def get_citation_count(submission_csv, citation_dict):
number = re.search(".*paper(\d+).csv", submission_csv)
if not number:
raise ValueError("Couldn't get paper number in string: " +
submission_csv)
N = int(number.group(1))
citation_dict[N] = {}
for r in iterate_csv(submission_csv):
email, citcount = r
citation_dict[N][email] = int(citcount)
def read_expertise_db(expertise_csv, exp_list):
db = {}
for r in iterate_csv(expertise_csv):
db[r[3]] = {}
if len(exp_list) != len(r[4:]):
print(exp_list)
print(r[4:])
raise ValueError("Exp list and expertise db topics do not match "
"%d vs. %d" % (len(exp_list), len(r[4:])))
db[r[3]]['expertises'] = [l for l, v in zip(exp_list, r[4:])
if (v.strip()=='1'or v.strip()=='2')]
return db
def main():
report = sys.argv[1]
output_csv = sys.argv[2]
output_lines = ['paper,assignment,email']
for r in iterate_csv(report):
valid, paper, email, reasons = r
if valid == 'x':
output_lines.append("%s,conflict,%s" % (paper, email))
with open(output_csv, 'w') as f:
f.write('\n'.join(output_lines))
f.write('\n')
def main():
report = sys.argv[1]
output_csv = sys.argv[2]
types = sys.argv[3]
output_lines = ['paper,assignment,email,conflicttype']
if types == "susp":
for r in iterate_csv(report):
valid, paper, email, reasons, comment = r
if valid == 'x':
output_lines.append("%s,conflict,%s" % (paper, email))
else:
for r in iterate_csv(report):
valid, paper, email, reasons = r
if valid == 'x':
output_lines.append("%s,conflict,%s,chair-confirmed" %
(paper, email))
with open(output_csv, 'w') as f:
f.write('\n'.join(output_lines))
f.write('\n')
def read_expertise_to_topics(topics_to_expertise_csv):
exp_list = []
topic_list = set()
t_to_e = {}
e_to_t = {}
for r in iterate_csv(topics_to_expertise_csv):
e = r[0]
exp_list.append(e)
ts = [t for t in r[1:] if t]
e_to_t[e] = ts
for t in ts:
topic_list.add(t)
if t not in t_to_e:
t_to_e[t] = []
t_to_e[t].append(e)
return e_to_t, t_to_e, exp_list, topic_list
def main():
institutions_csv = sys.argv[1]
submissions_json = sys.argv[2]
hotcrp_pc_member_csv = sys.argv[3]
pc_member_paper_db_csv = sys.argv[4]
out_proper = sys.argv[5]
out_paper_collabs_field = sys.argv[6]
out_pc_collabs_field = sys.argv[7]
out_dblp = sys.argv[8]
out_fake = sys.argv[9]
# Step 1: read all the inputs (institutions_csv, paper data from hotcrp,
# pc info from hotcrp and paper db from dblp):
print("Reading institutions csv...")
institutions = Institutions(institutions_csv)
print("Reading submissions:")
d = get_dict_json(submissions_json)
submissions = [Submission.from_json(p, institutions) for p in tqdm(d)]
print("Reading hotcrp pc members:")
hotcrp_pc_members = [PCMember.from_hotcrp_csv(line, institutions)
for line in tqdm(iterate_csv(hotcrp_pc_member_csv,
encoding='utf-8'))]
hotcrp_pc_members = {p.email: p for p in hotcrp_pc_members}
print("Reading pc papers:")
dblp_pc_members = {k: p.copy_no_conflicts()
for k, p in hotcrp_pc_members.items()}
for row in tqdm(iterate_csv(pc_member_paper_db_csv)):
(email, id, firstname, lastname, keys, valid,
pub_key, pub_title, pub_year, pub_authors) = row
if valid == "x":
pub = Publication.from_key(pub_key, institutions)
dblp_pc_members[email].add_publication(pub)
print("Cross referencing conflicts")
for s in tqdm(submissions):
# Step 2: list conflicts that are declared by authors properly
# Step 3: list conflicts that are declared by authors
# in the collaborators field
for k, v in hotcrp_pc_members.items():
s.add_collaborator_conflict(v)
# Step 4: list conflicts declared by pc members but undeclared
# by paper authors
for k, v in hotcrp_pc_members.items():
s.add_conflicts_from_pc_member(v)
# Step 5: list conflicts not declared by anyone but caught by dblp
for k, v in dblp_pc_members.items():
s.add_conflicts_from_dblp(v)
# Step 6:
for k, v in dblp_pc_members.items():
s.add_fake_conflicts(hotcrp_pc_members[k], v)
print_reports(submissions, 'proper', out_proper)
print_reports(submissions, 'collaborators_field', out_paper_collabs_field)
print_reports(submissions, 'declared_by_pc_members', out_pc_collabs_field)
print_reports(submissions, 'dblp', out_dblp)
print_reports(submissions, 'fake_conflicts', out_fake)
def __init__(self, csv):
insts = {}
for i in iterate_csv(csv):
insts[i[0]] = i
self.insts = insts