tt_parser.py

import bs4
import urllib2
import urlparse
import string
from tt_generator import *

class HTMLCourseParser:

	def __init__(self, url):
		self.url = fix_url(url)

	def parse(self):
		course = Course(extract_name_from_url(self.url))
		
		html = urllib2.urlopen(self.url)
		soup = bs4.BeautifulSoup(html)
		table = soup.body.find('table', attrs={'class' : 'table'})
		rows = table.find_all('tr')
		rows = rows[1:]

		shifts = []
		current_shift = Shift('None')
		for tr in rows:
			tds = tr.find_all('td')
		
			shift_name = str(tds[0].contents[0].strip())
			shift_time = str(tds[2].contents[0].encode("cp1252").strip())

			day		= parse_day(shift_time[0:3])
			start	  = parse_time(shift_time[5:10])
			end		= parse_time(shift_time[13:18])
			room	   = str(tds[3].a.get_text().encode("cp1252").strip()) if tds[3].find_all('a') else ""

			
			if shift_name != current_shift.name:   # new shift
				current_shift = Shift(shift_name)
				shifts.append(current_shift)

			current_shift.add_lesson_slot( \
				LessonSlot(day, start, end, room))


		blocks = {}
		for shift in shifts:
			category = extract_category(course.name, shift.name)
			if not category in blocks.keys():
				blocks[category] = LessonBlock(category)
			blocks[category].add_shift(shift)
		for block in blocks.values():
			course.add_lesson_block(block)

		course.name = course.lesson_blocks[0].shifts[0].name[:len(course.name)]
		course.long_name = soup.body.find('h2').contents[0].encode("utf-8").strip()

		return course


def extract_category(course_name, shift_name):
	return shift_name[len(course_name) : len(shift_name)-2]

def extract_name_from_url(url):
	splitted = url.split('/')
	return splitted[4]

def parse_day(day):
	dictionary = { 'Mon' : Weekday.MONDAY,
				   'Tue' : Weekday.TUESDAY,
				   'Wed' : Weekday.WEDNESDAY,
				   'Thu' : Weekday.THURSDAY,
				   'Fri' : Weekday.FRIDAY,
				   'Sat' : Weekday.SATURDAY,
				   'Seg' : Weekday.MONDAY,
				   'Ter' : Weekday.TUESDAY,
				   'Qua' : Weekday.WEDNESDAY,
				   'Qui' : Weekday.THURSDAY,
				   'Sex' : Weekday.FRIDAY,
				   'Sab' : Weekday.SATURDAY}
	return dictionary[day[:3]]

def parse_time(time):
	components = time.split(':')
	return Time(int(components[0]), int(components[1]))
	
def fix_url(url):
	parser = urlparse.urlparse(url)
	components = parser.path.split('/')[:5]
	fixed_path = string.join(components, '/') + "/turnos"
	return urlparse.urlunparse(parser[:2] + (fixed_path,) + parser[3:])