[postgis-devel] Re: [postgis-users] bounding box incorrect on circularstring

[Paul Ramsey]

Yes, ignore the_prep regression failures. I'm almost done moving to a
memcmp strategy, and I'll update regression then.

P.

On Tue, Oct 7, 2008 at 5:28 PM, Mark Leslie <mrk.leslie at gmail.com> wrote:
> Mark Cave-Ayland wrote:
>> Ken Southerland wrote:
>>> I am using Postgres 8.2.4 on Windows and am having an issue with the
>>> Circular Strings in the Postgis extension.
>>>
>>> Why is the bounding box of circularstrings incorrectly calculated.
>>> Here I have a circular string that represents a nearly closed circle...
>>>
>>> db=# SELECT astext( geom ) FROM geom_table WHERE id = 0;
>>>
>>> ---------------------------------------
>>>  CIRCULARSTRING(220268.439465645 150415.359530563,220227.333322076
>>> 150505.561285
>>> 879,220227.353105332 150406.434743975)
>>>
>>>
>>> ... but it's extent is a rectangle that only covers the two endpoints
>>> and ignores the middle point and the arc segments in-between...
>>>
>>> db=# SELECT astext( extent( geom ) ) FROM geom_table WHERE id = 0;
>>>
>>> ---------------------------------------------------------------
>>>  POLYGON((220227.34375 150406.421875,220227.34375
>>> 150415.375,220268.453125 15041
>>> 5.375,220268.453125 150406.421875,220227.34375 150406.421875))
>>>
>>> ... is this a known issue that is being addressed?
>>>
>>> Ken
>>
>> Hi Ken,
>>
>> Which PostGIS version are you using? Any chance you could file this on
>> the PostGIS bugtracker?
>>
>>
>> Many thanks,
>>
>> Mark.
>>
> I've created an issue for this
> <http://code.google.com/p/postgis/issues/detail?id=58>, and have a
> working patch for the 1.3 branch.  Unfortunately, I'm getting regression
> failures in regress_ogc_prep (both before and after the patch) that are
> preventing me from committing.  The errors I'm getting are...
>
>  ERROR:  function st_intersects(unknown, text, integer) does not exist
>  ERROR:  function st_intersects(geometry, geometry, integer) does not exist
>  ERROR:  function st_intersects(unknown, unknown, integer) does not exist
>
> ... for various combinations of geometry, unknown and text.  Are these a
> result of our ongoing prepared geometry API indecision and consequently
> ignorable?
>
> --
> Mark Leslie
> Geospatial Software Architect
> LISAsoft
>
> -------------------------------------------------------------
> Ph: +61 2 8570 5000 Fax: +61 2 8570 5099 Mob: +61
> Suite 112, Jones Bay Wharf 19-21 Pirrama Rd Pyrmont NSW 2009
> -------------------------------------------------------------
>
> LISAsoft is part of the A2end Group of Companies
> http://www.ardec.com.au
> http://www.lisasoft.com
> http://www.terrapages.com
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>