Checkout student submissions from Github Classroom

checkout.py

"""Check out student submissions for an assignment"""

# driven by two csv files:
#  - export from GitHub Classroom roster containing
#       identifier, github_username, github_id, name
#  - export from iLearn containing
#       'Email address', 'ID number'
#
# input classroom name and assignment name
# then checks out one repository per student into a directory
# named for the 'ID number'
# reports any missing or extra students
#

import csv
import subprocess
import os


def read_github_roster(csvfile):
    """Read the roster exported from github, return
    a dictionary with email addr as key and github username
    as the value"""

    result = {}
    with open(csvfile) as fd:
        reader = csv.DictReader(fd)
        for line in reader:
            if 'identifier' in line and not line['github_username'] == "":
                result[line['identifier']] = line['github_username']

    return result


def read_ilearn_export(csvfile):
    """Read the file exported from iLearn, return a
    dictionary with email address as the key and
    student id as the values"""

    result = {}
    with open(csvfile) as fd:
        reader = csv.DictReader(fd)
        for line in reader:
            if 'Email address' in line:
                result[line['Email address']] = line['ID number']

    return result


def merge_students(github, ilearn):
    """Generate one dictionary with info from both
    rosters, also return a dictionary with people not
    in both"""

    emails = set(github.keys())
    emails = set(emails.union(set(ilearn.keys())))

    roster = {}
    missing = {}
    for email in emails:
        if email in github and email in ilearn:
            roster[email] = {'github': github[email], 'ilearn': ilearn[email]}
        elif email in github:
            missing[email] = {'github': github[email]}
        else:
            missing[email] = {'ilearn': ilearn[email]}

    return roster, missing


#
# 		name=${assignment}-${line}
# 		git clone ${classroom}${name}
def checkout(config, githubid, studentid):
    """Checkout one student assignment into a directory
    named for the student id
    """

    if not os.path.exists(config['outdir']):
        os.makedirs(config['outdir'])

    repo = "%s/%s-%s" % (config['classroom'], config['assignment'], githubid)
    cmd = ['git', 'clone', repo, os.path.join(config['outdir'], studentid)]
    print(" ".join(cmd))
    p1 = subprocess.Popen(cmd, stdout=subprocess.PIPE)
    return p1.communicate()


def pull(config, githubid, studentid):
    """make sure we have the latest version of this
    repository, checkout if needed or just pull if
    we already have a version"""

    repodir = os.path.join(config['outdir'], studentid)
    if os.path.exists(repodir):
        wd = os.getcwd()
        try:
            os.chdir(repodir)
            cmd = ['git', 'pull']
            p1 = subprocess.Popen(cmd, stdout=subprocess.PIPE)
        finally:
            os.chdir(wd)
        return p1.communicate()
    else:
        return checkout(config, githubid, studentid)


if __name__=='__main__':

    import sys

    github = read_github_roster(sys.argv[1])
    ilearn = read_ilearn_export(sys.argv[2])

    students, missing = merge_students(github, ilearn)

    config = {
        'classroom': "git@github.com:MQCOMP257",
        'assignment': "data-science-portfolio",
        'outdir': 'submissions',
    }

    for key in students:
        checkout(config, students[key]['github'], students[key]['ilearn'])
        #pull(config, students[key]['github'], students[key]['ilearn'])

    #print("-------MISSING---------")
    #for key in missing:
    #    print(key, missing[key])