def _read_all_data(self):
"""
Read all data into dict all_data
"""
for train in self.all_trains:
self.all_data[train] = []
for filename in self.filelist:
infile = csv.reader(open(filename), delimiter=self.delim)
#TODO(Simon): verify code snippet:
#/start snippet
values = {}
for train in self.all_trains:
values[train] = False
# put rows into variable all_data
for row in infile:
if (row[0] and #skip empty rows
not str_eq(row[0], 'Tognr.')): # skip legend
self.all_data[row[0]].append(row)
values[row[0]] = True #remember which trains had data
#list missing values
for key in values:
if not values[key]:
self.all_data[key].append('missing')
def main(args):
"""
Main function for module train_sorting.py
"""
if len(args) > 1 and str_eq(args[1], 'test'):
test()
def main(args):
"""
Main function of module train_io.py
"""
if len(args) > 1 and str_eq(args[1], 'test'):
inout = TrainIo('Oslo')
inout.test()
def _cache_windows(self, scheduled, dep_windows, arr_windows):
"""
Finds windows for a dict/list of trains
re
"""
for sched in scheduled:
for train in self.get_all_trains():
if not str_eq(sched, train):
#Spags?
done = False
while not done:
for row in self.all_data[train]:
if not str_eq(row, 'missing'):
#Check if train departs within a 15 min window
#of scheds departure/arrival
if(row[self.legend['Pl.avg.']] and
abs(time_diff(row[self.legend['Pl.avg.']],
scheduled[sched]))
< self.windowsize/2.):
if not sched in dep_windows:
dep_windows[sched] = []
dep_windows[sched].append(train)
done = True
#Check if train arrives within a 15 min window
#of scheds departure/arrival
if(row[self.legend['Pl.ank.']] and
abs(time_diff(row[self.legend['Pl.ank.']]
, scheduled[sched]))
< self.windowsize/2.):
if not sched in arr_windows:
arr_windows[sched] = []
arr_windows[sched].append(train)
done = True
if done:
"""
Here I make the assumption that if a train has a
scheduled departure/arrival time then it is
stored in all the csv files. Also that the
planned departure/arrival time is the same
for all days, which is not always the case,
but for most trains it is.
"""#pylint:disable=pointless-string-statement
break
done = True
return dep_windows, arr_windows
def _list_all_trains(self):
"""
List all the trains
"""
for filename in self.filelist:
infile = csv.reader(open(filename), delimiter=self.delim)
for row in infile:
if row[0] and not str_eq(row[0], 'Tognr.'):
if row[0] not in self.all_trains:
self.all_trains.append(row[0])
def _cache_sched(self):
"""
Reads all_data to find all scheduled departures
and arrivals
"""
for train in self.get_all_trains():
for row in self.all_data[train]:
if not str_eq(row, 'missing'):
#Remember trains arr time if scheduled
if row[self.legend['Pl.ank.']]:
self.sched_arrs[train] = row[self.legend['Pl.ank.']]
#Remember trains dep time if scheduled
if row[self.legend['Pl.avg.']]:
self.sched_deps[train] = row[self.legend['Pl.avg.']]
def wrt_trains(self, train1, train2, t1_dir='Departure',
t2_dir='Departure'):
"""
Writes data from train1 and train2 to
a file named train1_train2.txt
"""
t1_arr = None
t2_arr = None
#TODO(Simon): raise errors
if str_eq(t1_dir, 'Arrival'):
if str_eq(t2_dir, 'Arrival'):
t1_arr = self.data_to_array(train1, 'Pl.ank.',
'Fakt.ank.')
t2_arr = self.data_to_array(train2, 'Pl.ank.',
'Fakt.ank.')
elif str_eq(t2_dir, 'Departure'):
t1_arr = self.data_to_array(train1, 'Pl.ank.',
'Fakt.ank.')
t2_arr = self.data_to_array(train2, 'Pl.avg.',
'Fakt.avg.')
else:
#Raise some kind of error
pass
elif str_eq(t1_dir, 'Departure'):
if str_eq(t2_dir, 'Arrival'):
t1_arr = self.data_to_array(train1, 'Pl.avg.',
'Fakt.avg.')
t2_arr = self.data_to_array(train2, 'Pl.ank.',
'Fakt.ank.')
elif str_eq(t2_dir, 'Departure'):
t1_arr = self.data_to_array(train1, 'Pl.avg.',
'Fakt.avg.')
t2_arr = self.data_to_array(train2, 'Pl.avg.',
'Fakt.avg.')
else:
#Raise some knd of error
pass
else:
#Raise some kind of error
pass
#Filter out days with missing values(what does this do?)
outdata = train_sorting.sort_out_complete_pairs(t1_arr, t2_arr)
outfile = open(train1 + '_' + train2 + '.txt', 'w')
outfile.write(train1 + ' ' + train2 + '\n')
for i in range(len(outdata[0])):
outfile.write(str(outdata[0][i]) + ' ' +
str(outdata[1][i]) + '\n')
outfile.close()
def data_to_array(self, train, plan, act):
"""
Stores the delay times for a specified
trainnumber read from dict all_data
"""
train_data = self.all_data[train]
delay_list = []
for row in train_data:
if str_eq('missing', row):
delay_list.append(float('Nan'))
else:
try:
delay_list.append(time_diff(row[self.legend[act]],
row[self.legend[plan]]))
except ValueError:
delay_list.append(float('Nan'))
return numpy.array(delay_list) #pylint:disable=no-member
def main(args):
"""
Main
"""
if len(args) > 1 and str_eq(args[1], 'test'):
test()
def main(args):
"""
Main function for train_dependencies.py
"""
if len(args) > 1 and str_eq(args[1], 'test'):
test()
def main(args):
"""
Main function of train_algos.py
"""
if len(args) > 1 and str_eq(args[1], 'test'):
test()
