def merge_course_if_fullyear(self):
if self.course.num[-1] == "A":
other_half_letter = "B"
elif self.course.num[-1] == "B":
other_half_letter = "A"
else:
#Not a full year course
return
other_half_key = self.course.get_key()[:-1] + other_half_letter
if other_half_key not in self.courses_dict:
#Haven't scraped the other course yet
print "Haven't found other half yet."
return
other_half = self.courses_dict[other_half_key]
print "Merging with other half: %s" % (other_half.get_key())
#Remove the other course from the dict so we can re-add the full course
del self.courses_dict[other_half_key]
del self.courses_dict[self.course.get_key()]
merged = SolusModels.SolusCourse()
merged.add_merged_info(other_half, self.course)
self.add_course(merged)
def read_from_file(self):
with open(self.read_file_name) as f:
all_str = f.read()
all_list = json.loads(all_str)
for course_dict in all_list:
course = SolusModels.SolusCourse(course_dict)
self.courses_dict[course.get_key()] = course
print len(self.courses_dict)
def scrape_single_section_component(self, piece_array):
if len(piece_array) < 6:
import pdb; pdb.set_trace()
components = []
#Date range
m = re.search('^([\S]+)\s*-\s*([\S]+)$', piece_array.pop())
start_date = m.group(1)
end_date = m.group(2)
instructor = piece_array.pop()
room = piece_array.pop()
#Timeslot
end = piece_array.pop()
start = piece_array.pop()
#Sometimes day is e.g. "MoTuWeSaSu"
all_days = piece_array.pop()
if all_days == "TBA":
all_days = "TB"
while len(all_days) > 0:
day = SolusModels.index_of_day_abbr(all_days[-2:])
all_days = all_days[:-2]
section_component = SolusModels.SectionComponent()
components.append(section_component)
section_component.start_date = start_date
section_component.end_date = end_date
section_component.instructor = instructor
section_component.room = room
section_component.timeslot = SolusModels.timeslot_index_by_components(day, start, end)
return components
def scrape_section_header(self, piece_array, section):
section_info = piece_array.pop()
m = re.search('^([\S]+)-([\S]+)\s+\((\S+)\)$', section_info)
while not m:
section_info = piece_array.pop()
m = re.search('^([\S]+)-([\S]+)\s+\((\S+)\)$', section_info)
section.index = m.group(1)
section.type = SolusModels.section_type_index_by_key(m.group(2))
section.id = m.group(3)
def scrape_section_header(self, piece_array, section):
section_info = piece_array.pop()
m = re.search('^([\S]+)-([\S]+)\s+\((\S+)\)$', section_info)
while not m:
section_info = piece_array.pop()
m = re.search('^([\S]+)-([\S]+)\s+\((\S+)\)$', section_info)
section.index = m.group(1)
section.type = SolusModels.section_type_index_by_key(m.group(2))
section.id = m.group(3)
def scrape_section_page(self):
section_pieces = self.section_pieces_from_page()
while len(section_pieces) > 0:
section = SolusModels.Section()
self.course.sections.append(section)
section.term = self.current_term
self.scrape_single_section(section_pieces, section)
def scrape_single_section_component(self, piece_array):
if len(piece_array) < 6:
import pdb
pdb.set_trace()
components = []
#Date range
m = re.search('^([\S]+)\s*-\s*([\S]+)$', piece_array.pop())
start_date = m.group(1)
end_date = m.group(2)
instructor = piece_array.pop()
room = piece_array.pop()
#Timeslot
end = piece_array.pop()
start = piece_array.pop()
#Sometimes day is e.g. "MoTuWeSaSu"
all_days = piece_array.pop()
if all_days == "TBA":
all_days = "TB"
while len(all_days) > 0:
day = SolusModels.index_of_day_abbr(all_days[-2:])
all_days = all_days[:-2]
section_component = SolusModels.SectionComponent()
components.append(section_component)
section_component.start_date = start_date
section_component.end_date = end_date
section_component.instructor = instructor
section_component.room = room
section_component.timeslot = SolusModels.timeslot_index_by_components(
day, start, end)
return components
def scrape_sections(self):
sel = self.selenium
term_options = sel.get_select_options("id=DERIVED_SAA_CRS_TERM_ALT")
for option in term_options:
if not len(term_options) == 1:
sel.select("id=DERIVED_SAA_CRS_TERM_ALT", "label=%s" % option)
sel.click("id=DERIVED_SAA_CRS_SSR_PB_GO$92$")
sel.wait_for_page_to_load(self.timeout_milliseconds)
self.current_term = SolusModels.term_index_by_key(option)
self.scrape_term()
def scrape_sections(self):
sel = self.selenium
term_options = sel.get_select_options("id=DERIVED_SAA_CRS_TERM_ALT")
for option in term_options:
if not len(term_options) == 1:
sel.select("id=DERIVED_SAA_CRS_TERM_ALT", "label=%s" % option)
sel.click("id=DERIVED_SAA_CRS_SSR_PB_GO$92$")
sel.wait_for_page_to_load(self.timeout_milliseconds)
self.current_term = SolusModels.term_index_by_key(option)
self.scrape_term()
def scrape_subjects_for_alphanum(self, alphanum):
sel = self.selenium
sel.click("id=DERIVED_SSS_BCC_SSR_ALPHANUM_" + alphanum)
sel.wait_for_page_to_load(self.timeout_milliseconds)
#Prepare to traverse all links
link_number = self.starting_subject_index
link_name_base = "name=DERIVED_SSS_BCC_GROUP_BOX_1$84$$%d"
link_name = link_name_base % (link_number,)
while sel.is_element_present(link_name):
#Store subject title
m = re.search("^([^-]*) - (.*)$", sel.get_text(link_name).strip())
subject_key = m.group(1)
subject_title = m.group(2)
print "\nSubject: %s: %s" % (subject_key, subject_title)
if subject_key not in self.ignored_subjects:
self.subject_index = SolusModels.subject_index_by_key(subject_key)
SolusModels.Subject.subjects[self.subject_index].title = subject_title
#Open the dropdown
sel.click(link_name)
sel.wait_for_page_to_load(self.timeout_milliseconds)
#Traverses all course links in the dropdown
self.scrape_single_dropdown()
#Close the dropdown
try:
sel.click(link_name)
except:
print "FAILURE %s" % link_name
time.sleep(100)
sel.wait_for_page_to_load(self.timeout_milliseconds)
else:
print "Ignored"
#Go to next link
link_number += 1
if self.max_subjects_per_letter and link_number >= self.max_subjects_per_letter + self.starting_subject_index:
break
link_name = link_name_base % (link_number,)
def scrape_subjects_for_alphanum(self, alphanum):
sel = self.selenium
sel.click("id=DERIVED_SSS_BCC_SSR_ALPHANUM_" + alphanum)
sel.wait_for_page_to_load(self.timeout_milliseconds)
#Prepare to traverse all links
link_number = self.starting_subject_index
link_name_base = "name=DERIVED_SSS_BCC_GROUP_BOX_1$84$$%d"
link_name = link_name_base % (link_number, )
while sel.is_element_present(link_name):
#Store subject title
m = re.search("^([^-]*) - (.*)$", sel.get_text(link_name).strip())
subject_key = m.group(1)
subject_title = m.group(2)
print "\nSubject: %s: %s" % (subject_key, subject_title)
if subject_key not in self.ignored_subjects:
self.subject_index = SolusModels.subject_index_by_key(
subject_key)
SolusModels.Subject.subjects[
self.subject_index].title = subject_title
#Open the dropdown
sel.click(link_name)
sel.wait_for_page_to_load(self.timeout_milliseconds)
#Traverses all course links in the dropdown
self.scrape_single_dropdown()
#Close the dropdown
try:
sel.click(link_name)
except:
print "FAILURE %s" % link_name
time.sleep(100)
sel.wait_for_page_to_load(self.timeout_milliseconds)
else:
print "Ignored"
#Go to next link
link_number += 1
if self.max_subjects_per_letter and link_number >= self.max_subjects_per_letter + self.starting_subject_index:
break
link_name = link_name_base % (link_number, )
def scrape_single_dropdown(self):
sel = self.selenium
#Prepare to traverse all links
link_number = self.starting_course_index
link_name_base = "id=CRSE_TITLE$%d"
link_name = link_name_base % (link_number, )
while sel.is_element_present(link_name):
#Go into the course
sel.click(link_name)
sel.wait_for_page_to_load(self.timeout_milliseconds)
self.course = SolusModels.SolusCourse()
SolusModels.SolusCourse.num_courses += 1
self.course.subject = self.subject_index
#Scrape info from course
try:
self.scrape_single_course()
self.course.clean()
self.add_course(self.course)
self.merge_course_if_fullyear()
except SolusModels.UselessCourseException as e:
print "Ignored"
SolusModels.SolusCourse.num_courses -= 1
#Back out from course page
sel.click("id=DERIVED_SAA_CRS_RETURN_PB")
sel.wait_for_page_to_load(self.timeout_milliseconds)
#Go to next course
link_number += 1
if self.max_courses_per_subject and link_number >= self.max_courses_per_subject + self.starting_course_index:
break
link_name = link_name_base % (link_number, )
def scrape_title(self):
sel = self.selenium
raw_title = sel.get_text("css=span.PALEVEL0SECONDARY").strip()
m = re.search('^([\S]+)\s+([\S]+)\s+-\s+(.*)$', raw_title)
#Subject is assigned earlier
#self.course.subject = SolusModels.subject_index_by_key(m.group(1))
self.course.subject_description = m.group(1)
self.course.num = m.group(2)
self.course.title = m.group(3)
print ""
print "%s/%s %s - %s" % (self.course.subject_description,
self.course.subject, self.course.num,
self.course.title)
if re.search('^(UNSP)|(.*UNS)$', self.course.num):
raise SolusModels.UselessCourseException(
"%s %s" % (self.course.subject, self.course.num))
