QGIS/python/plugins/processing/algs/qgis/PointDistance.py

# -*- coding: utf-8 -*-

"""
***************************************************************************
    PointDistance.py
    ---------------------
    Date                 : August 2012
    Copyright            : (C) 2012 by Victor Olaya
    Email                : volayaf at gmail dot com
***************************************************************************
*                                                                         *
*   This program is free software; you can redistribute it and/or modify  *
*   it under the terms of the GNU General Public License as published by  *
*   the Free Software Foundation; either version 2 of the License, or     *
*   (at your option) any later version.                                   *
*                                                                         *
***************************************************************************
"""

__author__ = 'Victor Olaya'
__date__ = 'August 2012'
__copyright__ = '(C) 2012, Victor Olaya'

# This will get replaced with a git SHA1 when you do a git archive

__revision__ = '$Format:%H$'

import math
from qgis.core import QgsFeatureRequest, QgsFeature, QgsGeometry, QgsDistanceArea
from processing.core.GeoAlgorithm import GeoAlgorithm
from processing.core.parameters import ParameterNumber
from processing.core.parameters import ParameterVector
from processing.core.parameters import ParameterSelection
from processing.core.parameters import ParameterTableField
from processing.core.outputs import OutputTable
from processing.tools import dataobjects, vector


class PointDistance(GeoAlgorithm):

    INPUT_LAYER = 'INPUT_LAYER'
    INPUT_FIELD = 'INPUT_FIELD'
    TARGET_LAYER = 'TARGET_LAYER'
    TARGET_FIELD = 'TARGET_FIELD'
    MATRIX_TYPE = 'MATRIX_TYPE'
    NEAREST_POINTS = 'NEAREST_POINTS'
    DISTANCE_MATRIX = 'DISTANCE_MATRIX'

    MAT_TYPES = ['Linear (N*k x 3) distance matrix',
                 'Standard (N x T) distance matrix',
                 'Summary distance matrix (mean, std. dev., min, max)']

    def defineCharacteristics(self):
        self.name = 'Distance matrix'
        self.group = 'Vector analysis tools'

        self.addParameter(ParameterVector(self.INPUT_LAYER,
            self.tr('Input point layer'), [ParameterVector.VECTOR_TYPE_POINT]))
        self.addParameter(ParameterTableField(self.INPUT_FIELD,
            self.tr('Input unique ID field'), self.INPUT_LAYER,
            ParameterTableField.DATA_TYPE_ANY))
        self.addParameter(ParameterVector(self.TARGET_LAYER,
            self.tr('Target point layer'), ParameterVector.VECTOR_TYPE_POINT))
        self.addParameter(ParameterTableField(self.TARGET_FIELD,
            self.tr('Target unique ID field'), self.TARGET_LAYER,
            ParameterTableField.DATA_TYPE_ANY))
        self.addParameter(ParameterSelection(self.MATRIX_TYPE,
            self.tr('Output matrix type'), self.MAT_TYPES, 0))
        self.addParameter(ParameterNumber(self.NEAREST_POINTS,
            self.tr('Use only the nearest (k) target points'), 0, 9999, 0))

        self.addOutput(OutputTable(self.DISTANCE_MATRIX, self.tr('Distance matrix')))

    def processAlgorithm(self, progress):
        inLayer = dataobjects.getObjectFromUri(
            self.getParameterValue(self.INPUT_LAYER))
        inField = self.getParameterValue(self.INPUT_FIELD)
        targetLayer = dataobjects.getObjectFromUri(
            self.getParameterValue(self.TARGET_LAYER))
        targetField = self.getParameterValue(self.TARGET_FIELD)
        matType = self.getParameterValue(self.MATRIX_TYPE)
        nPoints = self.getParameterValue(self.NEAREST_POINTS)

        outputFile = self.getOutputFromName(self.DISTANCE_MATRIX)

        if nPoints < 1:
            nPoints = len(vector.features(targetLayer))

        self.writer = outputFile.getTableWriter([])

        if matType == 0:
            # Linear distance matrix
            self.linearMatrix(inLayer, inField, targetLayer, targetField,
                matType, nPoints, progress)
        elif matType == 1:
           # Standard distance matrix
            self.regularMatrix(inLayer, inField, targetLayer, targetField,
                nPoints, progress)
        elif matType == 2:
            # Summary distance matrix
            self.linearMatrix(inLayer, inField, targetLayer, targetField,
                matType, nPoints, progress)

    def linearMatrix(self, inLayer, inField, targetLayer, targetField,
                     matType, nPoints, progress):
        if matType == 0:
            self.writer.addRecord(['InputID', 'TargetID', 'Distance'])
        else:
            self.writer.addRecord(['InputID', 'MEAN', 'STDDEV', 'MIN', 'MAX'])

        index = vector.spatialindex(targetLayer)

        inIdx = inLayer.fieldNameIndex(inField)
        outIdx = targetLayer.fieldNameIndex(targetField)

        outFeat = QgsFeature()
        inGeom = QgsGeometry()
        outGeom = QgsGeometry()
        distArea = QgsDistanceArea()

        features = vector.features(inLayer)
        current = 0
        total = 100.0 / float(len(features))
        for inFeat in features:
            inGeom = inFeat.geometry()
            inID = unicode(inFeat.attributes()[inIdx])
            featList = index.nearestNeighbor(inGeom.asPoint(), nPoints)
            distList = []
            vari = 0.0
            for i in featList:
                request = QgsFeatureRequest().setFilterFid(i)
                outFeat = targetLayer.getFeatures(request).next()
                outID = outFeat.attributes()[outIdx]
                outGeom = outFeat.geometry()
                dist = distArea.measureLine(inGeom.asPoint(),
                                            outGeom.asPoint())
                if matType == 0:
                    self.writer.addRecord([inID,unicode(outID),unicode(dist)])
                else:
                    distList.append(float(dist))

            if matType != 0:
                mean = sum(distList) / len(distList)
                for i in distList:
                    vari += (i - mean) * (i - mean)
                vari = math.sqrt(vari / len(distList))
                self.writer.addRecord([inID, unicode(mean),
                                      unicode(vari), unicode(min(distList)),
                                      unicode(max(distList))])

            current += 1
            progress.setPercentage(int(current * total))

    def regularMatrix(self, inLayer, inField, targetLayer, targetField,
                      nPoints, progress):
        index = vector.spatialindex(targetLayer)

        inIdx = inLayer.fieldNameIndex(inField)

        outFeat = QgsFeature()
        inGeom = QgsGeometry()
        outGeom = QgsGeometry()
        distArea = QgsDistanceArea()

        first = True
        current = 0
        features = vector.features(inLayer)
        total = 100.0 / float(len(features))

        for inFeat in features:
            inGeom = inFeat.geometry()
            inID = unicode(inFeat.attributes()[inIdx])
            featList = index.nearestNeighbor(inGeom.asPoint(), nPoints)
            if first:
                first = False
                data = ['ID']
                for i in range(len(featList)):
                    data.append('DIST_{0}'.format(i+1))
                self.writer.addRecord(data)

            data = [inID]
            for i in featList:
                request = QgsFeatureRequest().setFilterFid(i)
                outFeat = targetLayer.getFeatures(request).next()
                outGeom = outFeat.geometry()
                dist = distArea.measureLine(inGeom.asPoint(),
                                            outGeom.asPoint())
                data.append(unicode(float(dist)))
            self.writer.addRecord(data)

            current += 1
            progress.setPercentage(int(current * total))