[GRASS-user] v.isochrones : error message

[Moritz Lennert]

On 15/04/15 15:00, image93 wrote:
> Hello,
>
> Thank you Moritz. I succeed in getting your last v.isochrones version. I
> have just one last question. According to you, it would be possible to use
> v.isochrones extension via Qgis interface?


You should be able to easily integrate it into the processing menu by 
creating a file v.isochrones.txt that contains the information about the 
parameters.

Here's an (untested) example which would call v.isochrones with the 
v.net.iso method.
v.isochrones
v.isochrones - Creates isochrones from a road map and starting points
Vector (v.*)
ParameterVector|map|Roads with speed attribute|1|False
ParameterVector|start_points|Vector map with starting points for 
isochrones|0|False
ParameterTableField|cost_column|Name of cost column (in 
minutes)|input|-1|False
ParameterString|time_steps|Time steps of isochrones (in minutes)|None
OutputVector|isochrones|Output vector layer

In my Debian installation of QGIS, this file then goes to

/usr/share/qgis/python/plugins/processing/algs/grass7/description

However, IIUC, this would currently not work from QGIS as QGIS layers 
are always either point, line or polygon, so having a road network with 
a nodes layer is not possible. We would have to modify v.isochrones to 
just add the starting points to the network at each run. I wanted to 
avoid that because of the time it takes and the fact that in many cases, 
people run many analyses on the same network, so currently v.isochrones 
in its v.net.iso method just looks for the node on the network that is 
closest to the starting point and goes from there.

The r.cost method does not have this limit, so would be easier to 
integrate into QGIS in the current state of the module.

Moritz