def process_shuffle_csv(course, flags):
if flags.feedback:
what = 'feedback'
submit_assign = flags.feedback
else:
what = 'assessment'
submit_assign = flags.shuffle
submit_assign_id = course.find_assignment_id(submit_assign)
if submit_assign_id is None:
return
# get both current and former students, since someone may have uploaded feedback before dropping the course
submissions = course.fetch_assignment_submissions(submit_assign_id)
spreadsheets = []
submit_grades = {}
for sub in submissions:
if sub['workflow_state'] != 'submitted':
continue
uid = sub['user_id']
late_seconds = sub['seconds_late']
late_days = (late_seconds + 86100) // 86400
login = course.student_login(uid)
attachments = sub['attachments']
spreadsheet_url = None
upload_filename = None
filename = None
suffix = None
for attach in attachments:
if 'filename' not in attach or 'content-type' not in attach or 'url' not in attach:
continue
mimetype = attach['content-type']
if 'spreadsheet' in mimetype:
spreadsheet_url = attach['url']
filename = attach['filename']
(base, period, suffix) = filename.rpartition('.')
if spreadsheet_url:
destfile = '{}/{}_{}.{}'.format(flags.dir, login, what, suffix)
print('Downloading', what, 'by', login)
try:
urllib.request.urlretrieve(spreadsheet_url, filename=destfile)
except Exception as err:
print(err)
else:
spreadsheets += [destfile]
if flags.verbose:
print(spreadsheet_url, '->', destfile)
uid = course.get_id_for_student(login)
if uid is not None and uid > 0:
gr = Grade()
gr.add(10, '', 0, course.late_penalty_by_days(late_days))
submit_grades[uid] = gr
else:
print('-', login, 'did not upload', what, 'spreadsheet')
print('')
print('Uploading grades for submitting', what)
course.batch_upload_grades(submit_grades, 1, submit_assign_id)
print('')
csv_files = CanvasCSV.convert_files_to_csv(spreadsheets, flags.dir)
process_grades(course, flags, csv_files)
return
def build_feedback(partnum, total, mcq, coding, subscores):
subscores = [x for x in subscores if 'Not Applicable' not in str(x)]
comment = 'Part {}: '.format(partnum + 1) if partnum >= 0 else ''
if mcq is None:
comment += '{} points'.format(int(total))
else:
comment += '{}mcq+{}code'.format(Grade.drop_decimals(mcq),
Grade.drop_decimals(coding))
if sum(1 for x in subscores if x is not None) > 1:
subs = (Grade.drop_decimals(x) if x is not None else '-'
for x in subscores)
comment += '; per-Q: ' + ':'.join(subs)
return comment
def parse_hackerrank(course, csv, grades, partnum, verbose=False):
## extract the column numbers of interest from the first row
row = csv.next_row()
idx_date = csv.get_index('Date taken')
idx_email = csv.get_index('Login ID')
idx_andrew = csv.get_index('Andrew')
if idx_andrew == -1:
idx_andrew = csv.get_index('AndrewID')
if idx_andrew == -1:
idx_andrew = csv.get_index('Andrew ID')
idx_mcq = csv.get_index('MCQ')
idx_coding = csv.get_index('Coding')
idx_score = csv.get_index('Total score')
idx_date = csv.get_index('Date taken') # was the assignment late?
idx_qs = [csv.get_index('Question ' + str(num)) for num in range(1, 100)]
while not csv.eof:
row = csv.next_row()
if not row:
break
submit_date = csv.get_field(idx_date)
email = csv.get_field(idx_email)
andrew = csv.get_field(idx_andrew)
mcq = csv.get_field(idx_mcq)
coding = csv.get_field(idx_coding)
total = csv.get_field(idx_score)
subscores = [csv.get_field(i) for i in idx_qs if i >= 0]
uid = course.get_student_id(email, andrew)
if uid is None:
continue # non-existent or dropped student
comment = build_feedback(partnum, total, mcq, coding, subscores)
if uid in grades:
gr = grades[uid]
else:
gr = Grade()
grades[uid] = gr
if verbose:
print(' adding', uid, partnum, email, total, comment)
if total is not None:
pn = partnum if partnum >= 0 else 0
gr.add(total, comment, pn, course.late_penalty(submit_date))
return
def reassign(flags, remargs):
if flags.shuffle is None or flags.feedback is None:
print(
'You must specify both shuffle and feedback assignments with --shuffle and --feedback'
)
return
if not remargs:
print('No reassignments specified')
return
course = Course(HOST, COURSE_NAME, verbose=flags.verbose)
course.simulate(flags.dryrun)
course.mail_address(MAIL)
# get the assignment IDs for the interviewee assessment and interviewer feedback
assessment_id = course.find_assignment_id(flags.shuffle)
if assessment_id is None:
print('Did not find assignment', flags.shuffle, 'for shuffle')
return
feedback_id = course.find_assignment_id(flags.feedback)
if feedback_id is None:
print('Did not find assignment', flags.feedback, 'for feedback')
return
# retrieve the current peer associations ##FIXME
# peer_feedbacks = course.fetch_reviews(feedback_id,True)
# peer_assessments = course.fetch_reviews(assessment_id,True)
# process the requested reassignments
interviewers = {}
assessment_comments = {}
feedback_comments = {}
assessment_reviews = {}
feedback_reviews = {}
for pair in remargs:
interviewer, _, interviewee = pair.partition(':')
interviewer_uid = course.get_student_id(interviewer)
if interviewer_uid is None or interviewer_uid < 0:
print('Interviewer "{}" not found'.format(interviewer))
continue
interviewee_uid = course.get_student_id(interviewee)
if interviewee_uid is None or interviewee_uid < 0:
print('Interviewee "{}" not found'.format(interviewer))
continue
interviewers[interviewer_uid] = interviewer
# build the link to the peer review page for the interviewer
fb_link = course.peer_review_user_link(feedback_id, interviewer_uid)
# store the new assignments
assessment_reviews[interviewee] = interviewer
feedback_reviews[interviewer] = interviewee
# add comments informing the students of the new assignments
new_interviewer_comment = ('You have been assigned a new interviewer: {}\n'+ \
'The feedback peer review link is: {}\n').format(interviewer,fb_link)
if interviewee_uid not in assessment_comments:
assessment_comments[interviewee_uid] = Grade()
assessment_comments[interviewee_uid].add(0, new_interviewer_comment)
new_interviewee_comment = 'You have been re-assigned to interview: {}'.format(
interviewee)
if interviewer_uid not in assessment_comments:
assessment_comments[interviewer_uid] = Grade()
assessment_comments[interviewer_uid].add(0, new_interviewee_comment)
feedback_comments[interviewee_uid] = Grade(0,('Your interview has been re-assigned.\n'+ \
'Your new interviewer is: {}\n'+ \
'The feedback link is: {}\n').format(interviewer,fb_link))
if False:
# remove existing peer-review assignments for interviewees from the Canvas assignment named by
# the --shuffle flag, then upload the new peer-review assignments
print("removing peer reviewers in assignment", flags.shuffle)
course.remove_peer_reviewers(feedback_comments.keys(), assessment_id)
course.add_peer_reviewers(assessment_reviews, assessment_id)
# upload a comment for each student informing them of their new interviewee
course.batch_upload_grades(assessment_comments, 0, assessment_id)
# remove existing peer reviews for interviewer from the Canvas assignment named by the --feedback flag
# then upload the new peer-review assignments
print("removing peer reviewers in assignment", flags.feedback)
course.remove_peer_reviewers(interviewers.keys(), feedback_id)
course.add_peer_reviewers(feedback_reviews, feedback_id)
# upload a comment for each student informing them of their new interviewer
course.batch_upload_grades(feedback_comments, 0, feedback_id)
return
def make_shuffle(course, flags):
if not flags.feedback:
print('You must specify the feedback assignment with --feedback')
return
feedback_assign_id = course.find_assignment_id(flags.feedback)
if feedback_assign_id is None:
print('The requested feedback assignment was not found')
return
questions = flags.questions
if questions is None:
questions = 'problems.txt'
with open(questions) as f:
q = f.read().split('\n')
students = flags.students
if students is None:
student_ids = course.fetch_active_students()
interviewers = [email_to_AndrewID(login) for login in student_ids]
else:
if students == '.' or students == '=':
students = 'students.txt'
with open(students) as f:
interviewers = f.read().split('\n')
interviews = assign_interviews(interviewers, q)
# create output directory
directory = flags.dir
if directory is None or directory == '' or directory == '.':
directory = 'shuffle'
if not os.path.exists(directory):
os.makedirs(directory)
log = open('assignments.log', 'a')
log.write('Shuffle assignments for {}/{}\n'.format(flags.assignment,
flags.feedback))
assessment_reviews = {
} # map from student to their interviewer, who will leave a peer review for them
feedback_reviews = {
} # map from student to their interviewee, who will leave a peer review for them
grades = {}
feedback_links = {}
for (interviewer, iview) in interviews.items():
log.write('{} -> {} Q1: {} Q2: {} Q3: {}\n'.format(
interviewer, iview[0], iview[2], iview[3], iview[4]))
if flags.verbose:
print(interviewer, '->', iview[0], ' Q1:', iview[2], ' Q2:',
iview[3], ' Q3:', iview[4])
interviewee = iview[0]
rev_interviewer = iview[1]
assessment_reviews[interviewee] = interviewer
feedback_reviews[interviewer] = interviewee
uid = course.get_id_for_student(interviewer)
interviewer_uid = course.get_id_for_student(rev_interviewer)
fb_link = course.peer_review_user_link(feedback_assign_id,
interviewer_uid)
if uid is not None and uid > 0:
interviewee_uid = course.get_id_for_student(interviewee)
assess_link = course.peer_review_user_link(course.assignment_id,
interviewee_uid)
grades[uid] = Grade(
0, ('You will interview: {}\n' +
'The questions to ask: {} {} {}\n' +
'You will be interviewed by: {}\n\n' +
'The feedback peer review link is: {}').format(
interviewee, iview[2], iview[3], iview[4],
rev_interviewer, fb_link))
#FIXME
# grades[uid] = Grade(0,('You will interview: {}\n'+
# 'The questions to ask: {} {} {}\n'+
# 'You will be interviewed by: {}\n\n'+
# 'The assessment peer review link is: {}')
# 'The feedback peer review link is: {}')
# .format(interviewee,iview[2],iview[3],iview[4],rev_interviewer,assess_link,fb_link))
if len(fb_link) > 0:
feedback_links[uid] = Grade(0,
('You will be interviewed by: {}\n' +
'The feedback link is: {}\n').format(
rev_interviewer, fb_link))
log.close()
# upload peer-review assignments to the Canvas assignment named by the -a flag
#FIXME course.add_peer_reviewers(assessment_reviews,course.assignment_id)
# upload a comment for each student informing them of their interviewee/questions/interviewer
course.batch_upload_grades(grades, 0)
# upload peer-review assignments to the Canvas assignment named by the --feedback flag
course.add_peer_reviewers(feedback_reviews, feedback_assign_id)
# upload a comment for each student informing them of their interviewer and providing the link to the peer review
course.batch_upload_grades(feedback_links, 0, feedback_assign_id)
return
def process_shuffle_assessment(submissions,
rubric_def,
rubric_grades,
submit_grades,
assessors,
submit_points,
course,
verbose=False):
if verbose:
print(len(submissions), 'total submissions retrieved')
for sub in submissions:
if sub['id'] not in assessors:
continue
pts = submit_points
remarks = ''
attachments = []
if 'submission_comments' in sub:
for com in sub['submission_comments']:
if 'attachments' in com and com['attachments'] and com[
'author']['display_name'] not in COURSE_STAFF:
for a in com['attachments']:
ctype = 'unknown'
if 'content-type' in a:
ctype = a['content-type']
attachments += (a['filename'], ctype, a['mime_class'],
a['url'])
if verbose:
print('Attachments:', attachments)
uid = sub['user_id']
reviewer, _, criteria = assessors[sub['id']]
crit_points = {}
for crit in criteria:
if 'points' not in crit or 'criterion_id' not in crit or crit[
'points'] < 0.0:
continue # N/A or non-scored criterion
points = crit['points']
c_id = crit['criterion_id']
name = rubric_def.get_name(c_id)
if not name:
continue
if name[0] == 'Q' and name[2] == ':':
name = name[4:]
if 'Suggestions' in name:
if len(crit['comments']) < 8:
pts = pts - (0.05 * submit_points)
remarks += 'Did not provide suggested improvements (-{})\n'.format(
Grade.drop_decimals(0.05 * submit_points))
continue
if 'Location' in name:
if len(crit['comments']) < 8:
pts = pts - (0.05 * submit_points)
remarks += 'Did not specify location/time (-{})\n'.format(
Grade.drop_decimals(0.05 * submit_points))
continue
if name in crit_points:
crit_points[name] = crit_points[name] + [points]
else:
crit_points[name] = [points]
total_points = 0
possible = 0
if verbose:
print('crit_points:', crit_points)
for itemname, pointlist in crit_points.items():
total_points += mean(pointlist)
possible += 3
if total_points == 0:
print('! {} ({}) received a zero score'.format(
course.student_login(uid), uid))
if possible > 0:
grade = 12.0 * (total_points / possible)
rubric_grades[uid] = 12.0 * (total_points / possible)
else:
grade = None
print('empty rubric for {} ({})'.format(course.student_login(uid),
uid))
if reviewer is not None:
if attachments == []: ##FIXME
pts = pts - (0.30 * submit_points)
remarks += 'Did not upload photo (-{})'.format(
Grade.drop_decimals(0.30 * submit_points))
submit_grades[reviewer] = Grade(pts, remarks)
if verbose or True: ##DEBUG
print(' ', course.student_login(reviewer), 'entered', grade,
'for', course.student_login(uid))
return
def parse_shuffle_feedback(course, csv, filename, grades, verbose=False):
if '_feedback' in filename:
user = filename.rsplit('_feedback', -1)[0]
else:
user = None
## extract the column numbers of interest from the first row
row = csv.next_row()
if "Interviewee:" in row or "Location" in row[3]:
print('*', extract_andrew_from_filename(filename),
"submitted an Interviewee Assessment")
return
# skip the second row, which contains user instructions
row = csv.next_row()
if 'Andrew' in row:
# file format error? Possibly shifted contents of spreadsheet
print(filename, "is an unrecognized file")
return
# third row contains interviewer's AndrewID and overall criterion scores + final average
row = csv.next_row()
andrew = row[0]
# massage Andrew ID
andrew = email_to_AndrewID(andrew)
overall_score = row[14]
# skip empty row
csv.next_row()
# skip header line
csv.next_row()
crit = [0] * 4
for i in range(4):
row = csv.next_row()
if row[0] == '':
print('!', filename, ' contains blank entry in rubric')
crit[i] = 0
else:
try:
crit[i] = float(row[0])
except:
print('*', filename, ' contains invalid entry in rubric')
return
# skip next header line
csv.next_row()
# read comments line
row = csv.next_row()
comment = '(no comment given)' if row is None else ''.join(row)
## validity checks
if andrew is None or andrew == "":
print('*', filename, " does not contain an Interviewer ID")
return
if andrew == user:
print("= user", andrew, "entered own ID as interviewer")
return
uid = course.get_id_for_student(andrew)
if uid is None:
print('*', filename, "contains an unknown Andrew ID:", andrew)
return
if uid in grades:
print('+', filename, "contains an Andrew ID we've already seen:",
andrew)
return
## OK, we've got a good feedback form, so massage the data
## we're changing the score in Canvas to be 0-12 to be compatible with an upcoming switch to rubric scoring,
## so just add up the subscores
overall_score = sum(crit)
feedback = 'Subscores: {}/{}/{}/{}, Feedback: {}'.format(
crit[0], crit[1], crit[2], crit[3], comment)
if verbose:
print(" adding", uid, andrew, overall_score, feedback)
grades[uid] = Grade(overall_score, feedback)
return
def parse_shuffle_assessment(course, csv, filename, grades, verbose=False):
# read header line
row = csv.next_row()
if "Feedback" in row[0]:
print('*', extract_andrew_from_filename(filename, 'assessment'),
"submitted an Interviewer Feedback")
return
# skip second header line
csv.next_row()
# read AndrewIDs and Q1/Q2/Q3/Overall scores
row = csv.next_row()
andrew = email_to_AndrewID(row[1])
##FIXME: massage AndrewID
try:
totalscore = float(row[14])
except:
print('Bad totalscore in', filename)
totalscore = -1
had_error = True
# skip empty line
csv.next_row()
# skip header line
csv.next_row()
# skip second header line
csv.next_row()
q1 = []
q2 = []
q3 = []
had_error = False
for i in range(7):
# get rubric score i for the three questions
row = csv.next_row()
if row is None:
print('Bad data in', filename)
had_error = True
q1 += ['?']
q2 += ['?']
q3 += ['?']
else:
q1 += [row[0]]
q2 += [row[1]]
q3 += [row[2]]
# skip empty line
csv.next_row()
# skip header line
csv.next_row()
# skip second header line
csv.next_row()
overall = []
for i in range(5):
row = csv.next_row()
if row is None:
had_error = True
else:
overall += [row[0]]
# skip empty line
csv.next_row()
# skip header line
csv.next_row()
# read feedback text
row = csv.next_row()
if row is None:
had_error = True
feedback = ''
else:
feedback = row[3]
try:
# strip decimal part of total score
base_score = int(totalscore)
except:
base_score = 0
if base_score == 0:
print('-', filename, 'gives zero score')
# validity checks
if andrew is None or andrew == '':
print('*', filename, "does not include interviewee's Andrew ID")
return
uid = course.get_id_for_student(andrew)
if uid is None:
print('*', filename, 'contains an unknown Andrew ID:', andrew)
return
if uid in grades:
print('=', filename, "contains an Andrew ID we've already seen:",
andrew)
return
# reformat the information into a comment for the gradebook
comment = 'Q1: {}/{}/{}/{}/{}/{}/{}'.format(q1[0], q1[1], q1[2], q1[3],
q1[4], q1[5], q1[6])
if ''.join(q2) != '':
comment += '\nQ2: {}/{}/{}/{}/{}/{}/{}'.format(q2[0], q2[1], q2[2],
q2[3], q2[4], q2[5],
q2[6])
if ''.join(q3) != '':
comment += '\nQ3: {}/{}/{}/{}/{}/{}/{}'.format(q3[0], q3[1], q3[2],
q3[3], q3[4], q3[5],
q3[6])
comment += '\nOverall: {}/{}/{}/{}/{}'.format(overall[0], overall[1],
overall[2], overall[3],
overall[4])
comment += '\nFeedback: {}'.format(feedback)
## insert score and comment into 'grades'
if had_error:
print('Bad data in', filename, ', best guess is:')
print(' ', andrew, totalscore, comment)
if verbose:
print(" adding", uid, andrew, totalscore, comment)
grades[uid] = Grade(totalscore, comment)
return