Add space as delimiter in csv.writer

Question

I am trying to output a combination from my_list 1 & 2

import itertools
import csv

my_list1=["+Red"]
my_list2=["+Lip +Stick","+Magic"]

# Please note: Space is kept intentionally in "+Lip +Stick" above

combinations=itertools.product(my_list1,my_list2)
with open('Txt_of_Keywords.txt','w') as f1:
    writer=csv.writer(f1,lineterminator='\n',)
    for c in combinations:
        writer.writerow(c)

Output without delimiter:

+Red,+Lip +Stick
+Red,+Magic

I want to replace comma with space in the above output.

I tried adding space using delimiter as below:

writer=csv.writer(f1,delimiter=' ',lineterminator='\n',)

Output with delimiter:

+Red "+Lip +Stick"
+Red +Magic

Above output incorrectly writes quotation marks. This is because "+Lip +Stick" has space in between while defining my_list2. Sadly, i want that space there.

Please help me to replace comma with space (as highlighted above in bold).

Answer 1

Why are the quotes 'incorrect'? You wrote a value containing a space, the writer uses quotes to make sure that space is not seen as a delimiter.

If you don't want to have quoting, you may as well just write your values directly to the file. If you use the print() function , you'll get newlines and spaces for free:

with open('Txt_of_Keywords.txt','w') as f1:
    for c in combinations:
        print(*c, file=f1)

You can't tell the csv.writer() object to ignore delimiters in values; your choices are between quoting, using an escape character, or if you disabled both, an error being raised. Your other option is to split your columns on the delimiter, but that's just more work just to make the csv module work, where you don't really need to use that module at all .

Answer 2

you're trying to lure csv module into not quoting/escaping the space. Setting quoting to csv.QUOTENONE and/or csv.quotechar to "empty" doesn't work, raising exceptions like:

quotechar must be set if quoting enabled

or

need to escape, but no escapechar set

csv has checks to ensure that it cannot happen, for reversibility reasons.

So the best way would be to split each value and flatten the list of the tokens obtained with this generator comprehension fed to itertools.chain.from_iterable :

writer.writerow(list(itertools.chain.from_iterable(x.split() for x in c)))

full example:

my_list1=["+Red"]
my_list2=["+Lip +Stick","+Magic"]

# Please note: Space is kept intentionally in "+Lip +Stick" above

combinations=itertools.product(my_list1,my_list2)
with open('Txt_of_Keywords.txt','w') as f1:
        writer=csv.writer(f1,lineterminator='\n',delimiter=" ")
        for c in combinations:
            writer.writerow(list(itertools.chain.from_iterable(x.split() for x in c)))

now I'm getting:

+Red +Lip +Stick
+Red +Magic

You could also write your list without the csv module at this point, since the main point of csv is to easily escape chars that could interact with the delimiter and handle multi-line (see Martjin new answer for that).