Get the official campus emails for students in a Berkeley course

enrollment-emails

#!/usr/bin/python

# Example usage:
#   $0 -H headers.json -s compsci -c c8  
#   $0 -H headers.json -s stat    -c c100

import sys
import os
import json
import requests
import argparse

def get_descriptors(uri, params, headers):
	'''Get SIS descriptor IDs.'''
	r = requests.get(uri, params=params, headers=headers)
	data = r.json()
	return data['apiResponse']['response']['fieldValues']

def extract_lecture_codes(sections):
	return list(
		map(lambda x: x['code'],
			filter(lambda x: ' LEC ' in x['description'], sections)
		)
	)

def get_enrollments(uri, params, headers):
	if args.verbose:
		print("get_enrollments: {} {}".format(uri, params))
	r = requests.get(uri, params=params, headers=headers)

	if r.status_code == 404:
		if args.debug: print('No more enrollments')
		# there is no more data
		return []

	data = r.json()

	# Return if there is no response (e.g. 404)
	if 'response' not in data['apiResponse']:
		if args.debug: print('404 No response')
		print('404 No response')
		return[]

	# Return if the UID has no enrollments
	if 'classSectionEnrollments' not in data['apiResponse']['response']:
		if args.debug: print('No enrollments')
		print('No enrollments')
		return []

	enrollments = \
		data['apiResponse']['response']['classSectionEnrollments']

	params['page-number'] += 1
	enrollments += get_enrollments(uri, params, headers)

	if args.debug: print('enrollments batch: {}'.format(len(enrollments)))

	return enrollments

def enrollment_emails(enrollments):
	'''Extract unique student emails from enrollments.'''
	emails = set()
	for e in enrollments:
		for identifier in e['student']['identifiers']:
			for email in e['student']['emails']:
				if email['type']['code'] == 'CAMP':
					emails.add(email['emailAddress'])
					break
	return emails

## MAIN
# there's an api for this too
TERM_FALL_2016 = 2168
TERM_SPRING_2017 = 2172
TERM_SUMMER_2017 = 2178
TERM_FALL_2017 = 2178
SUBJECT_AREA = 'STAT'

# Headers file contains requests json. Example:
# { "Accept": "application/json",
#   "app_id": "YOUR_SIS_APP_ID",
#	"app_key": "YOUR_SIS_APP_KEY" }
HEADERS_FILE = 'fetch-enrollments.json'

parser = argparse.ArgumentParser(description="Print enrolled students' email addresses.")
parser.add_argument('-t', dest='term_id', type=int, default=TERM_FALL_2017,
	help='SIS Term ID (default=%(default)s)')
parser.add_argument('-s', dest='sa_filter', default=SUBJECT_AREA,
	help='subject area filter (default=%(default)s)')
parser.add_argument('-c', dest='course', default='',
	help='Course number, e.g. c8, 133, 135')
parser.add_argument('-H', dest='headers_file', default=HEADERS_FILE,
	help='JSON file containing requests headers')
parser.add_argument('-v', dest='verbose', action='store_true', help='Be verbose.')
parser.add_argument('-d', dest='debug', action='store_true', help='Debug.')
args = parser.parse_args()

## main

# api secrets
headers = json.loads(open(args.headers_file).read())

enrollments_uri = "https://apis.berkeley.edu/sis/v2/enrollments"
descriptors_uri = enrollments_uri + '/terms/{}/classes/sections/descriptors'
sections_uri = enrollments_uri + "/terms/{}/classes/sections/{}"

params = {
	"subject-area-code": args.sa_filter,
	"catalog-number": args.course,
}

# Retrieve the "section IDs" associated with the course.
# This gets both lecture and sections.
uri = descriptors_uri.format(args.term_id)
sections = get_descriptors(uri, params, headers)

# We only care about the lecture.
lecture_codes = extract_lecture_codes(sections)

# Get the enrollments
params['page-number'] = 1
params['page-size'] = 100
enrollments = []
for lecture_code in lecture_codes:
	uri = sections_uri.format(args.term_id, lecture_code)
	enrollments += get_enrollments(uri, params, headers)

# Get the emails
emails = enrollment_emails(enrollments)
print('\n'.join(e for e in emails))