def timestamp_feature(catalog, soup):
"""The datetime the xml file was last modified.
"""
# there's really no "time created", we're using the date the courses are listed for...
epoch = 1318790434
catalog.timestamp = int(float(soup.title.text)) + epoch
catalog.datetime = datetime.datetime.fromtimestamp(catalog.timestamp)
logger.info('Catalog last updated on %s' % catalog.datetime)
def course_feature(catalog, soup):
"""Parses all the courses (AKA, the most important part).
"""
courses = {}
course_crns = {}
for course in soup.findAll('course'):
c = Course.from_soup_tag(course)
courses[str(c)] = c
catalog.courses = courses
catalog.courses
logger.info('Catalog has %d courses' % len(courses))
def course_feature(catalog, soup):
"""Parses all the courses (AKA, the most important part).
"""
courses = {}
count = 0
for course_data in parse_tables(soup):
c = create_course(course_data)
count += 1
courses[str(c)] = c
catalog.courses = FrozenDict(courses)
logger.info('Catalog has %d courses (manual: %d)' % (len(courses), count))
def semester_feature(catalog, soup):
"""The year and semester information that this xml file hold courses for.
"""
catalog.name = soup.find('h3').text.strip()
raw = soup.find('h3').text.split(' Session ')
catalog.year = int(raw[1])
month_mapping = {'Spring': 1, 'Summer': 5, 'Fall': 9}
catalog.semester = raw[0]
catalog.month = month_mapping[raw[0]]
logger.info('Catalog type: %s' % catalog.name)
def semester_feature(catalog, soup):
"""The year and semester information that this xml file hold courses for.
"""
raw = soup.coursedb['semesternumber']
catalog.year = int(raw[:4])
month_mapping = {1: 'Spring', 5: 'Summer', 9: 'Fall'}
catalog.month = int(raw[4:])
catalog.semester = month_mapping[catalog.month]
catalog.name = soup.coursedb['semesterdesc']
logger.info('Catalog type: %s' % catalog.name)
def crosslisting_feature(catalog, soup):
"""Parses all the crosslistings. These refer to the similar CRNs,
such as a grad & undergrad level course.
"""
listing = {}
for elem in soup.coursedb.findAll('crosslisting'):
seats = int(elem['seats'])
crns = [safeInt(crn.string) for crn in elem.findAll('crn')]
# we want to refer to the same object to save space
cl = CrossListing(crns, seats)
for crn in crns:
listing[crn] = cl
catalog.crosslistings = FrozenDict(listing)
logger.info('Catalog has %d course crosslistings' % len(catalog.crosslistings))
def semester_feature(catalog, soup):
"""The year and semester information that this xml file hold courses for.
"""
raw = _text(soup.findAll('h3')).split('\n')[1]
match = RE_SEMESTER_RANGE.match(raw)
catalog.year = int(match.group('year'))
#month_mapping = {'Spring': 1, 'Summer': 5, 'Fall': 9}
month_mapping = {'january': 1, 'may': 5, 'august': 9}
catalog.month = month_mapping[match.group('start_month').lower()]
if catalog.url:
match = RE_SEMESTER_URL.match(catalog.url)
if match:
catalog.year = int(match.group('year'))
catalog.month = int(match.group('month'))
semester_mapping = {1: 'Spring', 5: 'Summer', 9: 'Fall'}
catalog.semester = semester_mapping[catalog.month]
catalog.name = '%s %d' % (catalog.semester, catalog.year)
logger.info('Catalog type: %s' % catalog.name)
def timestamp_feature(catalog, soup):
"""The datetime the xml file was last modified.
"""
catalog.timestamp = int(soup.coursedb['timestamp'])
catalog.datetime = datetime.datetime.fromtimestamp(catalog.timestamp)
logger.info('Catalog last updated on %s' % catalog.datetime)