[postgis-devel] Contributing code to PostGIS Raster

[Pierre Racine]

>> Sounds good to me.  Should I be providing patches once all the
>> functions of a FV is complete or whenever I'm complete with a
>> function?
>>
>
>Good to me too. I'm with #837 and #851. Then #650, #762 and GDAL
>driver. The last task oriented to help gvSIG/QGIS people with their
>plugins, and provide writing support (keeping an eye in GSoC too).

Frank did not have time to complete the support for float or double in GDALPolygonize during the code sprint. I don't know if he will continue to work on this in the coming weeks. We'll see if he manifest himself. So you might want to put #650 on ice onless you are willing to look at it yourself.

Another useful thing which is planned and sister to ST_Resample is ST_SameAlignment which is a preriquisite for PL/pgSQL functions (and the future two rasters version of ST_MapAlgebra) to determine if they need to ST_Resample first.

Otherwise I guess Objective Objective 2.0.01, 2.0.03 and 2.0.05 will not make it in PostGIS 2.0. (We must keep some work for PostGIS 2.1 after all :-)

David was suppose to work on 2.0.01 during the code sprint but we find having a mapalgebra working with a function much more useful.

The PL/pgSQL prototype for the two raster version of ST_MapAlgebra (Objective 2.0.03) has still to be finished (plpgsql/st_mapalgebra_optimized.sql).

2.0.05 would be very nice to have for PL/pgSQL developpers but it might not do it to PostGIS 2.0.

>Bborie, I personally prefer a patch per FV. Pierre?

Bborie, please coordinate with Jorge for patches.

Pierre