2

I have a rather large text file where there is an extra space between every character;

I t   l o o k s   l i k e   t h i s .

I'd like to remove those extra characters so

It looks like this.

via the Linux terminal. I can't seem to find anyway to do this without removing all of the whitespaces. I'm willing to try any solution at this point. I'd appreciate any nudge in the right direction.

Matthew.Paul.P
  • 23
  • 3
  • 1
    Are you sure they are spaces and not NUL characters? It sounds like you have something encoded using UTF-16, in which case the solution would be to change the encoding to UTF-8 or whatever. – ikegami Nov 30 '13 at 21:26

5 Answers5

5
$ echo 'I t   l o o k s   l i k e   t h i s .  ' | sed 's/\(.\) /\1/g'
It looks like this.
John Kugelman
  • 349,597
  • 67
  • 533
  • 578
3

Are you certain that the intermediate characters are spaces? It is most likely that this is a UTF-16 file.

I suggest you use a capable editor to open it as such and convert it to UTF-8.

Borodin
  • 126,100
  • 9
  • 70
  • 144
  • 1
    Borodin is right. try to type 'file your_file.txt' to get the encoding. you don't want to corrupt your data right? – Pierre Nov 30 '13 at 19:18
3

An awksolution

echo "I t   l o o k s   l i k e   t h i s ." | awk '{for (i=1;i<=NF;i+=2) printf $i;print ""}' FS=""
It looks like this.
Jotne
  • 40,548
  • 12
  • 51
  • 55
1

As long as it's every other character you want to get rid of, you can use python.

>>> s = "I t   l o o k s   l i k e   t h i s ."
>>> print s[0::2]
It looks like this.

If you wanted to do this for the text file, do the following:

with open("/path/to/file.txt") as f:
    f = f.readlines()

with open("/path/to/new.txt") as g:
    for i in f:
        g.write(str(i)[0::2]+"\n")
TW80000
  • 1,507
  • 1
  • 11
  • 18
1
perl -pe 's|(\s+)| " "x (length($1)>1) |ge' file
mpapec
  • 50,217
  • 8
  • 67
  • 127