3

I've got a distance matrix with entities with an Input ID, target ID, and distance. I'm looking for a way to find out the relative direction of the input ID columnn to the target ID column.

Question has been answered and works, but this question has alternative solutions for anyone who is looking for a similar process.

4
  • The distance matrix was created with one point layer as input or two?
    – Bera
    Commented Sep 11, 2020 at 16:57
  • two point layers were used to make it, but I've got the matrix plotted as a point layer on QGIS (as well as the original two layers)
    – James
    Commented Sep 11, 2020 at 17:05
  • Is the distance matrix generated from an initial layer? And these "target ids", do they come from a different layer?
    – wfgeo
    Commented Sep 11, 2020 at 17:09
  • My distance matrix is generated from two initial layers, Layer 1 has the Input IDs and Layer 2 has the Target IDs.
    – James
    Commented Sep 11, 2020 at 17:22

1 Answer 1

2

This became a little bit convoluted but it seems to be working.

For each pair of points azimuth is calculated then translated into directions (N, E ,S or W)

Change layer and field names to match your data:

dm = QgsProject.instance().mapLayersByName('Distance matrix')[0]
pointlyr1 = QgsProject.instance().mapLayersByName('ns_point')[0] #Input point layer in distance matrix
pointlyr1_idfield = 'ogc_fid'
pointlyr2 = QgsProject.instance().mapLayersByName('js_point')[0] #Target 
pointlyr2_idfield = 'ogc_fid'

new_field_name = 'Direction' #Add this text field to distance matrix before executing the code

dms = [[int(f['InputID']),int(f['TargetID'])] for f in dm.getFeatures()] #List all input and target ids
p1dict = {int(f[pointlyr1_idfield]):f.geometry() for f in pointlyr1.getFeatures()} #Create a dictionary of id and geometry for input
p2dict = {int(f[pointlyr2_idfield]):f.geometry() for f in pointlyr2.getFeatures()} #Same for target

directions = [] #List to hold directions (N,E,S,W)

for id1, id2 in dms:
    #For each pair of points calculate azimuth (-180 - +180)
    az = p1dict[id1].asPoint().azimuth(p2dict[id2].asPoint())
    def calcdir(a):
        #Translate azimuth to compass direction
        if -45<a<45:
            v = 'N'
        elif 45<=a<135:
            v = 'E'
        elif 135<=a<=180 or -180<=a<=-135:
            v = 'S'
        else:
            v = 'W'
        return v
    directions.append(calcdir(az))
#Write directions list to a column in Distance matrix layer
with edit(dm):
    for feat, direct in zip(dm.getFeatures(), directions):
        #print(feat.id(), direct)
        feat.setAttribute(feat.fieldNameIndex(new_field_name), direct) 
        dm.updateFeature(feat)

enter image description here

0

Not the answer you're looking for? Browse other questions tagged or ask your own question.