Re: [gdal-dev] GNM Usage

2015-03-12 Thread ChiefDan
Thanks for the response. The source data is in EPSG 900913, the network data
according to ogrinfo:
PROJCS["WGS_84_Pseudo_Mercator",
GEOGCS["GCS_WGS_1984",
DATUM["WGS_1984",
SPHEROID["WGS_84",6378137,298.257223563]],
PRIMEM["Greenwich",0],
UNIT["Degree",0.017453292519943295]],
PROJECTION["Mercator_1SP"],
PARAMETER["central_meridian",0],
PARAMETER["false_easting",0],
PARAMETER["false_northing",0],
UNIT["Meter",1],
PARAMETER["latitude_of_origin",0.0]]

Looking at the individual point values, they seem to be EPSG 4326:
OGRFeature(gnm_Frog_Branches_point):1
  name (String) = 78B-S-0
  gfid (Integer) = 68
  POINT (12.305749698885625 45.450093091932679)

For a line with multiple point objects on it, will the GNM processing break
the line up into multiple segments creating the line segments between the
points?
Also, if there are overlapping lines that form a Y intersection, can the GNM
processing recognize that as a potential path?




--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192757.html
Sent from the GDAL - Dev mailing list archive at Nabble.com.
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev


Re: [gdal-dev] GNM Usage

2015-03-12 Thread ChiefDan
Looking at the _gnm_graph table, I think that is the issue. THere are some
connections, but only 8 out of ~300 points.
Is it required that the points layer be on the end of a line? The points
layer I have are on top of the lines, however a single line may have
multiple points along its length.

Dan



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192666.html
Sent from the GDAL - Dev mailing list archive at Nabble.com.
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev


Re: [gdal-dev] GNM Usage

2015-03-12 Thread Mikhail Gusev
The topology building algorithm firstly takes one of the line features,
determines its start and end point coordinates, than forms two areas with
the center in those points and in size with the snap tolerance value and
than searches two point features which are situated in the formed areas.
The connection will be recorded to the _gnm_graph only if the two according
point features were found.
So in your case you should verify the snap tolerance parameter. If your
network has the projection SRS it is set in linear units.

2015-03-11 0:27 GMT+04:00 ChiefDan :

> Looking at the _gnm_graph table, I think that is the issue. THere are some
> connections, but only 8 out of ~300 points.
> Is it required that the points layer be on the end of a line? The points
> layer I have are on top of the lines, however a single line may have
> multiple points along its length.
>
> Dan
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192666.html
> Sent from the GDAL - Dev mailing list archive at Nabble.com.
> ___
> gdal-dev mailing list
> gdal-dev@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Re: [gdal-dev] GNM Usage

2015-03-11 Thread Dmitriy Baryshnikov

Hi,

I also would like to note, that this is the simple and sample algorithm. 
You free to use the GNM API to create more complex connections. For 
example, the GNM allows to connect some polygons (i.e. countries) 
without physical  geometry intersect/connect (logical connection). So, 
you have more capabilities, but (by now) using API of GNM.


Best regards,
Dmitry

11.03.2015 15:01, ChiefDan пишет:

Thanks for the info!



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192776.html
Sent from the GDAL - Dev mailing list archive at Nabble.com.
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev



___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Re: [gdal-dev] GNM Usage

2015-03-11 Thread ChiefDan
Thanks for the info!



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192776.html
Sent from the GDAL - Dev mailing list archive at Nabble.com.
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev


Re: [gdal-dev] GNM Usage

2015-03-11 Thread Mikhail Gusev
No, it will not break the line into segments. All point features that are
geometrically on the line feature except endings will be ignored.
For Y intersection the connections will be created only for 3 lines with
the common end point.

2015-03-11 14:28 GMT+04:00 ChiefDan :

> Thanks for the response. The source data is in EPSG 900913, the network
> data
> according to ogrinfo:
> PROJCS["WGS_84_Pseudo_Mercator",
> GEOGCS["GCS_WGS_1984",
> DATUM["WGS_1984",
> SPHEROID["WGS_84",6378137,298.257223563]],
> PRIMEM["Greenwich",0],
> UNIT["Degree",0.017453292519943295]],
> PROJECTION["Mercator_1SP"],
> PARAMETER["central_meridian",0],
> PARAMETER["false_easting",0],
> PARAMETER["false_northing",0],
> UNIT["Meter",1],
> PARAMETER["latitude_of_origin",0.0]]
>
> Looking at the individual point values, they seem to be EPSG 4326:
> OGRFeature(gnm_Frog_Branches_point):1
>   name (String) = 78B-S-0
>   gfid (Integer) = 68
>   POINT (12.305749698885625 45.450093091932679)
>
> For a line with multiple point objects on it, will the GNM processing break
> the line up into multiple segments creating the line segments between the
> points?
> Also, if there are overlapping lines that form a Y intersection, can the
> GNM
> processing recognize that as a potential path?
>
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192757.html
> Sent from the GDAL - Dev mailing list archive at Nabble.com.
> ___
> gdal-dev mailing list
> gdal-dev@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Re: [gdal-dev] GNM Usage

2015-03-10 Thread Mikhail Gusev
Hello, Dan.
Have you checked whether the _gnm_graph table had been populated
successfully? May it be void? The possible reason of this can be that the
points are "too far from the line ends" so you could check if you have set
an appropriate for your data snap tolerance value.

2015-03-10 2:50 GMT+04:00 ChiefDan :

> One thing when building the network I did not do was use CreateRule to add
> costs to paths as this does not make sense in my use case.
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192408.html
> Sent from the GDAL - Dev mailing list archive at Nabble.com.
> ___
> gdal-dev mailing list
> gdal-dev@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Re: [gdal-dev] GNM Usage

2015-03-09 Thread ChiefDan
One thing when building the network I did not do was use CreateRule to add
costs to paths as this does not make sense in my use case.



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/GNM-Usage-tp5192401p5192408.html
Sent from the GDAL - Dev mailing list archive at Nabble.com.
___
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev