#!/usr/bin/python3

import sys, getopt, re, csv

the_list = "the963_text_document.txt"
the_list_csv = "the963.csv"
usage_text = 'Usage: specify "go" and we\'ll create a CSV from the list ['+the_list+'].'
savedict = {} # saving in dict with "num" as the index, for flexibility; expect 963 total when done.

def usage():
    print(usage_text)
    sys.exit(2)

def process(the_list):
    print("Processing ["+the_list+"]:")
    num = 0 # line number
    count = 0 # count of people
    person_rus_found = False
    person_eng_found = False
    title_rus_found = False
    comma_found = False
    with open(the_list, "r") as f:
        for line in f:
            line = line.rstrip()
            if line == "" or line == "–" or re.match("^.$", line): # empty, dash, or a single character
                #print( "INFO: skipping line")
                continue
            num += 1
            print(f'{num : >4}: [{line}]')
            # If previous line ended with a comma, add this line to it; it's part of the title/description.
            # Note that there's one entry (#664) with three lines, each of the first two end with commas,
            # so allow this to repeat as well.
            if comma_found:
                savedict[count]['title_rus'] += " "+line
                print(f"<<<COMMA IN ABOVE SO COMBINING, NEW title_rus is {savedict[count]['title_rus']}")
                if line[-1] == ",":
                    print(f"<<<ANOTHER COMMA FOUND {count}, {line}")
                    comma_found = True
                else:
                    comma_found = False
            elif not person_rus_found:
                # Some titles/descriptions have more than one line.  Second line starts with a "(", so easy to identify.
                # Need to add it to the previous description, and we haven't updated the count yet so it's that one.
                if re.match("^\(", line):
                    savedict[count]['title_rus'] += " "+line
                    print(f"<<<TITLE RUS UPDATED {count}, NOW {savedict[count]['title_rus']}")
                    continue
                count += 1
                print(f"<<<PERSON RUS FOUND {count}, {line}")
                person_rus_found = True
                # Define key for new dictionary entry
                savedict[count] = {}
                # Add values for new entry
                savedict[count]['name_rus'] = line
                # NOTE: sometimes, both names are on the same line, the English after it in parentheses.
                # So, check if it matches that pattern and if so, fill in both names in this section.
                result = re.match("(.*)\((.*)\)", line)
                if result:
                    savedict[count]['name_rus'] = result.group(1).rstrip()
                    savedict[count]['name_eng'] = result.group(2).rstrip()
                    print(f"<<<PERSON ENG FOUND ON SAME LINE AS PERSON RUS {count}, {line}")
                    person_eng_found = True
                if re.match("Шанг Юн БЭЙ", line):
                    print(f"<<<SPECIAL CASE MISSING ENG NAME #102 {count}, {line}")
                    # Special case: there's no English translation for this, so set to empty string and set flag showign we saw it.
                    savedict[count]['name_eng'] = "" # emptry string
                    person_eng_found = True
            elif not person_eng_found:
                print(f"<<<PERSON ENG FOUND {count}, {line}")
                person_eng_found = True
                # Remove leading and trailing parentheses from the name
                name_eng = line
                p=re.search("\((.*)\)",line)
                if p:
                    name_eng = p.group(1)
                # Add values for new entry
                savedict[count]['name_eng'] = name_eng
            elif not title_rus_found:
                print(f"<<<TITLE RUS FOUND {count}, {line}")
                title_rus_found = True
                # Add values for new entry
                savedict[count]['title_rus'] = line
                if line[-1] == ",":
                    print(f"<<<COMMA FOUND {count}, {line}")
                    comma_found = True
            if title_rus_found:
                # If we found the title/description in Russian, then that's the last so start looking for the next one.
                person_rus_found = False
                person_eng_found = False
                title_rus_found = False

def write_csv():
    with open(the_list_csv, 'w', newline='') as csvfile:
        spamwriter = csv.writer(csvfile, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
        for i in sorted(savedict):
            spamwriter.writerow([savedict[i]['name_rus'],savedict[i]['name_eng'],savedict[i]['title_rus']])

def main(argv):
    try:
        opts, args = getopt.getopt(argv,'')
    except getopt.GetoptError:
        usage()
    if len(args) == 1 and args[0] == 'go':
        process(the_list)
        print(f"Processed; total of {len(savedict)} entries in the savedict dictionary.  Now, to write the CSV file.")
        write_csv()

    else:
        usage()

if __name__ == "__main__":
   main(sys.argv[1:])