bus-service: comparison buses.py

-:9139a94e540c
+:3b86597c5a88
 all_clusters = set()
 viimeinen_käyttöpäivä = None
 clusters_by_name = {}
 services_for_day = {}
-def load_buses(gtfs_zip_path):
+def load_buses(gtfs_zip_path, regions):
 	global viimeinen_käyttöpäivä
 	from zipfile import ZipFile
 	with ZipFile(gtfs_zip_path) as gtfs_zip:
-		print('Ladataan linjat... ', file = stderr, end = '', flush = True)
+		print('Loading routes... ', file = stderr, end = '', flush = True)
 		with gtfs_zip.open('routes.txt') as file:
 			for row in read_csv(map(bytes.decode, file)):
 				route = BusRoute(row)
 				routes[route.reference] = route
 				routes_per_id[route.id] = route
-		print('%d linjaa' % len(routes), file = stderr)
+		print('%d routes' % len(routes), file = stderr)
+		print('Loading trips... ', file = stderr, end = '', flush = True)
-		print('Ladataan ajovuorot... ', file = stderr, end = '', flush = True)
 		shape_distances = {}
 		try:
 			with gtfs_zip.open('shapes.txt') as file:
 				for row in read_csv(map(bytes.decode, file)):
 					shape_distances[row['shape_id']] = max(shape_distances.get(row['shape_id'], 0), float(row['shape_dist_traveled']))
 				route.trips.add(trip)
 				if trip.name in all_trips:
 					print('Trip %s already exists' % trip.name)
 				else:
 					all_trips[trip.name] = trip
-		print('%d ajoa' % len(all_trips), file = stderr)
+		print('%d trips' % len(all_trips), file = stderr)
 		def read_date(teksti):
 			return date(int(teksti[:4]), int(teksti[4:6]), int(teksti[6:]))
 		def read_time(teksti):
-			tunti, minuutti, sekunti = map(int, teksti.split(':'))
+			hour, minute, second = map(int, teksti.split(':'))
-			return timedelta(hours = tunti, minutes = minuutti, seconds = sekunti)
+			return timedelta(hours = hour, minutes = minute, seconds = second)
 		print('Ladataan päiväykset... ', file = stderr, flush = True)
 		viimeinen_käyttöpäivä = date.today()
 		def services_available_at(day):
 			for service in services.values():
 				if day in service.dates:
 					yield service
-		print('Ladataan pysäkit... ', file = stderr, end = '', flush = True)
+		print('Loading stops... ', file = stderr, end = '', flush = True)
 		with gtfs_zip.open('stops.txt') as file:
 			for row in read_csv(map(bytes.decode, file)):
-				location = Sijainti(float(row['stop_lat']), float(row['stop_lon']))
+				location = Location(float(row['stop_lat']), float(row['stop_lon']))
 				stop = BusStop(
 					reference = row['stop_id'],
 					name = row['stop_name'],
 					location = location,
 					code = row.get('stop_code', row['stop_id']),
 					except KeyError:
 						pass
 			for bus_stop in bus_stops.values():
 				if not hasattr(bus_stop, 'region'):
 					bus_stop.region = None
-		print('%d pysäkkiä' % len(bus_stops), file = stderr)
+		print('%d stops' % len(bus_stops), file = stderr)
 		class BusStopCluster:
 			def __init__(self):
 				self.stops = set()
-				self._center = None
+				self.cached_center = None
 				self.name = None
 			@property
 			def url_name(self):
 				return self.name.lower().replace('(', '').replace(')', '').replace(' ', '-')
 			def add_stop(self, stop):
 				assert not stop.cluster
 				stop.cluster = self
 				self.stops.add(stop)
-				self._center = None
+				self.cached_center = None
 			@property
 			def center(self):
-				if not self._center:
+				if not self.cached_center:
 					if self.stops:
 						from statistics import median
 						pointtype = type(next(iter(self.stops)).location)
-						self._center = pointtype(
+						self.cached_center = pointtype(
 							median(stop.location.x for stop in self.stops),
 							median(stop.location.y for stop in self.stops),
 						)
 					else:
 						raise ValueError('an empty cluster has no center point')
-				return self._center
+				return self.cached_center
 			def merge(self, other):
 				for bus_stop in other.stops:
 					bus_stop.cluster = self
 				self.stops |= other.stops
 				other.stops = set()
-				other._center = None
+				other.cached_center = None
 			def schedule(self, *, max_amount = 50):
 				result = []
 				for stop in self.stops:
 					result += stop.schedule(max_amount = max_amount)
 				result.sort(key = lambda schedule_entry: schedule_entry['time'])
 			for bus_stop in sorted_bus_stops:
 				if not bus_stop.cluster:
 					stops_to_cluster = {bus_stop}
 					# etsi pysäkin samannimiset vastaparit
 					for pair_candidate in bus_stops_by_name[bus_stop.name]:
-						distance = pair_candidate.location.etäisyys(bus_stop.location)
+						distance = pair_candidate.location.distance(bus_stop.location)
 						if pair_candidate is not bus_stop and distance <= 0.4:
 							stops_to_cluster.add(pair_candidate)
 					for stop_to_cluster in stops_to_cluster:
 						if stop_to_cluster.cluster:
 							cluster = stop_to_cluster.cluster
 			for bus_stop in sorted_bus_stops:
 				if len(bus_stop.cluster.stops) == 1:
 					possibilities = set()
 					for cluster in all_clusters:
 						if cluster is not bus_stop.cluster:
-							distance = cluster.center.etäisyys(bus_stop.location)
+							distance = cluster.center.distance(bus_stop.location)
 							if distance <= 0.4:
 								possibilities.add((distance, cluster))
 					if possibilities:
 						best = min(possibilities)[1]
 						all_clusters.remove(bus_stop.cluster)
 				result |= pair[0] & pair[1]
 			return result
 		def name_clusters():
 			from collections import defaultdict
-			from pprint import pprint
 			clusters_per_name = defaultdict(set)
 			for cluster in all_clusters:
 				name_representing_stop = min((len(stop.reference), stop.reference, stop) for stop in cluster.stops)[2]
 				clusters_per_name[name_representing_stop.name].add(cluster)
 			for name, clusters in clusters_per_name.items():
 				if len(clusters) == 1:
-					# Ryhmä on ainoa jolla on varaus tälle nimelle. Sen kuin vaan.
+					# Simple case: this cluster is the only one that wants this name.
 					next(iter(clusters)).name = name
 				else:
 					if profile['regions']['use-regions']:
-						# Olisiko kaikki klusterit eri alueilla?
+						# Find out if all clusters are in different areas
 						common_regions = shared_elements_in_n_sets({stop.region for stop in cluster.stops} for cluster in clusters)
-						# Esitys: ryhmä -> ne alueet jotka ovat tälle ryhmälle ainutlaatuisia
+						# Proposal: cluster -> the areas unique to the cluster
 						proposal = {
 							cluster: {stop.region for stop in cluster.stops} - common_regions - {None}
 							for cluster in clusters
 						}
-						# Jos enintään yksi klusteri tässä esityksessä on kokonaan ilman omaa aluetta, jolla se voisi eritellä,
+						# If at most one cluster is without its own unique region, name the others by region and this one without any.
-						# niin nimetään klusterit näiden alueiden mukaan.
-						# Se klusteri jolla ei ole omaa aluetta (jos on) jätetään ilman aluepäätettä.
 						if sum([1 for unique_areas in proposal.values() if not unique_areas]) <= 1:
 							for cluster, unique_areas in proposal.items():
 								individual_cluster_name = name
 								if unique_areas:
 									individual_cluster_name += ' (' + min(unique_areas) + ')'
 								cluster.name = individual_cluster_name
 								break
-					# Typerä reunatapaus. Indeksoidaan numeroin...
+					# If all else fails, just number them.
 					for n, (_, cluster) in enumerate(sorted(
 						min((stop.reference.lower(), cluster) for stop in cluster.stops)
 						for cluster in clusters
 					), 1):
 						individual_cluster_name = name + '-' + str(n)
 						cluster.name = individual_cluster_name
-		print('Ryhmitellään pysäkit...')
+		print('Clustering bus stops...')
 		cluster_bus_stops()
 		name_clusters()
 		for cluster in all_clusters:
 			if cluster.url_name in clusters_by_name:
 				print('Warning: Clusters %r and %r share the same URL name: %r' % (cluster.name, clusters_by_name[cluster.url_name].name, cluster.url_name))
 			else:
 				clusters_by_name[cluster.url_name] = cluster
-		print('Ladataan aikataulut... ', end = '', flush = True, file = stderr)
+		print('Loading schedules... ', end = '', flush = True, file = stderr)
 		with gtfs_zip.open('stop_times.txt') as file:
 			row_count = sum(line.count(b'\n') for line in file)
 		with gtfs_zip.open('stop_times.txt') as file:
 			progress = 0
 			for row in read_csv(map(bytes.decode, file)):
 				traveled_distance = float(row.get('shape_dist_traveled', 1)) * float(profile['metrics']['shape-modifier'])
 				trip.schedule.append(BusHalt(arrival_time, departure_time, stop, trip, traveled_distance))
 				stop.involved_trips.add(trip)
 				progress += 1
 				if progress % 1000 == 0:
-					print('\rLadataan aikataulut... %.1f%%' % (progress * 100 / row_count), end = ' ', file = stderr)
+					print('\rLoading schedules... %.1f%%' % (progress * 100 / row_count), end = ' ', file = stderr)
-		print('\rLadataan aikataulut... ladattu', file = stderr)
+		print('\rLoading schedules... complete', file = stderr)
 		for trip in all_trips.values():
 			from busroute import simplify_name
 			schedule = trip.concise_schedule()
 			try:
 			trips_by_vehicle_info[(trip.block_id, trip.schedule[0].arrival_time)] = trip
 if __name__ == '__main__':
 	profile.read('profiles/föli.ini')
 	load_buses('gtfs.zip')
+	import busroute
+	from regions import parse_regions
+	busroute.regions = parse_regions('föli.osm')

Mercurial > bus-service / file comparison

comparison: buses.py

buses.py