def parse_special_events_html(self, special_events_html):
special_events = dict()
special_events["specialEvents"] = []
if not special_events_html:
return special_events
try:
soup = BeautifulSoup(special_events_html)
event_tables = soup.findAll("table", {"class": "event-item"})
for event_table in event_tables:
rows = event_table.findAll("tr")
event_time = ""
event_date = ""
lot_num_array = []
event_name = ""
for row_index in range(0, 3):
if row_index == 0: # we're on the header row
header_content = rows[row_index].find("th").string
header_array = header_content.split(":")
event_date = header_array[0]
event_name = header_array[1].replace(" ", "")
elif row_index == 1: # time row
cells = rows[row_index].findAll("td")
cell_content = cells[1].string
event_time_arr = cell_content.split(" ", 2)
event_time = event_time_arr[0] + event_time_arr[1].replace(".", "").upper()
elif row_index == 2: # lots row
cells = rows[row_index].findAll("td")
cell_content = cells[1].string
lot_num_array = cell_content.replace(" ", "").split(",")
# strip leading 0's out of lot number array
for index, item in enumerate(lot_num_array):
lot_num_array[index] = self.strip_leading_zeros_from_short_name(item)
try: # the most brittle part of the parse
if len(event_time) == 1:
event_time = "0" + event_time
event_datetime_tmp = event_date + " " + event_time
event_datetime_str = datetime.datetime.strptime(event_datetime_tmp, "%m/%d/%Y %I:%M%p").strftime(
"%Y-%m-%dT%H:%M:%S"
)
except ValueError:
logging.error("Error parsing campus special event date")
event_datetime_str = None
# Rather than exclude props not currently available via uw lots, going with "None"
# This will manifest in the json as "property":null which should be easily detectable via client JS
special_event = {
"eventName": event_name,
"parkingLocations": lot_num_array,
"eventDatetime": event_datetime_str,
"parkingStartDatetime": None,
"parkingEndDatetime": None,
"eventVenue": None,
"webUrl": self.parking_data["special_events_url"],
}
special_events["specialEvents"].append(special_event)
except (ValueError, AttributeError, TypeError, IndexError) as e:
# unlike availability, we eat this error. availability is still useful w/out events
logging.error("Error parsing scraped content from campus special events page." + str(e))
special_events["specialEvents"] = []
return special_events
def parse_special_events_html(self, special_events_html):
special_events = dict()
special_events['specialEvents'] = []
if not special_events_html:
return special_events
try:
soup = BeautifulSoup(special_events_html)
event_tables = soup.findAll('table', {'class': 'event-item'})
for event_table in event_tables:
rows = event_table.findAll('tr')
event_time = ''
event_date = ''
lot_num_array = []
event_name = ''
for row_index in range(0, 3):
if row_index == 0: # we're on the header row
header_content = rows[row_index].find('th').string
header_array = header_content.split(':')
event_date = header_array[0]
event_name = header_array[1].replace(' ', '')
elif row_index == 1: # time row
cells = rows[row_index].findAll('td')
cell_content = cells[1].string
event_time_arr = cell_content.split(' ', 2)
event_time = event_time_arr[0] + event_time_arr[
1].replace('.', '').upper()
elif row_index == 2: # lots row
cells = rows[row_index].findAll('td')
cell_content = cells[1].string
lot_num_array = cell_content.replace(' ',
'').split(',')
# strip leading 0's out of lot number array
for index, item in enumerate(lot_num_array):
lot_num_array[
index] = self.strip_leading_zeros_from_short_name(
item)
try: # the most brittle part of the parse
if len(event_time) == 1:
event_time = '0' + event_time
event_datetime_tmp = event_date + ' ' + event_time
event_datetime_str = datetime.datetime.strptime(
event_datetime_tmp,
'%m/%d/%Y %I:%M%p').strftime('%Y-%m-%dT%H:%M:%S')
except ValueError:
logging.error('Error parsing campus special event date')
event_datetime_str = None
# Rather than exclude props not currently available via uw lots, going with "None"
# This will manifest in the json as "property":null which should be easily detectable via client JS
special_event = {
'eventName': event_name,
'parkingLocations': lot_num_array,
'eventDatetime': event_datetime_str,
'parkingStartDatetime': None,
'parkingEndDatetime': None,
'eventVenue': None,
'webUrl': self.parking_data['special_events_url']
}
special_events['specialEvents'].append(special_event)
except (ValueError, AttributeError, TypeError, IndexError) as e:
# unlike availability, we eat this error. availability is still useful w/out events
logging.error(
'Error parsing scraped content from campus special events page.'
+ str(e))
special_events['specialEvents'] = []
return special_events
