[postgis-users] simplifying (homogenize) a polygon

[Nicolas Ribot]

You welcome.

In fact, thanks to you: I used to extract multipolygons boundaries
with such queries (!!):

st_exteriorRing((st_dumpRings(
            (st_dump(geom)).geom
        )).geom) as geom

until your use case makes me remember st_boundary function :)

Nicolas

On 21 November 2013 15:51, Denis Rouzaud <denis.rouzaud at gmail.com> wrote:
> wow, thanks a lot!
>
> indeed, much much nicer!
>
> thanks again,
>
> Denis
>
>
>
> On 21. 11. 13 15:47, Nicolas Ribot wrote:
>>
>> Thanks for the explanation.
>>
>>
>> You could correct the multipolygons in one step, by using the
>> st_buildArea function, that "merges" a linework of linestrings,
>> preserving holes.
>> To extract polygons boundaries as linestring, st_boundary() is very handy:
>>
>> The "bad_multipg" table contains invalid multipolygons with adjacent
>> parts (http://imgur.com/MP6Cy4f)
>>
>> select gid, st_buildArea(st_boundary(geom))
>> from bad_multipg
>> group by gid;
>>
>> Gives: http://imgur.com/RqlowFI
>>
>> Nicolas
>>
>> On 21 November 2013 13:50, Denis Rouzaud <denis.rouzaud at gmail.com> wrote:
>>>
>>> Hi Nicolas,
>>>
>>> I edit my multi polygons in QGIS, and you're right the geometry becomes
>>> invalid if some parts have adjacent segments.
>>>
>>> In a simple case, I would have a multi polygons with 3 parts with 2 parts
>>> adjacent.
>>> http://imgur.com/zvxPFeR
>>>
>>> Although the geometry is invalid, my intention is to make it valid, and
>>> therefore merge the two parts, which is impossible in QGIS - AFAIK.
>>>
>>> So my intention was to save the geometry as this, and make it valid by a
>>> command.
>>>
>>> I could achieve this with a function:
>>> CREATE OR REPLACE FUNCTION distribution.simplemultipoly(geom geometry)
>>> RETURNS geometry AS
>>> $BODY$
>>>      DECLARE
>>>          num_geom integer;
>>>      BEGIN
>>>          SELECT ST_NumGeometries(geom) into num_geom;
>>>          SELECT
>>> ST_Multi(ST_CollectionHomogenize(ST_Union(ST_GeometryN(geom,n))))
>>>              FROM generate_series(1,num_geom) n
>>>              INTO geom;
>>>          RETURN geom;
>>>      END;
>>> $BODY$
>>> LANGUAGE plpgsql;
>>>
>>> But I wondered if there was a more direct way.
>>>
>>> Cheers,
>>>
>>> Denis
>>>
>>>
>>> On 21. 11. 13 12:42, Nicolas Ribot wrote:
>>>
>>> Hi Denis,
>>>
>>> In your example, I don't understand how the parts of your
>>> multipolygons can be adjacent. Are they valid ?
>>> Do you have a visual example ?
>>>
>>> Thank you
>>>
>>> Nicolas
>>>
>>>
>>> On 21 November 2013 07:32, Denis Rouzaud <denis.rouzaud at gmail.com> wrote:
>>>
>>> Hello Rémi,
>>>
>>> I was hoping a simplest request without enabling topology but thanks
>>> anyway!
>>>
>>> Cheers,
>>>
>>> Denis
>>>
>>>
>>> On 20. 11. 13 18:26, Rémi Cura wrote:
>>>
>>>
>>>  From what I understand of your needs, Postigs topology was designed for
>>> this.
>>> Cheers,
>>>
>>> Rémi-C
>>>
>>>
>>> 2013/11/20 Denis Rouzaud <denis.rouzaud at gmail.com>
>>>
>>> Hi all,
>>>
>>> I am drawing some multipolygons in QGIS and sometimes, I have parts of
>>> them which are adjacent and I'd like to homogenize them to have less
>>> parts
>>> and no adjacent parts.
>>>
>>> I could do this with a quite complex method:
>>>
>>> 1. get the number of parts Z:
>>> SELECT ST_NumGeometries(geometry) FROMmytable WHERE id=123;
>>>
>>> 2. do the simplification using ST_CollectionHomogenize:
>>> SELECT
>>> ST_Multi(ST_CollectionHomogenize(ST_Union(ST_GeometryN(geometry,n))))
>>> FROM
>>> mytable CROSS JOIN generate_series(1,Z) n WHERE id = 123;
>>>
>>> 3. update manually the result in the table (copy paste).
>>>
>>>
>>> Does someone has something more straightforward???
>>>
>>> Thanks a lot!
>>>
>>> Best regards,
>>>
>>> Denis
>>> _______________________________________________
>>> postgis-users mailing list
>>> postgis-users at lists.osgeo.org
>>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>>>
>>>
>>>
>>> _______________________________________________
>>> postgis-users mailing list
>>> postgis-users at lists.osgeo.org
>>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>>>
>>>
>>>
>>> _______________________________________________
>>> postgis-users mailing list
>>> postgis-users at lists.osgeo.org
>>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>>>
>>> _______________________________________________
>>> postgis-users mailing list
>>> postgis-users at lists.osgeo.org
>>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>>>
>>>
>>>
>>> _______________________________________________
>>> postgis-users mailing list
>>> postgis-users at lists.osgeo.org
>>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>>
>> _______________________________________________
>> postgis-users mailing list
>> postgis-users at lists.osgeo.org
>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users