delete-phrases-repetition.py

#!/usr/bin/python3
#"C:\Users\flo-f\AppData\Roaming\Microsoft\Windows\Start Menu\Programs\Python 3.6\Python 3.6 (32-bit).lnk"   C:\Users\flo-f\Downloads\delete-phrases-repetition.py "C:\Users\flo-f\Desktop\Corpus ricostruito (FG).txt" 20 "C:\Users\flo-f\Desktop\Corpus nuovo.txt"
import sys
import os.path

wnumber = 10
searchthis = " " #this is the separator: if " " you look for words, if "." you look for phrases

def nth_replace(string, old, new, n=1, option='only nth'):
    """
    This function replaces occurrences of string 'old' with string 'new'.
    There are three types of replacement of string 'old':
    1) 'only nth' replaces only nth occurrence (default).
    2) 'all left' replaces nth occurrence and all occurrences to the left.
    3) 'all right' replaces nth occurrence and all occurrences to the right.
    """
    if option == 'only nth':
        left_join = old
        right_join = old
    elif option == 'all left':
        left_join = new
        right_join = old
    elif option == 'all right':
        left_join = old
        right_join = new
    else:
        print("Invalid option. Please choose from: 'only nth' (default), 'all left' or 'all right'")
        return None
    groups = string.split(old)
    nth_split = [left_join.join(groups[:n]), right_join.join(groups[n:])]
    return new.join(nth_split)

filein = ""
fileout = ""
text = ""

if len(sys.argv) > 1:
    filein = sys.argv[1]
if filein == "-h":
    print("Usage:\n delete-phrases-repetition.py \"fileinput.txt\" 10 \"fileoutput.txt\" \" \"\n delete-phrases-repetition.py \"fileinput.txt\" 1 \"fileoutput.txt\" \".\"\nNOTE: The number represents occurrences of the separator to get a phrase. If you use \".\" as separator, the number should be 1.")
    sys.exit()

if len(sys.argv) > 2:
   wnumber = int(sys.argv[2])

if len(sys.argv) > 3:
    fileout = sys.argv[3]

if len(sys.argv) > 4:
    searchthis = sys.argv[4]

if filein != "" and os.path.isfile(filein):
    text_file = open(filein , "r")
    #text = text_file.read().replace("\n", "")
    text = text_file.read()
    text_file.close()
else:
    text = input("Testo:")

if wnumber < 0:
    wnumber = 10


active = 1
pos = 0
while active:
    wpos = pos
    npos = pos
    #read a specific number of words
    for i in range(wnumber):
        wpos = text.find(searchthis, npos+1)
        if wpos > 0:
            npos = wpos
    #check if we reached someway the end of text
    if npos > len(text)-1:
        if pos > len(text)-1:
            break
        else:
            npos = len(text)-1
    #read this phrase
    tmpstring = text[pos:npos]
    #replace all occurrences of the phrase, after the first one, with nothing
    if tmpstring != "":
        newtext = nth_replace(text, tmpstring, "", 2, "all right")
        text = newtext
    pos = text.find(searchthis, pos+1)+1 #continue from next word
    if pos <= 0:
        pos = len(text)

#delete double spaces
newtext = newtext.replace("  ", " ")

if fileout:
    text_file = open(fileout, "w")
    text_file.write(newtext)
    text_file.close()
else:
    print(newtext)