def fetch_timetable(browser, link):
# Utility method to return a nice Dataset from a timetable url
if debug:
puts('Fetching timetable from %s' % link)
response = browser.follow_link(link)
soup = BeautifulSoup(response.read())
table = soup.find('table')
timetable = []
for row in table.findAll('tr'):
title = None
title_test = row.find('td')
if title_test.find('span'):
title = title_test.getText()
values = []
for col in row.findAll('td')[1:]:
value = col.getText()
if value == ' ':
value = None
if isinstance(value, basestring) and ':' in value:
try:
time = value.strip().split(':')
time = datetime.time(*[int(s) for s in time])
except:
pass
else:
value = time
values.append(value)
timetable.append((title, values))
while len(timetable):
if 'TRAIN NO.' not in timetable[0][0]:
del timetable[0]
else:
break
train_nums = timetable[0]
data = Dataset()
data.headers = train_nums[1]
if debug:
puts(repr(data.headers))
for place, times in timetable[1:]:
if debug:
puts(repr((place, times)))
data.rpush(times, tags=[place.title().replace('`S', "'s")])
#Strip out TRAIN NO. columns
while 1:
try:
del data['TRAIN NO.']
except:
break
return data
