[gdal-dev] GDAL with OGDI

[Malm Paul]

I looked at table soundp in postgres, created with FWTOOLS which works, and all hdh, numeric(15,6), values are 1.000000.
Could it be that the all values are nan in the original data, and were previously changed in OGDI to 1.0000?

By the way, can I ommit the hdh column with ogr2ogr when importing the DNC chart to Postgres?

Chaitanya, how do I register to create tickets, and
do you still want me to do that?

/Paul 

-----Original Message-----
From: fwarmerdam at gmail.com [mailto:fwarmerdam at gmail.com] On Behalf Of Frank Warmerdam
Sent: den 30 juni 2010 08:34
To: Chaitanya kumar CH
Cc: Malm Paul; gdal-dev at lists.osgeo.org
Subject: Re: [gdal-dev] GDAL with OGDI

On Wed, Jun 30, 2010 at 8:00 AM, Chaitanya kumar CH <chaitanya.ch at gmail.com> wrote:
> Paul,
>
>
> I think that the problem is with the SQL command built by the 
> PostgreSQL driver.
> In the INSERT command a numeric field with the value NaN should be 
> mentioned in quotes.

Chaitanya,

It might be prudent for the postgres driver to check for nan's, but it isn't usual for OGR drivers to anticipate nan values in numeric fields.

>> It is the same with ogdi-3.1.5 and ogdi-3.2.0-beta.
>> why is hdh = nan (the corresponding hdp is correct), it works when 
>> importing with ogr2ogr from FWTOOLS.

Regardless of what might be done to deal with the nan value, I think there is an underlying issue to investigate, which is why a nan is being produced by OGDI for a field that apparently has a valid value.

Best regards,
-- 
---------------------------------------+--------------------------------
---------------------------------------+------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent