def output_areas(request, title, format, areas, **kwargs): areas = add_codes(areas) if format == 'map.html': format = 'html' kwargs['show_map'] = True if format == 'html': return output_html(request, title, areas, **kwargs) return output_json(iterdict((area.id, area.as_dict()) for area in areas))
def convert_address(request, format='json'): address = request.GET.get('address') if not address: raise ViewException(format, 'No address was provided.', 400) converter = AddressConverter() locations = converter.resolve_address(address, partial=bool(request.GET.get('partial'))) # this is a copy from mapit.views.areas.areas_by_point # because it's hard to reuse their code :( if PYGDAL: from osgeo import gdal gdal.UseExceptions() # we find areas for every lat/long coord we got back areas = [] type = request.GET.get('type', '') for coords in locations: location = Point(float(coords['lng']), float(coords['lat']), srid=4326) try: location.transform(settings.MAPIT_AREA_SRID, clone=True) except: raise ViewException(format, _('Point outside the area geometry'), 400) args = query_args(request, format) if type: args = dict(("area__%s" % k, v) for k, v in args.items()) # So this is odd. It doesn't matter if you specify types, PostGIS will # do the contains test on all the geometries matching the bounding-box # index, even if it could be much quicker to filter some out first # (ie. the EUR ones). coords['areas'] = [] args['polygon__bbcontains'] = location shapes = Geometry.objects.filter(**args).defer('polygon') for shape in shapes: try: area = Area.objects.get(polygons__id=shape.id, polygons__polygon__contains=location) coords['areas'].append(str(area.id)) areas.append(area) except: pass else: geoms = list(Geometry.objects.filter(polygon__contains=location).defer('polygon')) args['polygons__in'] = geoms matches = Area.objects.filter(**args).all() coords['areas'] = [str(m.id) for m in matches] areas.extend(matches) areas = add_codes(areas) if format == 'html': return output_html(request, _("Areas matching the address '{0}'").format(address), areas, indent_areas=True) # hack to include the geocoded addresses in the results data = iterdict(chain( ((area.id, area.as_dict()) for area in areas), [("addresses", locations)])) return output_json(data)
def convert_address(request, format='json'): address = request.GET.get('address') if not address: raise ViewException(format, 'No address was provided.', 400) converter = AddressConverter() locations = converter.resolve_address(address, partial=bool( request.GET.get('partial'))) # this is a copy from mapit.views.areas.areas_by_point # because it's hard to reuse their code :( if PYGDAL: from osgeo import gdal gdal.UseExceptions() # we find areas for every lat/long coord we got back areas = [] type = request.GET.get('type', '') for coords in locations: location = Point(float(coords['lng']), float(coords['lat']), srid=4326) try: location.transform(settings.MAPIT_AREA_SRID, clone=True) except: raise ViewException(format, _('Point outside the area geometry'), 400) args = query_args(request, format) if type: args = dict(("area__%s" % k, v) for k, v in args.items()) # So this is odd. It doesn't matter if you specify types, PostGIS will # do the contains test on all the geometries matching the bounding-box # index, even if it could be much quicker to filter some out first # (ie. the EUR ones). coords['areas'] = [] args['polygon__bbcontains'] = location shapes = Geometry.objects.filter(**args).defer('polygon') for shape in shapes: try: area = Area.objects.get( polygons__id=shape.id, polygons__polygon__contains=location) coords['areas'].append(str(area.id)) areas.append(area) except: pass else: geoms = list( Geometry.objects.filter( polygon__contains=location).defer('polygon')) args['polygons__in'] = geoms matches = Area.objects.filter(**args).all() coords['areas'] = [str(m.id) for m in matches] areas.extend(matches) areas = add_codes(areas) if format == 'html': return output_html( request, _("Areas matching the address '{0}'").format(address), areas, indent_areas=True) # hack to include the geocoded addresses in the results data = iterdict( chain(((area.id, area.as_dict()) for area in areas), [("addresses", locations)])) return output_json(data)