Processing a CSV file.

process_csv.py

import csv
import codecs
import re

def load_dictionary(csv_file, idx, lbl):
	d = {}
	f = open(csv_file, encoding="utf-8")
	s = csv.reader(f)
	for r in s:
		i = r[idx] # Read index
		l = r[lbl] # Read label
		d[i] = l
	f.close()
	return d

def load_csv(csv_data):
	f = open(csv_data, encoding="utf-8")
	rows = csv.reader(f)
	for row in rows:
		extract_data(row)
	f.close()

	return

def extract_data(data_row):

	topic = data_row[7]
	rank = data_row[4]
	word_type = data_row[8]
	audio_file = data_row[9]
	article = data_row[11]
	citation = data_row[12]
	en_gloss = data_row[13]
	in_situ = data_row[14]
	blank = data_row[15]
	focal = data_row[16]

	if word_type in word_types: word_type = word_types[word_type]
	if topic in topics: topic = topics[topic]

	filled = blank

	if citation != "":
		# Remove redundant spaces
		blank = re.sub(r' +',' ',blank)
		citation = re.sub(r' +',' ',citation)

		regex = r"_+"
		madlibs = re.findall(regex, blank)

		fills = in_situ.split(" ")

		if len(fills) == len(madlibs):
			for (i, madlib) in enumerate(madlibs):
				filled = filled.replace(madlib, "{"+fills[i]+"}", 1)

		stripped = re.sub(r'{','',filled)
		stripped = re.sub(r'}','',stripped)

		if stripped != citation:
			print (fills)
			print (blank)
			print (filled)
			print (citation)
			print ("")

	if 1:
		print (rank, article, focal, "> "+en_gloss)
		print (topic)
		print (filled)
		print ("")

	return

if __name__ == '__main__':

	wortart_file = "Lextra-Wortart.csv"
	word_types = load_dictionary(wortart_file, 0, 3) # The arguments: file, index column, label column

	themen_file = "Lextra-Themen.csv"
	topics = load_dictionary(themen_file, 0, 5)

	main_file = "Lextra-Words.csv"
	load_csv(main_file)

	print ("Valmis!")