#!/usr/bin/env python

# Intersects two map spreadsheets A0->B and A1->C to A->B, with B overwriting C

# Multi-safe (supports an input appearing multiple times).

# Case- and punctuation-insensitive.

import csv

import os.path

import sys

sys.path.append(os.path.dirname(__file__)+"/../lib")

import maps

import util

def main():

    try: _prog_name, map_1_path = sys.argv[:2]

    except ValueError:

        raise SystemExit('Usage: '+sys.argv[0]+' <map_0 map_1 '

            '[compare_col_num...] [| '+sys.argv[0]

            +' map_2]... >intersect_map')

    compare_col_nums = map(int, sys.argv[2:]) # 0-based

    if compare_col_nums == []:

        compare_col_nums = None # list_subset() value for all columns

    def compare_on(row): return tuple(map(maps.simplify,

        util.list_subset(row, compare_col_nums)))

    headers = [None]*2

    # Get map 1

    compare_cols = set()

    stream = open(map_1_path, 'rb')

    reader = csv.reader(stream)

    headers[1] = reader.next()

    for row in reader:

        if row[0] != '':

            compare_cols.add(compare_on(row))

    stream.close()

    # Open map 0

    reader = csv.reader(sys.stdin)

    headers[0] = reader.next()

    # Check col labels

    combinable = maps.combinable(*headers)

    if not combinable:

        raise SystemExit('Map error: '

        'Map 0 column 0 label doesn\'t contain map 1 column 0 label')

    # Add map 0 to map 1, overwriting existing entries

    writer = csv.writer(sys.stdout)

    writer.writerow(headers[0])

    for row in reader:

        if not combinable or compare_on(row) in compare_cols:

            # not combinable, or in map 1

            writer.writerow(row)

main()