
Geometry : 
Subject: [ggl] Polygon DP simplification
From: V D (zedxz2)
Date: 20110930 14:36:13
*Typo:
boost::geometry(p1,p2, my_dist_strat);
Should read
boost::geometry::distance(p1,p2, my_dist_strat);
Sorry about that
v
On Sep 30, 2011, at 10:56 AM, V D wrote:
> Hi,
>
> I'm trying to supply my own strategy to work around the problem for now.
> Using the same code as before, except now I'm trying:
>
>
> 27 typedef boost::geometry::strategy::distance::pythagoras<point_xy, point_xy, double> dist_strat;
> 28 typedef boost::geometry::strategy::simplify::douglas_peucker<point_xy, dist_strat > simp_strat;
>
> 30 simp_strat my_simplify_strat_test;
> 31 boost::geometry::simplify(original_1, simplified_1, 1.0 / 2048.0, my_simplify_strat_test);
>
>
> But I cannot get the strategy to compile, I get a pretty nasty compile error :
>
>
> instantiated from ?void boost::geometry::simplify(const Geometry&, Geometry&, const Distance&, const Strategy&) [with Geometry = main()::multi_polygon, Distance = double, Strategy = main()::simp_strat]?
> simp_test.cpp:31: instantiated from here
> /opt/local/include/boost/geometry/strategies/concepts/distance_concept.hpp:232: error: no type named ?type? in ?struct boost::geometry::strategy::distance::services::strategy_point_point<boost::geometry::strategy::distance::comparable::pythagoras<boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, double> >?
> /opt/local/include/boost/geometry/strategies/concepts/distance_concept.hpp:233: error: no type named ?type? in ?struct boost::geometry::strategy::distance::services::strategy_point_point<boost::geometry::strategy::distance::comparable::pythagoras<boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, double> >?
> /opt/local/include/boost/geometry/strategies/concepts/distance_concept.hpp:244: error: no matching function for call to ?boost::geometry::strategy::distance::comparable::pythagoras<boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, double>::apply(const boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>&, const boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>&, const boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>&)?
> /opt/local/include/boost/geometry/strategies/cartesian/distance_pythagoras.hpp:92: note: candidates are: static typename boost::geometry::select_calculation_type<S1, S2, CalculationType>::type boost::geometry::strategy::distance::comparable::pythagoras<Point1, Point2, CalculationType>::apply(const Point1&, const Point2&) [with Point1 = boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, Point2 = boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, CalculationType = double]
>
>
> The dist_strat type works fine, for example:
>
> point_xy p1(0.0, 0.0)
> point_xy p2(5.0, 0.0)
>
> dist_strat my_dist_strat;
> boost::geometry(p1,p2, my_dist_strat);
>
> But I can't seem to supply it to my deuglas peucker strategy ?
>
> v
>
>
>
>
> On Sep 30, 2011, at 6:04 AM, Mateusz Loskot wrote:
>
>> On 30/09/11 03:47, V D wrote:
>>> Hum, if the problem is with the pythagoras distance, why is the
>>> following snippet returning the expected value, 5.0.
>>> Given the current error in the source, shouldn't it return 25 ?
>>>
>>> #include <boost/geometry.hpp>
>>> #include <boost/geometry/strategies/cartesian/distance_pythagoras.hpp>
>>> #include <boost/geometry/geometries/point_xy.hpp>
>>> #include <iostream>
>>>
>>> int main()
>>> {
>>> typedef boost::geometry::model::d2::point_xy<double> point_xy;
>>>
>>> point_xy p1(0.0, 0.0);
>>> point_xy p2(5.0, 0.0);
>>>
>>> boost::geometry::strategy::distance::pythagoras<point_xy, point_xy,
>>> double> my_strategy;
>>> std::cout << boost::geometry::distance(p1,p2, my_strategy) << std::endl;
>>>
>>> return 0;
>>> }
>>>
>>>
>>> Is it only in the use of DP simplification that the pythagoras strategy
>>> is problematic ?
>>
>> The difference is in the actual strategies that kick in.
>> I explained it in my last comment:
>>
>> https://svn.boost.org/trac/boost/ticket/5954#comment:3
>>
>> So, yes, my patch does not fix the problem, but sheds some light where
>> is the bug. Though, I'm having problems with finding out why the
>> different strategy is generated for simplify, not having enough time to
>> dig it now.
>>
>> p.s. Please, take care to respond to the mailing list too. I added ggl address to my response here.
>>
>> Best regards,
>> 
>> Mateusz Loskot, http://mateusz.loskot.net
>> Charter Member of OSGeo, http://osgeo.org
>> Member of ACCU, http://accu.org
>
Geometry list run by mateusz at loskot.net