[GRASS-dev] problem initializing temporal gis

Sören Gebbert soerengebbert at googlemail.com
Sat Jun 28 03:32:56 PDT 2014


Hi Luca,

2014-06-27 15:08 GMT+02:00 Luca Delucchi <lucadeluge at gmail.com>:
> On 24 June 2014 17:10, Sören Gebbert <soerengebbert at googlemail.com> wrote:
>> Hi Luca,
>
> Hi Soeren,
>
>> there is no dedicated field for unit specification. You can use the
>> description, or maybe better, the semantic field in the base
>> information object:
>>
>> {{{
>>>>> import grass.temporal as tgis
>>>>> tgis.init()
>>>>> stds = tgis.SpaceTimeRasterDataset("a at PERMANENT")
>>>>> stds.base.set_semantic_type("unit:degree celsius")
>>>>> stds.base.get_semantic_type()
>> 'unit:degree celsius'
>> }}}
>>
>
> ok thanks,
> do you think could be possible to add a field for unit in future?

Yes, it is possible to add any amount of fields in the future. Older
databases must then be updated to support the new fields, otherwise
the updated framework will not work with them.

However, a specific unit field that can be set by the user will
conflict with the unit field of raster and voxel maps. IMHO the unit
of a space time raster dataset should be derived from the units of the
registered raster maps. Well i did not implemented this, since from my
experience are units used very seldom for maps. So the correct
implementation would add a units fields to raster and 3d raster map
tables and to STRDS and STR3DS tables. The STDS SQL code must be
updated to aggregate the unit information. But what to do if map layer
have different units? Or if only a subset of the registered maps have
units?

So maybe the semantic type option is sufficient right now?

Best regards
Soeren

>
>>
>> Best regards
>> Soeren
>>
>
> --
> ciao
> Luca
>
> http://gis.cri.fmach.it/delucchi/
> www.lucadelu.org


More information about the grass-dev mailing list