easychair2acm.py

#!/usr/bin/python3
# -*- coding: utf-8 -*-
""" Convert EasyChair data into ACM data for eRights generation

   Run: python3 easychair2acm.py <submitted_papers.csv> <authors.csv> <accepted_papers.csv>
"""

import sys

assert len(sys.argv) == 4

_papertype_mappings = {}

_papertype_mappings["experience-full"] = "Full Paper"
_papertype_mappings["experience-short"] = "Short Paper"

_papertype_mappings["research-full"] = "Full Paper"
_papertype_mappings["research-short"] = "Short Paper"

_papertype_mappings["tool-full"] = "Full Paper"
_papertype_mappings["tool-short"] = "Short Paper"


_id_papertype_mappings = {}

with open (sys.argv[1], mode='r', encoding='utf-8-sig') as submitted_file:
    for line in submitted_file: 
        # PAPER ID;PAPER TITLE;PAPER-TYPE
        record = line.rstrip().split(";")
        # PAPER ID -> PAPER-TYPE
        _id_papertype_mappings[record[0].strip()] = record[2].strip()

# id -> [Author1, Author2,..., AuthorN] (should be in order)             
_id_authornames_mappings = {}
# id -> [email1, email2,..., emailN] (should be in order)    
_id_authoremails_mappings = {}


class AuthorAffiliation:
    def __init__(self,first_name,last_name,affiliation):
        self.first_name = first_name
        self.last_name = last_name
        self.affiliation = affiliation
    
    def __str__(self):
        return '{} {}:{}'.format(self.first_name, self.last_name, self.affiliation)
    def __repr__(self):
        return '{} {}:{}'.format(self.first_name, self.last_name, self.affiliation)
    
# WARNING: This assumes the entries in the file are ordered (lead author first)
with open (sys.argv[2],  mode='r', encoding='utf-8-sig') as authors_file:
    for line in authors_file: 
        # PAPER ID;PAPER TITLE;FIRSTNAME;LASTNAME;EMAIL;AFFILIATION;COUNTRY_CODE
        record = line.rstrip().split(";")
        id = record[0].strip()
        first_name = record[2].strip()
        last_name = record[3].strip()
        email = record[4].strip()
        aff = record[5].strip()

        if id in _id_authoremails_mappings:
            _id_authornames_mappings[id].append(AuthorAffiliation(first_name,last_name,aff))
        else:
            authors = list()
            authors.append(AuthorAffiliation(first_name,last_name,aff))
            _id_authornames_mappings[id] = authors
        
        if id in _id_authoremails_mappings:
            _id_authoremails_mappings[id].append(email)
        else:
            emails = list()
            emails.append(email)
            _id_authoremails_mappings[id] = emails       


def getACMType(paper_id):
    type = _id_papertype_mappings[paper_id]
    if type in _papertype_mappings:
        return _papertype_mappings[type]
    else:
        return "!!!MISSING!!!"
    
acmcsv = open("acm-eRights.csv", "w")

with open (sys.argv[3],  mode='r', encoding='utf-8-sig') as accepted_file:
    for line in accepted_file:
        # PAPER ID;PAPER TITLE;STATUS
        record = line.rstrip().split(";")
        id = record[0].strip()
        title = record[1].strip()
        assert record[2] == "ACCEPT"
        authors = [str(author) for author in _id_authornames_mappings[id]]
        emails = [ '{}'.format(email) for email in _id_authoremails_mappings[id]]
        lead_email = emails.pop(0)
        assert len(emails) + 1 == len(authors)
        entry = '"{}","{}","{}","{}","{}"\n'.format(
            getACMType(id),
            title,
            ';'.join(authors),
            lead_email,
            ';'.join(emails)
            )
        acmcsv.write(entry)

acmcsv.close()