#! /usr/bin/env python2.7
"""
TeX authors affiliations aggregation script

Script accepts input files in the following format:

    Oleg ; EPFL
    Andrew ; Leiden
    Andrew ; Kiev
    Alexey ; Leiden
    Alexey ; EPFL
    Alexey ; MIT

And produces the output in stdout:

    \author[1]{Oleg}
    \author[2,3]{Andrew}
    \author[2,1,4]{Alexey}

    \affiliation[1]{EPFL}
    \affiliation[2]{Leiden}
    \affiliation[3]{Kiev}
    \affiliation[4]{MIT}

Empty lines are skipped, malformed lines are reported without interruption.
"""

__author__ = "Andrii Magalich"
__version__ = "1.0"
__date__ = "17th Dec 2014"

if __name__ == '__main__':

    import sys
    from collections import defaultdict

    if len(sys.argv) == 1:
        input_file = "authors.csv"
    else:
        input_file = sys.argv[1]

    output_file = "ship_authors.tex"
    
    input = open(input_file, 'r').read()
    lines = input.split("\n")

    authors_order = []
    authors = defaultdict(lambda: [])
    affiliations = []

    for line in lines:
        if not line:
            continue

        try:
            author, affiliation = line.split(";")
        except ValueError:
            print("////Error processing line:\n    {}".format(line))
        else:
            author = author.strip()
            affiliation = affiliation.strip()
            if not affiliation in affiliations:
                affiliations.append(affiliation)
            if not author in authors_order:
                authors_order.append(author)
            authors[author].append(str(affiliations.index(affiliation)+1))

    print("-"*80)

    for author in authors_order:
        print("\\author[{}]{{{},}}".format(",".join(authors[author]), author))

    print("")

    for i, affiliation in enumerate(affiliations):
        print("\\affiliation[{}]{{{}}}".format(i+1, affiliation))