def stops(tables, static_feed, arg_agency_id):
columns = ['agency_id', 'stop_id', 'stop_code', 'stop_name', 'stop_desc', 'stop_lat', 'stop_lon',
'lat_lon', 'stop_url', 'location_type', 'parent_station', 'wheelchair_boarding', 'version']
# tables['Stops'] = pd.DataFrame(index=np.r_[0:len(static_feed['stops'].index)], columns=columns)
tables['Stops'] = pd.DataFrame()
print static_feed['stops'].columns
for i, row in static_feed['stops'].iterrows():
new_row = {}
new_row['agency_id'] = arg_agency_id
# new_row['stop_id'] = str(row['stop_id'])
# print "Name of first row: " + row.index[0]
if row.index[0] == 'stop_id':
new_row['stop_id'] = row['stop_id']
else:
new_row['stop_id'] = row[0]
new_row['stop_code'] = str(helper.optional_field(i, 'stop_code', static_feed['stops']))
new_row['stop_name'] = str(row['stop_name'])
new_row['stop_desc'] = str(helper.optional_field(i, 'stop_desc', static_feed['stops']))
new_row['stop_lat'] = float(row['stop_lat'])
new_row['stop_lon'] = float(row['stop_lon'])
# new_row['stop_url'] = str(helper.optional_field(i, 'stop_url', static_feed['stops']))
new_row['location_type'] = int(helper.optional_field(i, 'location_type', static_feed['stops'], 0))
new_row['parent_station'] = int(helper.optional_field(i, 'parent_station', static_feed['stops'], 0))
new_row['wheelchair_boarding'] = int(helper.optional_field(i, 'wheelchair_boarding', static_feed['stops'], 0))
tables['Stops'] = tables['Stops'].append(pd.Series(new_row), ignore_index=True)
helper.write_table(tables, "Stops")
print "SUCCESS with stops"
def route_stop_seq(tables, static_feed, trip_update_feed, alert_feed,
vehicle_position_feed, agency_id, trip2pattern):
count = 0
columns = [
'agency_id', 'route_short_name', 'route_dir', 'pattern_id', 'stop_id',
'seq', 'is_time_point', 'version', 'trip_id'
]
tables['Route_stop_seq'] = pd.DataFrame()
for i, row in static_feed['routes'].iterrows(
): #iterate through the different routes
route_id = row['route_id']
patterns = [] #the patterns
for j, subrow in static_feed['trips'].loc[
static_feed['trips']['route_id'] == route_id].iterrows():
trip_id = subrow['trip_id'] #get the trip_id
print trip_id
direction_id = subrow[
'direction_id'] if 'direction_id' in subrow else 0 #get the direction id in the trip
trip_id_block = static_feed['stop_times'].loc[
static_feed['stop_times']['trip_id'] == trip_id]
sequence = trip_id_block['stop_id'].tolist()
if str(sequence) not in patterns:
patterns += [str(sequence)]
pattern_num = patterns.index(str(sequence)) + 1
route_short_name = str(
helper.optional_field(i, 'route_long_name',
static_feed['routes']))
pattern_id = "{0}_{1}_{2}".format(route_short_name,
direction_id, pattern_num)
for k, subsubrow in trip_id_block.iterrows():
new_row = {}
new_row['trip_id'] = trip_id
new_row['agency_id'] = agency_id
new_row['route_short_name'] = route_short_name
new_row['route_dir'] = direction_id
new_row['pattern_id'] = pattern_id
new_row['stop_id'] = str(subsubrow['stop_id'])
new_row['seq'] = subsubrow['stop_sequence']
new_row['is_time_point'] = int(
helper.optional_field(k, 'timepoint',
static_feed['stop_times'], 0))
new_row['version'] = 1
#replace later
tables["Route_stop_seq"] = tables["Route_stop_seq"].append(
pd.Series(new_row), ignore_index=True)
count += 1
trip2pattern[trip_id] = pattern_id
with open('Trip2Pattern.csv', 'wb') as f:
writer = csv.writer(f)
writer.writerow(["trip_id", "pattern_id"])
for key, value in trip2pattern.items():
writer.writerow([key, value])
helper.write_table(tables, 'Route_stop_seq')
print "SUCCESS with Route Stop Seq"
def route_stop_seq(tables, static_feed, arg_agency_id, arg_route_id, trip2pattern):
count = 0
columns = ['agency_id', 'route_short_name', 'route_dir',
'pattern_id', 'stop_id', 'seq', 'is_time_point', 'version', 'trip_id']
tables['Route_stop_seq'] = pd.DataFrame()
for i, row in static_feed['routes'].iterrows(): #iterate through the different routes
if arg_route_id == 'all':
route_id = row['route_id']
else :
route_id = int(arg_route_id)
print route_id
patterns = [] #the patterns
for j, subrow in static_feed['trips'].loc[static_feed['trips']['route_id'] == route_id].iterrows():
trip_id = subrow['trip_id'] #get the trip_id
print trip_id
direction_id = subrow['direction_id'] if 'direction_id' in subrow else 0 #get the direction id in the trip
trip_id_block = static_feed['stop_times'].loc[static_feed['stop_times']['trip_id'] == trip_id]
sequence = trip_id_block['stop_id'].tolist()
if str(sequence) not in patterns:
patterns += [str(sequence)]
pattern_num = patterns.index(str(sequence)) + 1
route_short_name = str(helper.optional_field(i, 'route_long_name', static_feed['routes']))
pattern_id = "{0}_{1}_{2}".format(route_short_name, direction_id, pattern_num)
for k, subsubrow in trip_id_block.iterrows():
new_row = {}
new_row['trip_id'] = trip_id
new_row['agency_id'] = arg_agency_id
new_row['route_short_name'] = route_short_name
new_row['route_dir'] = direction_id
new_row['pattern_id'] = pattern_id
new_row['stop_id'] = str(subsubrow['stop_id'])
new_row['seq'] = subsubrow['stop_sequence']
new_row['is_time_point'] = int(helper.optional_field(k, 'timepoint', static_feed['stop_times'], 0))
new_row['version'] = 1; #replace later
tables["Route_stop_seq"] = tables["Route_stop_seq"].append(pd.Series(new_row), ignore_index=True)
count += 1
trip2pattern[trip_id] = pattern_id
if arg_route_id != 'all':
break
with open('Trip2Pattern.csv', 'wb') as f:
writer = csv.writer(f)
writer.writerow(["trip_id", "pattern_id"])
for key, value in trip2pattern.items():
writer.writerow([key, value])
helper.write_table(tables, 'Route_stop_seq')
print "SUCCESS with Route Stop Seq"
def runPattern(tables, static_feed, arg_agency_id):
columns = ['agency_id', 'route_short_name', 'start_date', 'end_date',
'service_id', 'day', 'route_dir', 'run', 'pattern_id', 'trip_headsign', 'trip_id', 'version']
tables['RunPattern'] = pd.DataFrame(index=np.r_[0:len(static_feed['trips'].index)], columns=columns)
run_count = {}
#open the trip2pattern file
trip2pattern = helper.csv2df("Trip2Pattern.csv") #load the trip2pattern csv
#iterate through eveyr trip
for i, row in static_feed['trips'].iterrows():
#get this specific row
new_row = tables["RunPattern"].loc[i]
new_row['agency_id'] = arg_agency_id
#getting the name of the route
j = np.where(static_feed['routes']['route_id'] == row['route_id'])[0][0]
new_row['route_short_name'] = str(helper.optional_field(j, 'route_short_name', static_feed['routes'], static_feed['routes'].iloc[j]['route_long_name']))
#use calendar.txt
new_row['service_id'] = row['service_id']
calendar = static_feed['calendar'].loc[static_feed['calendar']['service_id'] == row['service_id']].iloc[0]
new_row['start_date'] = datetime.datetime.strptime(str(calendar['start_date']), "%Y%m%d")
new_row['end_date'] = datetime.datetime.strptime(str(calendar['end_date']), "%Y%m%d")
new_row['route_dir'] = int(helper.optional_field(i, 'direction_id', static_feed['trips'], 0))
new_row['day'] = "{0}{1}{2}{3}{4}{5}{6}".format(calendar['monday'], calendar['tuesday'], calendar['wednesday'], calendar['thursday'], calendar['friday'], calendar['saturday'], calendar['sunday'])
#calculating the runs...Each run is unqiuely identified by
#1. route_short_name
#2. service_id
#3. route_dir
if new_row['route_short_name'] not in run_count.keys():
run_count[new_row['route_short_name']] = {new_row['service_id']: {new_row['route_dir'] : 1}}
if new_row['service_id'] not in run_count[new_row['route_short_name']].keys():
run_count[new_row['route_short_name']] = {new_row['service_id']: {new_row['route_dir'] : 1}}
if new_row['route_dir'] not in run_count[new_row['route_short_name']][new_row['service_id']].keys():
run_count[new_row['route_short_name']] = {new_row['service_id']: {new_row['route_dir'] : 1}}
#Todo: order runs by time of first stop in pattern
new_row['run'] = run_count[new_row['route_short_name']][new_row['service_id']][new_row['route_dir']]
run_count[new_row['route_short_name']][new_row['service_id']][new_row['route_dir']] += 1 #increment the run because we've seen this before....
new_row['pattern_id'] = str(trip2pattern[trip2pattern['trip_id'] == row['trip_id']].iloc[0]['pattern_id'])
# new_row['trip_headsign'] = helper.optional_field(i, 'trip_headsign', static_feed['trips'], static_feed['stop_times'].loc[static_feed['stop_times']['trip_id'] == row['trip_id']]['stop_headsign'].iloc[0])
new_row['trip_id'] = str(row['trip_id'])
new_row['version'] = 1
helper.write_table(tables, 'RunPattern')
print "SUCCESS with RunPatterns"
def agencies(tables, static_feed, arg_agency_id):
columns = ['agency_id', 'agency_name', 'agency_url', 'agency_timezone', 'agency_lang',
'agency_phone', 'timezone_name']
tables['Agency'] = pd.DataFrame(index=np.r_[0:len(static_feed['agency'].index)], columns=columns)
for i, row in static_feed['agency'].iterrows():
new_row = tables["Agency"].loc[i] #instantiate a NEW ROW
new_row['agency_id'] = arg_agency_id
new_row['agency_name'] = row['agency_name']
new_row['agency_url'] = row['agency_url']
timezone = pytz.timezone(row['agency_timezone'])
new_row['agency_timezone'] = timezone
new_row['agency_lang'] = helper.optional_field(i, 'agency_lang', static_feed['agency'])
new_row['agency_phone'] = helper.optional_field(i, 'agency_phone', static_feed['agency'])
new_row['timezone_name'] = row['agency_timezone']
helper.write_table(tables, 'Agency')
print "SUCCESS finished with agencies"
def stops(tables, static_feed, trip_update_feed, alert_feed,
vehicle_position_feed, agency_id):
columns = [
'agency_id', 'stop_id', 'stop_code', 'stop_name', 'stop_desc',
'stop_lat', 'stop_lon', 'lat_lon', 'stop_url', 'location_type',
'parent_station', 'wheelchair_boarding', 'version'
]
# tables['Stops'] = pd.DataFrame(index=np.r_[0:len(static_feed['stops'].index)], columns=columns)
tables['Stops'] = pd.DataFrame()
print static_feed['stops'].columns
for i, row in static_feed['stops'].iterrows():
new_row = {}
new_row['agency_id'] = agency_id
# new_row['stop_id'] = str(row['stop_id'])
# print "Name of first row: " + row.index[0]
if row.index[0] == 'stop_id':
new_row['stop_id'] = row['stop_id']
else:
new_row['stop_id'] = row[0]
new_row['stop_code'] = str(
helper.optional_field(i, 'stop_code', static_feed['stops']))
new_row['stop_name'] = str(row['stop_name'])
new_row['stop_desc'] = str(
helper.optional_field(i, 'stop_desc', static_feed['stops']))
new_row['stop_lat'] = float(row['stop_lat'])
new_row['stop_lon'] = float(row['stop_lon'])
# new_row['stop_url'] = str(helper.optional_field(i, 'stop_url', static_feed['stops']))
new_row['location_type'] = int(
helper.optional_field(i, 'location_type', static_feed['stops'], 0))
new_row['parent_station'] = int(
helper.optional_field(i, 'parent_station', static_feed['stops'],
0))
new_row['wheelchair_boarding'] = int(
helper.optional_field(i, 'wheelchair_boarding',
static_feed['stops'], 0))
tables['Stops'] = tables['Stops'].append(pd.Series(new_row),
ignore_index=True)
helper.write_table(tables, "Stops")
print "SUCCESS with stops"
def agencies(tables, static_feed, arg_agency_id):
columns = [
'agency_id', 'agency_name', 'agency_url', 'agency_timezone',
'agency_lang', 'agency_phone', 'timezone_name'
]
tables['Agency'] = pd.DataFrame(
index=np.r_[0:len(static_feed['agency'].index)], columns=columns)
for i, row in static_feed['agency'].iterrows():
new_row = tables["Agency"].loc[i] #instantiate a NEW ROW
new_row['agency_id'] = arg_agency_id
new_row['agency_name'] = row['agency_name']
new_row['agency_url'] = row['agency_url']
timezone = pytz.timezone(row['agency_timezone'])
new_row['agency_timezone'] = timezone
new_row['agency_lang'] = helper.optional_field(i, 'agency_lang',
static_feed['agency'])
new_row['agency_phone'] = helper.optional_field(
i, 'agency_phone', static_feed['agency'])
new_row['timezone_name'] = row['agency_timezone']
helper.write_table(tables, 'Agency')
print "SUCCESS finished with agencies"
def routes(tables, static_feed, trip_update_feed, alert_feed,
vehicle_position_feed, agency_id):
columns = [
'agency_id', 'route_short_name', 'route_dir', 'route_type',
'route_long_name', 'route_desc', 'route_url', 'route_color',
'route_text_color', 'route_id', 'version'
]
tables['Routes'] = pd.DataFrame()
for i, row in static_feed['routes'].iterrows():
for direction_id in static_feed['trips'].loc[
static_feed['trips']['route_id'] ==
row['route_id']]['direction_id'].unique():
new_row = {}
new_row['agency_id'] = agency_id
new_row['route_short_name'] = str(
helper.optional_field(
i, 'route_short_name', static_feed['routes'],
static_feed['routes'].iloc[i]['route_long_name']))
new_row['route_dir'] = direction_id
new_row['route_type'] = int(row['route_type'])
new_row['route_long_name'] = str(
helper.optional_field(
i, 'route_long_name', static_feed['routes'],
static_feed['routes'].iloc[i]['route_short_name']))
new_row['route_desc'] = helper.optional_field(
i, 'route_desc', static_feed['routes'])
new_row['route_url'] = helper.optional_field(
i, 'route_url', static_feed['routes'])
new_row['route_color'] = helper.optional_field(
i, 'route_color', static_feed['routes'],
default='FFFFFF').upper()
new_row['route_text_color'] = helper.optional_field(
i, 'route_text_color', static_feed['routes'],
default='000000').upper()
new_row['route_id'] = str(row['route_id'])
new_row['version'] = 1
tables['Routes'] = tables['Routes'].append(pd.Series(new_row),
ignore_index=True)
helper.write_table(tables, 'Routes')
print "SUCCESS with routes"
def routes(tables, static_feed, arg_agency_id):
columns = ['agency_id', 'route_short_name', 'route_dir',
'route_type', 'route_long_name', 'route_desc',
'route_url', 'route_color', 'route_text_color', 'route_id', 'version']
tables['Routes'] = pd.DataFrame()
for i, row in static_feed['routes'].iterrows():
for direction_id in static_feed['trips'].loc[static_feed['trips']['route_id'] == row['route_id']]['direction_id'].unique():
new_row = {}
new_row['agency_id'] = arg_agency_id
new_row['route_short_name'] = str(helper.optional_field(i, 'route_short_name', static_feed['routes'], static_feed['routes'].iloc[i]['route_long_name']))
new_row['route_dir'] = direction_id
new_row['route_type'] = int(row['route_type'])
new_row['route_long_name'] = str(helper.optional_field(i, 'route_long_name', static_feed['routes'], static_feed['routes'].iloc[i]['route_short_name']))
new_row['route_desc'] = helper.optional_field(i, 'route_desc', static_feed['routes'])
new_row['route_url'] = helper.optional_field(i, 'route_url', static_feed['routes'])
new_row['route_color'] = helper.optional_field(i, 'route_color', static_feed['routes'], default='FFFFFF').upper()
new_row['route_text_color'] = helper.optional_field(i, 'route_text_color', static_feed['routes'], default='000000').upper()
new_row['route_id'] = str(row['route_id'])
new_row['version'] = 1
tables['Routes'] = tables['Routes'].append(pd.Series(new_row), ignore_index=True)
helper.write_table(tables, 'Routes')
print "SUCCESS with routes"
def schedules(tables, static_feed, arg_agency_id, trip2pattern):
try:
login = {'host': "localhost", 'user': SQL_USER,
'passwd': SQL_PWD, 'db': "PATHTransit"}
run_pattern_df = helper.sql2df('RunPattern', login)
agency_name = static_feed['agency'].iloc[0]['agency_name']
trip2pattern = helper.csv2df("Trip2Pattern.csv") #load the trip2pattern csv
route_stop_seq_df = helper.sql2df('Route_stop_seq', login)
except Exception as e:
print e
columns = ['agency_id', 'route_short_name', 'start_date', 'end_date', 'day',
'route_dir', 'run', 'pattern_id', 'seq', 'stop_id',
'is_time_point', 'pickup_type', 'drop_off_type',
'arrival_time', 'departure_time', 'stop_headsign', 'trip_id']
tables["Schedules"] = pd.DataFrame()
pattern_id_memoization = {} #to save calculations and memory
counter = 0
#use the unique tripid's to find the the patternid's specific to that trip
for a, row in static_feed['stop_times'].iterrows():
print counter
counter += 1
#find the pattern_id(What is the point of pattern id? we already have the stops.)
pattern_id = trip2pattern[trip2pattern['trip_id'] == row['trip_id']]['pattern_id']
new_row = {}
#agency_id basic info
new_row['agency_id'] = arg_agency_id
#information from pattern_id
current_trip_id = row['trip_id']
if current_trip_id in pattern_id_memoization.keys():
new_row['pattern_id'] = pattern_id_memoization[current_trip_id]
print current_trip_id
else: #access trip2pattern.csv to match the trip_id with the pattern
pattern_id = trip2pattern[trip2pattern['trip_id'] == current_trip_id].iloc[0]['pattern_id']
pattern_id_memoization[current_trip_id] = pattern_id
new_row['pattern_id'] = pattern_id_memoization[current_trip_id]
print "current trip id: " + str(current_trip_id)
# print route_stop_seq_df
#get information by matching with route_stop_seq matching by trip_id
route_stop_seq_stop_time_specific = route_stop_seq_df[route_stop_seq_df['trip_id'] == current_trip_id]
if not route_stop_seq_stop_time_specific.empty:
route_stop_seq_stop_time_specific = route_stop_seq_stop_time_specific.iloc[0]
new_row['route_short_name'] = route_stop_seq_stop_time_specific['route_short_name']
new_row['route_dir'] = route_stop_seq_stop_time_specific['route_dir']
#get information using run pattern by matching with trip_id
run_pattern_stop_time_specific = run_pattern_df[run_pattern_df['trip_id'] == current_trip_id]
if not run_pattern_stop_time_specific.empty:
run_pattern_stop_time_specific = run_pattern_stop_time_specific.iloc[0]
new_row['start_date'] = run_pattern_stop_time_specific['start_date']
new_row['end_date'] = run_pattern_stop_time_specific['end_date']
new_row['day'] = run_pattern_stop_time_specific['day']
new_row['run'] = run_pattern_stop_time_specific['run']
#all information in stop_times.txt
new_row['seq'] = row['stop_sequence']
new_row['stop_id'] = row['stop_id']
#time point is optional
new_row['is_time_point'] = helper.optional_field(a, 'time_point', static_feed['stop_times'])
# new_row['is_time_point'] = row['timepoint']
new_row['pickup_type'] = helper.optional_field(a, 'pickup_type', static_feed['stop_times'])
# new_row['pickup_type'] = row['pickup_type']
# new_row['drop_off_type'] = row['drop_off_type']
new_row['pickup_type'] = helper.optional_field(a, 'drop_off_type', static_feed['stop_times'])
new_row['arrival_time'] = row['arrival_time']
new_row['departure_time'] = row['departure_time']
# new_row['stop_headsign'] = row['stop_headsign']
new_row['stop_headsign'] = helper.optional_field(a, 'stop_headsign', static_feed['stop_times'])
new_row['trip_id'] = row['trip_id']
tables["Schedules"] = tables["Schedules"].append(pd.Series(new_row), ignore_index=True)
helper.write_table(tables, "Schedules")
print "Sucess with Schedules"
def schedules(tables, static_feed, trip_update_feed, alert_feed,
vehicle_position_feed, agency_id, trip2pattern):
try:
login = {
'host': "localhost",
'user': "******",
'passwd': "root",
'db': "TrafficTransit"
}
run_pattern_df = helper.sql2df('RunPattern', login)
agency_name = static_feed['agency'].iloc[0]['agency_name']
trip2pattern = helper.csv2df(
"Trip2Pattern.csv") #load the trip2pattern csv
route_stop_seq_df = helper.sql2df('route_stop_seq', login)
except Exception as e:
print e
columns = [
'agency_id', 'route_short_name', 'start_date', 'end_date', 'day',
'route_dir', 'run', 'pattern_id', 'seq', 'stop_id', 'is_time_point',
'pickup_type', 'drop_off_type', 'arrival_time', 'departure_time',
'stop_headsign', 'trip_id'
]
tables["Schedules"] = pd.DataFrame()
pattern_id_memoization = {} #to save calculations and memory
counter = 0
#use the unique tripid's to find the the patternid's specific to that trip
for a, row in static_feed['stop_times'].iterrows():
print counter
counter += 1
#find the pattern_id(What is the point of pattern id? we already have the stops.)
pattern_id = trip2pattern[trip2pattern['trip_id'] ==
row['trip_id']]['pattern_id']
new_row = {}
#agency_id basic info
new_row['agency_id'] = agency_id
#information from pattern_id
current_trip_id = row['trip_id']
if current_trip_id in pattern_id_memoization.keys():
new_row['pattern_id'] = pattern_id_memoization[current_trip_id]
print current_trip_id
else: #access trip2pattern.csv to match the trip_id with the pattern
pattern_id = trip2pattern[trip2pattern['trip_id'] ==
current_trip_id].iloc[0]['pattern_id']
pattern_id_memoization[current_trip_id] = pattern_id
new_row['pattern_id'] = pattern_id_memoization[current_trip_id]
print "current trip id: " + str(current_trip_id)
# print route_stop_seq_df
#get information by matching with route_stop_seq matching by trip_id
route_stop_seq_stop_time_specific = route_stop_seq_df[
route_stop_seq_df['trip_id'] == current_trip_id]
if not route_stop_seq_stop_time_specific.empty:
route_stop_seq_stop_time_specific = route_stop_seq_stop_time_specific.iloc[
0]
new_row['route_short_name'] = route_stop_seq_stop_time_specific[
'route_short_name']
new_row['route_dir'] = route_stop_seq_stop_time_specific[
'route_dir']
#get information using run pattern by matching with trip_id
run_pattern_stop_time_specific = run_pattern_df[
run_pattern_df['trip_id'] == current_trip_id]
if not run_pattern_stop_time_specific.empty:
run_pattern_stop_time_specific = run_pattern_stop_time_specific.iloc[
0]
new_row['start_date'] = run_pattern_stop_time_specific[
'start_date']
new_row['end_date'] = run_pattern_stop_time_specific['end_date']
new_row['day'] = run_pattern_stop_time_specific['day']
new_row['run'] = run_pattern_stop_time_specific['run']
#all information in stop_times.txt
new_row['seq'] = row['stop_sequence']
new_row['stop_id'] = row['stop_id']
#time point is optional
new_row['is_time_point'] = helper.optional_field(
a, 'time_point', static_feed['stop_times'])
# new_row['is_time_point'] = row['timepoint']
new_row['pickup_type'] = helper.optional_field(
a, 'pickup_type', static_feed['stop_times'])
# new_row['pickup_type'] = row['pickup_type']
# new_row['drop_off_type'] = row['drop_off_type']
new_row['pickup_type'] = helper.optional_field(
a, 'drop_off_type', static_feed['stop_times'])
new_row['arrival_time'] = row['arrival_time']
new_row['departure_time'] = row['departure_time']
# new_row['stop_headsign'] = row['stop_headsign']
new_row['stop_headsign'] = helper.optional_field(
a, 'stop_headsign', static_feed['stop_times'])
new_row['trip_id'] = row['trip_id']
tables["Schedules"] = tables["Schedules"].append(pd.Series(new_row),
ignore_index=True)
helper.write_table(tables, "Schedules")
print "Sucess with Schedules"
def runPattern(tables, static_feed, agency_id):
columns = [
'agency_id', 'route_short_name', 'start_date', 'end_date',
'service_id', 'day', 'route_dir', 'run', 'pattern_id', 'trip_headsign',
'trip_id', 'version'
]
tables['RunPattern'] = pd.DataFrame(
index=np.r_[0:len(static_feed['trips'].index)], columns=columns)
run_count = {}
#open the trip2pattern file
trip2pattern = helper.csv2df(
"Trip2Pattern.csv") #load the trip2pattern csv
#iterate through eveyr trip
for i, row in static_feed['trips'].iterrows():
#get this specific row
new_row = tables["RunPattern"].loc[i]
new_row['agency_id'] = agency_id
#getting the name of the route
j = np.where(
static_feed['routes']['route_id'] == row['route_id'])[0][0]
new_row['route_short_name'] = str(
helper.optional_field(
j, 'route_short_name', static_feed['routes'],
static_feed['routes'].iloc[j]['route_long_name']))
#use calendar.txt
new_row['service_id'] = row['service_id']
calendar = static_feed['calendar'].loc[
static_feed['calendar']['service_id'] == row['service_id']].iloc[0]
new_row['start_date'] = datetime.datetime.strptime(
str(calendar['start_date']), "%Y%m%d")
new_row['end_date'] = datetime.datetime.strptime(
str(calendar['end_date']), "%Y%m%d")
new_row['route_dir'] = int(
helper.optional_field(i, 'direction_id', static_feed['trips'], 0))
new_row['day'] = "{0}{1}{2}{3}{4}{5}{6}".format(
calendar['monday'], calendar['tuesday'], calendar['wednesday'],
calendar['thursday'], calendar['friday'], calendar['saturday'],
calendar['sunday'])
#calculating the runs...Each run is unqiuely identified by
#1. route_short_name
#2. service_id
#3. route_dir
if new_row['route_short_name'] not in run_count.keys():
run_count[new_row['route_short_name']] = {
new_row['service_id']: {
new_row['route_dir']: 1
}
}
if new_row['service_id'] not in run_count[
new_row['route_short_name']].keys():
run_count[new_row['route_short_name']] = {
new_row['service_id']: {
new_row['route_dir']: 1
}
}
if new_row['route_dir'] not in run_count[new_row['route_short_name']][
new_row['service_id']].keys():
run_count[new_row['route_short_name']] = {
new_row['service_id']: {
new_row['route_dir']: 1
}
}
#Todo: order runs by time of first stop in pattern
new_row['run'] = run_count[new_row['route_short_name']][
new_row['service_id']][new_row['route_dir']]
run_count[new_row['route_short_name']][new_row['service_id']][new_row[
'route_dir']] += 1 #increment the run because we've seen this before....
new_row['pattern_id'] = str(trip2pattern[
trip2pattern['trip_id'] == row['trip_id']].iloc[0]['pattern_id'])
# new_row['trip_headsign'] = helper.optional_field(i, 'trip_headsign', static_feed['trips'], static_feed['stop_times'].loc[static_feed['stop_times']['trip_id'] == row['trip_id']]['stop_headsign'].iloc[0])
new_row['trip_id'] = str(row['trip_id'])
new_row['version'] = 1
helper.write_table(tables, 'RunPattern')
print "SUCCESS with RunPatterns"