . Miller wrote:

On Tue, Jan 23, 2001 at 09:59:47PM +0000, Markus Neteler wrote:
> Hi all,
>
> as working with sites at time, I face the problem that I
> don't knwo how to code null values in a site_list.
> (Eric, we already talk once about this weeks ago).
>
> As we cannot simply invent a special character for
> this (maybe to complex to change the entire sites lib),
> what about using for GRASS 5.0 the value -0.0?
>
> Just an idea. Or is there another way?
> Using -9999 like ESRI does is of course unwanted and
> currently leads to nonsense results (as it is treated
> as true value).
>
> As -0.0 is currently also treated as true value, I
> would like to know if it could become a workaround unless
> it is improved in 5.1?

We've discussed this before. I don't know that modules will treat -0.0
appropriately (most likely they'll just treat it like zero). IMHO, the
whole conception of the sites format sucks from both a user's and a
programmer's perspective. When we get proper tabular data support built
into GRASS, this problem should be easy to handle. But that's going to
be a while, and I for one have yet to put my "money" where my mouth is
:wink: Hopefully, in 5.1 we can solve this. I don't think using -0.0 will
do much at this point, since most sites modules have some fundamental
assumptions about the data they are reading (which includes it not being
NULL and not being NaN). Also, this doesn't address NULL string values.

Sorry if I'm kind of a spoiler on the quick fix, but I think fundamental
changes are required.

I agree with Eric: the quick fix probably won't be "quick" for all the
programs using sites files. The quickest fix does though - multiple sites
files with single attributes and no NULL or NA values, i.e. subsetting the
sites. A more general solution will, I think, be reached with sites/vector
attribute tables, say by 5.1.

Roger

--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand@nhh.no
and: Department of Geography and Regional Development, University of
Gdansk, al. Mar. J. Pilsudskiego 46, PL-81 378 Gdynia, Poland.

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

4

Hi Markus, Hi all,

i agree with Eric and Roger that the quick fix is not advisable.

But there is another more general consideration to the NULLs in site
lists.
IMHO the site list format is for the storage of point data derived from
measurements. Usually the points are irregularily spaced. The site list
contains the spatial data (coordinates) and the attribute(s) in a sort
of 1:1 database. So if you have no value (measurement) at a given point,
there should be no site list entry/line.
The problem comes from changing the sites format from one attribute to
several attributes. This can not be modelled adequate with a 1:1
database structure.
Another approach would be to create a site list for each
attribute/variable and to ommit the sites where no measurement/variable
value is available (NULL attributes). I don't know if NULL values with
site files are conceptually correct.

Sorry, i am not able to explain this in better english.

cu,

Andreas

Markus Neteler wrote:

Hi all,

as working with sites at time, I face the problem that I
don't knwo how to code null values in a site_list.
(Eric, we already talk once about this weeks ago).

As we cannot simply invent a special character for
this (maybe to complex to change the entire sites lib),
what about using for GRASS 5.0 the value -0.0?

Just an idea. Or is there another way?
Using -9999 like ESRI does is of course unwanted and
currently leads to nonsense results (as it is treated
as true value).

As -0.0 is currently also treated as true value, I
would like to know if it could become a workaround unless
it is improved in 5.1?

Markus

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

--
Andreas Lange, 65187 Wiesbaden, Germany, Tel. +49 611 807850
Andreas.Lange@Rhein-Main.de - A.C.Lange@GMX.net

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'