[postgis-users] error running St_MapAlgebraExpr 2 raster version

[dustymugs]

The only problem you have in your query is the expression being provided
to ST_MapAlgebraExpr().  That expression should be:

[rast1.val]*[rast2.val]

The notice itself can safely be ignored.  In the normal use case, your
source rasters are tiled into smaller chucks.  And between these smaller
chucks, some of these tiles will not intersect.

+---+---+---+
|   |   |   |
+---+---+---+
|   |   |   |
+---*---*---*---+
|   |   |   |   |
+---*---*---*---+
    |   |   |   |
    +---+---+---+
    |   |   |   |
    +---+---+---+

If you were to look at the text figure above, the two rasters intersect.
 But some of the smaller chucks when compared would not.

-bborie

On 02/19/2013 03:30 PM, Silver Ashen wrote:
> Hello everybody,
> 
> I'm trying to test the St_MapAlgebraExpr 2 raster version in order to
> multiply the values of a raster by the values of an other one.
> 
> Here is the request :
> 
> SELECT St_MapAlgebraExpr(a.rast,b.rast,'(a.rast*b.rast)::double
> precision','32BF','INTERSECTION','NULL','NULL',-1) As rast
> FROM sigma.mnt_5m_2941x6197 as a, sigma.affor_aligned2 as b WHERE
> St_Intersects(a.rast,b.rast);
> 
> Note that affor_aligned2 was a vector rasterized with St_AsRaster in the
> way that it's fit the alignment of mnt_5m_2941x6197.
> 
> However, the function above returns the following :
> 
> "Notice : The two rasters provided have no intersection.  Returning no band
> raster"
> 
> I don't understand because when I look to the two raster in ArcGIS, there
> is no problem of srid. The two layers are well overlayed...
> 
>  I don't understand what I did wrong. Any idea ?
> 
> Thanks,
> Samy
> 
> 
> 
> _______________________________________________
> postgis-users mailing list
> postgis-users at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>