[GRASS-dev] vector display: attributes for areas

[Markus Metz]

On Tue, Aug 9, 2011 at 10:17 AM, Hamish <hamish_b at yahoo.com> wrote:
> Markus Metz wrote:
>> is there a particular reason why vector area attributes can
>> only be displayed if centroids are also displayed? Often I have
>> to set symbol size to zero to get a reasonably clean display,
>> centroids would only clutter the display and most of the time I
>> am not interested in the exact location of centroids when
>> displaying areas.
>>
>> Any objections if I add this functionality to d.vect:
>> display area attributes also if centroids are not displayed?
>
>
> do you mean the centroid "+" (or other symbol) must be displayed
> to also get the display=attribute for an area? if so that sounds
> line a recent regression.

d.vect behaves likes this since 6.0.2, maybe earlier.
>
> if you leave it as display=shape,attribute of course you will
> get the shape (ie symbol) as well.

What I mean is that
d.vect display=shape,attr type=area
does not display area attributes whereas
d.vect display=shape,attr type=area,centroid
does, but I usually don't want to display centroids when I want to
display area attributes.
>
>
> as you must well know, but for the sake of the discussion, areas
> are virtual features fundamentally tied to centroids.

Hm, for the sake of the discussion, areas are not fundamentally, but
optionally tied to centroids. Centroids in turn are fundamentally tied
to areas. IOW, areas without centroids are perfectly legal whereas
centroids outside any area or duplicate centroids violate topology.

Markus M