#!/usr/bin/env python

# Modifies a map spreadsheet A->B or any file using a replacements spreadsheet

# A->C or B->C

import csv

import HTMLParser

import os.path

import re

import sys

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

import maps

import opts

import strings

quote_re = '[\'"`]'

excluded_prefix_re = (

     '(?<![a-z] )' #not if it's a word in a sentence (uppercase SQL keywords OK)

    +'(?<!-)' # not if it's part of a '-'-separated identifier

    +'(?<!\*)' # don't double leading *

    )

excluded_suffix_re = (

     '(?! [a-z])' # not if it's a word in a sentence (uppercase SQL keywords OK)

    +'(?!-)' # not if it's part of a '-'-separated identifier

    )

def unescape_html(str_): return HTMLParser.HTMLParser().unescape(str_)

def repl_unescape_html(match): return unescape_html(match.group(0))

def main():

    env_names = []

    def usage_err():

        raise SystemExit('Usage: '+opts.env_usage(env_names, True)+' '

            +sys.argv[0]+' <map repl [col_num] [| '+sys.argv[0]

            +' repl_1 [col_num_1]]... >new_map')

    col_num = None

    try: col_num = sys.argv[2]

    except IndexError: pass

    if col_num != None: col_num = int(col_num) # 0-based

    # whether all patterns are plain text

    # defaults to on if matching entire cells in a spreadsheet (w/ col_num)

    text = opts.env_flag('text', col_num != None, env_names)

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

    except ValueError: usage_err()

    # Get replacements

    repls = []

    stream = open(repl_path, 'rb')

    reader = csv.reader(stream)

    reader.next() # skip header

    for row in reader:

        in_, out = row[:2]

        if in_ != '':

            is_word = re.match(r'^\w+$', in_)

            if text or is_word: # match as whole-word text (like SQL identifier)

                # this can also be done in Postgres with expression substitution

                # (wiki.vegpath.org/Postgres_queries#expression-substitution)

                in_str_re = re.escape(in_)

                q = quote_re

                in_ = '(?<='+q+')'+in_str_re+'(?='+q+')' # require quotes

                if is_word: # also match with quotes optional

                    # don't try to match word w/ suffix, because there are cases

                    # where a mapping adds a suffix which would cause the same

                    # replacement to be performed repeatedly

                    in_word_re = r'\b'+in_str_re+r'\b'

                    # only use excluded_prefix_re/excluded_suffix_re in text

                    # mode (used in renaming columns in SQL scripts), to prevent

                    # the special coding for column renames from also affecting

                    # regular regexp/word replacements

                    if text: in_word_re = (excluded_prefix_re+in_word_re

                        +excluded_suffix_re)

                    in_ = '(?:'+in_+'|'+in_word_re+')'

            repls.append((r'(?m)'+in_, out))

    stream.close()

    def repl_all(str_):

        str_ = strings.ustr(str_)

        for repl, with_ in repls:

            if with_ == 'unescape_html()': with_ = repl_unescape_html

            str_ = re.sub(repl, with_, str_)

        return str_

    # Modify map or file

    if col_num != None:

        reader = csv.reader(sys.stdin)

        writer = csv.writer(sys.stdout)

        cols = reader.next()

        writer.writerow(cols)

        for row in reader:

            row[col_num] = repl_all(row[col_num])

            writer.writerow(row)

    else: sys.stdout.write(strings.to_raw_str(repl_all(sys.stdin.read())))

main()