Hi,
I'm using Informix 9.30 with geodetic datablade (3.00). It seems
that the geodetic spatial functions, e.g., distance, doesn't
take the 3rd element of a point into account. The following
is the script I tried with the result.
create table test (location GeoObject);
insert into test values ('GeoPoint((3,4,10),ANY,ANY)');
insert into test values ('GeoPoint((3,5,1),ANY,ANY)');
CREATE INDEX test_idx
ON test (location geospcalt_ops)
USING RTREE ;
select location, distance(location, '((3,4,0),ANY, ANY)'::GeoPoint) from
test;
## RESULT
location GeoPoint((3,4,10),ANY,ANY)
(expression) 0.00
location GeoPoint((3,5,1),ANY,ANY)
(expression) 111167.9466418
If the 3rd element is considered, the distance between (3,4,0) and (3,5,1)
should be closer than that between (3,4,0) and (3,4,10), but it seems that
this is not the case for geodetic distance function. I'm wondering if this
is true that the geodetic spatial functions doesn't consider the 3rd
element in the point or if there is something that I'm missing here.
Any comments would be appreciated.
Thanks in advance,
Kiyoung..
Ken Tanaka - 28 Oct 2003 19:04 GMT
Kiyoung,
It looks like the altitude is not taken into account, but that distance
is measured only along the datum ellipsoid (geodesic). Look at page 1-12
(34) of the pdf document "Informix Geodetic DataBlade Module User's
Guide" at
http://www-3.ibm.com/software/data/informix/pubs/library/interim/8675-pdf.html
-Ken
> Hi,
>
[quoted text clipped - 33 lines]
> Thanks in advance,
> Kiyoung..
kiyoungy - 29 Oct 2003 00:09 GMT
Ken,
Thanks for the clarification. I'm wondering, then, if
there is any other function that take the altitude
into account when computing the distance. Do we have
to implement this ourselves?
Any comments would be appreciated.
Thanks in advance,
Kiyoung..
> Kiyoung,
>
[quoted text clipped - 44 lines]
> > Thanks in advance,
> > Kiyoung..
Ken Tanaka - 29 Oct 2003 18:43 GMT
Kiyoung,
I think you will have to implement it yourselves. You can let Informix
Geodetic provide the ground distance and then calculate in the altitude
difference. If a flat-earth approach is accurate enough, then this isn't
too hard, it's harder if you need to take into account the curvature of
the Earth.
-Ken
> Ken,
>
[quoted text clipped - 7 lines]
> Thanks in advance,
> Kiyoung..