Replace multiple consecutive white spaces with one comma in Unix

Question

I have the following sed command:

sed 's/\s/,/g' input > output.csv

(I got the command from this related topic)

which turns the following input:

SNP  A1  A2     FRQ    INFO      OR      SE       P
10:33367054   C   T  0.9275  0.9434  1.1685  0.1281  0.1843
10:33367707   G   A  0.9476  0.9436  1.0292  0.1530  0.8244
10:33367804   G   C  0.4193  1.0443  0.9734  0.0988  0.6443
10:33368119   C   A  0.9742  0.9343  1.0201  0.1822  0.9156

into:

SNP,,A1,,A2,,,,,FRQ,,,,INFO,,,,,,OR,,,,,,SE,,,,,,,P
10:33367054,,,C,,,T,,0.9275,,0.9434,,1.1685,,0.1281,,0.1843
10:33367707,,,G,,,A,,0.9476,,0.9436,,1.0292,,0.1530,,0.8244
10:33367804,,,G,,,C,,0.4193,,1.0443,,0.9734,,0.0988,,0.6443
10:33368119,,,C,,,A,,0.9742,,0.9343,,1.0201,,0.1822,,0.9156

I need a command that turns the multiple consecutive spaces into just one commma, to give me an output like this:

SNP,A1,A2,FRQ,INFO,OR,SE,P
10:33367054,C,T,0.9275,0.9434,1.1685,0.1281,0.1843
10:33367707,G,A,0.9476,0.9436,1.0292,0.1530,0.8244
10:33367804,G,C,0.4193,1.0443,0.9734,0.0988,0.6443
10:33368119,C,A,0.9742,0.9343,1.0201,0.1822,0.9156

Any ideas?

Answer 1

If you want to use sed, you can use this one:

$ sed 's/ \{1,\}/,/g' file
SNP,A1,A2,FRQ,INFO,OR,SE,P
10:33367054,C,T,0.9275,0.9434,1.1685,0.1281,0.1843
10:33367707,G,A,0.9476,0.9436,1.0292,0.1530,0.8244
10:33367804,G,C,0.4193,1.0443,0.9734,0.0988,0.6443
10:33368119,C,A,0.9742,0.9343,1.0201,0.1822,0.9156

It is based on glenn jackman's answer to How to strip multipe spaces to one using sed?.

It can also be like

sed 's/[[:space:]]\{1,\}/,/g' file

And note you can use sed -i.bak '...' file to get an in place edit, so that the original file will be backed up as file.bak and file will have the edited content.

---

But I think it is more clear with tr. With it, you can squeeze the spaces and then replace each one of them with a comma:

$ tr -s ' ' < file | tr ' ' ','
SNP,A1,A2,FRQ,INFO,OR,SE,P
10:33367054,C,T,0.9275,0.9434,1.1685,0.1281,0.1843
10:33367707,G,A,0.9476,0.9436,1.0292,0.1530,0.8244
10:33367804,G,C,0.4193,1.0443,0.9734,0.0988,0.6443
10:33368119,C,A,0.9742,0.9343,1.0201,0.1822,0.9156

By pieces:

$ tr -s ' ' < file
SNP A1 A2 FRQ INFO OR SE P
10:33367054 C T 0.9275 0.9434 1.1685 0.1281 0.1843
10:33367707 G A 0.9476 0.9436 1.0292 0.1530 0.8244
10:33367804 G C 0.4193 1.0443 0.9734 0.0988 0.6443
10:33368119 C A 0.9742 0.9343 1.0201 0.1822 0.9156

From man tr:

tr [OPTION]... SET1 [SET2]

Translate, squeeze, and/or delete characters from standard input, writing to standard output.

-s, --squeeze-repeats

replace each input sequence of a repeated character that is listed in SET1 with a single occurrence of that character

Answer 2

If you enable extended regular expressions with -r, then you can just add + to \s which means one or more:

$ sed -r 's/\s+/,/g' file.txt
SNP,A1,A2,FRQ,INFO,OR,SE,P
10:33367054,C,T,0.9275,0.9434,1.1685,0.1281,0.1843
10:33367707,G,A,0.9476,0.9436,1.0292,0.1530,0.8244
10:33367804,G,C,0.4193,1.0443,0.9734,0.0988,0.6443
10:33368119,C,A,0.9742,0.9343,1.0201,0.1822,0.9156

For reference:

-r, --regexp-extended
    use extended regular expressions in the script.

---

Note: On Mac OS X, sed is based on BSD and does not have the GNU extensions so you will have to use the -E flag:

-E    Interpret regular expressions as extended (modern) regular expressions rather
      than basic regular expressions (BRE's). The re_format(7) manual page fully 
      describes both formats.

Answer 3

Here is a very simple solution with awk

awk '{$1=$1}1' OFS=, file
SNP,A1,A2,FRQ,INFO,OR,SE,P
10:33367054,C,T,0.9275,0.9434,1.1685,0.1281,0.1843
10:33367707,G,A,0.9476,0.9436,1.0292,0.1530,0.8244
10:33367804,G,C,0.4193,1.0443,0.9734,0.0988,0.6443
10:33368119,C,A,0.9742,0.9343,1.0201,0.1822,0.9156

$1=$1 reformat the file so that all extra spaces are set to one space.