python：用其他文件中的单词替换文件中的单词

Question

I have a large text-file in which there are words I want to replace. I put those words in a csv-file, because I'm constantly adding and changing words and do not want to put the words in the python script itself. On each line is a word I want to replace, followed by the word I want to replace it with. Like this:

A_old,A_new
another word,another new word
something old,something new
hello,bye

I know how to replace single words in files with python with the string replace function, but I don't know how to do this when the words are listed in a different file. I tried my best, but I can't wrap my head around how to work with dictionaries/lists/tuples. I am rather new to python, and until now I managed with examples from around the internet, but this is beyond my capabilities. I got all kinds of errors like 'unhashable type: list' and 'expected a character buffer object'. The last thing I tried was the most succesful in that I didn't get any errors, but then nothing happened either. This is the code. I'm sure it's ugly, but I hope it's not entirely hopeless.

reader = csv.reader(open('words.csv', 'r'))
d = {}
for row in reader:
    key, value = row
    d[key] = value

newwords = str(d.keys())
oldwords = str(d.values())

with open('new.txt', 'wt') as outfile:
    with open('old.txt', 'rt') as infile:
        for line in infile:
            outfile.write(line.replace(oldwords,newwords))

The reason I am doing this is because I'm working on a cookbook with an ingredient based index, and I don't want an index with both 'carrot' and 'carrots', instead I want to change 'carrot' into 'carrots', and so on for all the other ingredients. Thanks a bunch for a nudge in the right direction!

Answer 1

First you make a list of pairs (old_word, new_word) from 'word.csv' :

old_new = [i.strip().split(',') for i in open('words.csv')]

Then, you can replace line by line :

with open('new.txt', 'w') as outfile, open('old.txt') as infile:
    for line in infile:
        for oldword, newword in old_new:
            line = line.replace(oldword, newword)
        outfile.write(line)

or in the whole file at once :

with open('new.txt', 'w') as outfile, open('old.txt') as infile:
    txt = infile.read()
    for oldword, newword in old_new:
        txt = txt.replace(oldword, newword)    
    outfile.write(txt)

but you have to replace one word at a time.

Answer 2

In your code example you read the replacement word pairs into a dictionary, and then into two lists with keys and values. I'm not sure why.

I propose to read the replacement words into a list of tuples.

with open('words.csv', 'rb') as rep_words:
    rep_list = []
    for rep_line in rep_words:
        rep_list.append(tuple(rep_line.strip().split(',')))

Then you can open the old.txt and new.txt files and perform the replacement using a nested for loop

with open('old.txt', 'rb') as old_text:
    with open('new.txt', 'wb') as new_text:
        for read_line in old_text:
            new_line = read_line
            for old_word, new in rep_list:
                new_line = new_line.replace(old_word, new_word))
            new_text.write(new_line)