[QGIS-Developer] closestSegmentWithContext() with epsilon in Python

[Radim Blazek]

You are right, it was simple, https://github.com/qgis/QGIS/pull/5755

I have added also leftOf as Out as it can be also useful.

Radim

On Fri, Nov 24, 2017 at 3:38 PM, Radim Blazek <radim.blazek at gmail.com> wrote:
> It looks like this:
>
> #ifndef SIP_RUN
>     double closestSegmentWithContext( const QgsPointXY &point,
> QgsPointXY &minDistPoint, int &afterVertex, double *leftOf = nullptr,
> double epsilon = DEFAULT_SEGMENT_EPSILON ) const;
> #else
>     double closestSegmentWithContext( const QgsPointXY &point,
> QgsPointXY &minDistPoint SIP_OUT, int &afterVertex SIP_OUT ) const;
> #endif
>
> maybe trivial for who knows how to do it, that is why I asked. How to
> skip leftOf and make epsilon keyword argument?
>
> Radim
>
> On Fri, Nov 24, 2017 at 1:16 PM, Matthias Kuhn <matthias at opengis.ch> wrote:
>> Hi Radim,
>>
>> I don't know why the epsilon is excluded from the python bindings, the
>> docs say nothing.
>> Adjusting it should be trivial, just edit qgsgeometry.h (the
>> prepare-commit script will do the heavy lifting)
>>
>> Matthias
>>
>> On 11/24/2017 12:58 PM, Radim Blazek wrote:
>>> Hi,
>>>
>>> I found that closestSegmentWithContext() in Python is useless for
>>> geometries in geographic crs, because small squared distances are set
>>> to zero because of epsilon threshold. See for example:
>>> https://user-images.githubusercontent.com/29453443/32953189-c395d8a0-cbaf-11e7-8d71-b5869392f08b.PNG
>>> points are not assigned to lines correctly because square distances are wrong.
>>>
>>> Would it be possible to add epsilon as keyword argument to
>>> closestSegmentWithContext() and possibly other methods taking epsilon
>>> in C++?
>>>
>>> I am not sip/python expert, no idea how to do that.
>>>
>>> Thanks
>>> Radim
>>> _______________________________________________
>>> QGIS-Developer mailing list
>>> QGIS-Developer at lists.osgeo.org
>>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>>
>> _______________________________________________
>> QGIS-Developer mailing list
>> QGIS-Developer at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer