#!/usr/bin/env python

# Canonicalizes a spreadsheet column to a vocabulary.

# The column header is also canonicalized. CSVs without a header are supported.

# Unrecognized names are left untouched, permitting successive runs on different

# vocabularies.

# Raises an error if two input terms map to the same simplified string.

# Case- and punctuation-insensitive.

import csv

import re

import sys

class OnceOnlyDict(dict):

    '''A dict that only allows each key to be assigned once (no overwriting).'''

    def __setitem__(self, key, value):

        if key in self: raise KeyError(key)

        dict.__setitem__(self, key, value)

def simplify(str_): return re.sub(r'[^a-zA-Z0-9]+', r'', str_.lower())

def main():

    try: _prog_name, col_num, vocab_path = sys.argv

    except ValueError: raise SystemExit('Usage: '+sys.argv[0]

        +' <in col# vocab [| '+sys.argv[0]+' col# vocab_2]... >out')

    col_num = int(col_num)

    # Get vocab

    dict_ = OnceOnlyDict()

    stream = open(vocab_path, 'rb')

    reader = csv.reader(stream)

    for row in reader: dict_[simplify(row[0])] = row[0]

    stream.close()

    # Canonicalize input

    reader = csv.reader(sys.stdin)

    writer = csv.writer(sys.stdout)

    for row in reader:

        term = simplify(row[col_num])

        try: row[col_num] = dict_[term]

        except KeyError: pass

        writer.writerow(row)

main()