Boost logo

Geometry :

Subject: Re: [geometry] Result change in geometry::centroid between version 1.53 and 1.58
From: Adam Wulkiewicz (adam.wulkiewicz_at_[hidden])
Date: 2015-04-24 11:36:39


Eric.noirfalise wrote:
> Hi Adam,
>
> Done for ticket : https://svn.boost.org/trac/boost/ticket/11236

Thanks!

Actually it's a manifestation of a different bug, in arithmetic
functions, in this particular case in add_point().
See: https://github.com/boostorg/geometry/pull/283

You wrote in a ticket that the compiler reports a warning.
Could you paste it here or there?

Regards,
Adam

P.S. Top-posting is discouraged on this list

>
> Thank you,
> Eric
>
> Selon Adam Wulkiewicz <adam.wulkiewicz_at_[hidden]>:
>
>> Hi Eric,
>>
>> Eric.noirfalise wrote:
>>> Dear All,
>>>
>>> I'm having a result change issue in geometry::centroid between boost
>> version
>>> 1.53 and 1.58 when I use input polygon composed of integer points and an
>> output
>>> double point.
>>>
>>> Here is my code :
>>>
>>> #include <boost/geometry.hpp>
>>> #include <boost/geometry/geometries/point_xy.hpp>
>>> #include <boost/geometry/geometries/polygon.hpp>
>>>
>>>
>>> int main()
>>> {
>>> typedef boost::geometry::model::d2::point_xy<int> PointInt;
>>> typedef boost::geometry::model::d2::point_xy<double> PointDouble;
>>> typedef boost::geometry::model::polygon<PointInt> polygon_type;
>>>
>>> polygon_type poly;
>>> boost::geometry::read_wkt(
>>> "POLYGON((1 1, 2 1, 2 2, 1 2, 1 1))", poly);
>>>
>>> PointDouble p;
>>> typedef
>> boost::geometry::strategy::centroid::bashein_detmer<PointDouble,
>>> PointInt, double> Strategy;
>>> boost::geometry::centroid(poly, p, Strategy());
>>>
>>> std::cout << "centroid: " << boost::geometry::dsv(p) << std::endl;
>>>
>>> return 0;
>>> }
>>>
>>> The result for boost 1.53 (which seems good for me) :
>>> centroid: (1.5, 1.5)
>>>
>>> And the result for boost 1.58 (which seems wrong for me) :
>>> centroid: (1, 1)
>>> I can also see new warnings during compilation with version 1.58.
>>>
>>> Could you confirm me that this is a bug ?
>> Yes, it's a bug.
>> Introduced in this PR https://github.com/boostorg/geometry/pull/155
>> This was released with Boost 1.57.
>>
>> Could you please create a ticket?
>>
>> Regards,
>> Adam
>> _______________________________________________
>> Geometry mailing list
>> Geometry_at_[hidden]
>> http://lists.boost.org/mailman/listinfo.cgi/geometry
>>
>
> _______________________________________________
> Geometry mailing list
> Geometry_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/geometry
>


Geometry list run by mateusz at loskot.net