non-square shapefiles and tileindex

[Frank Warmerdam]

Ara.T.Howard wrote:
> that the roi is, in fact, covered on that western edge.
> 
> i'm at a loss as to how that command could 'partially' work.  is that the
> result you got?  or was the entire longitudal range covered?

Ara,

Yes, I got a similar result.

> also very strange, is that requests for individual orbits sometime work and
> sometimes do not

The problem is that mapserver does not understand that in geographic coordinate
systems the dateline is a "wrap point".  So for instance for some of your
images the footprint polygon goes well past 180 degrees east, but MapServer
does not know that this overlaps the area just east of the west dateline.

One not-so-pleasant workaround for this is to create duplicate versions of
datasets that cross the dateline, with one dataset using negative longitudes
and one using positive longitudes.  This can be accomplished using GDAL
VRT files if you want to avoid duplicating the actual data.

Generally speaking "world level" mapping in cartesian software packages
designed for projected coordinate systems or more local mapping is fraught
with peril.

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    | President OSGeo, http://osgeo.org