I there a way to split single line into multiple lines with 3 columns.
New line characters are missing at the end of all the lines in the file.

I tried using awk, but it is splitting each column as one row instead of 3 columns in each row.

awk ' gsub(",", "n") 6' filename

where filename's content looks like:

A,B,C,D,E,F,G,H,I,J,K,L,M,N,O

Desired output has 3 columns in each line:

A,B,C
D,E,F
G,H,I
J,K,L
M,N,O

Using awk

$ awk -v RS='[,n]' 'a=$0;getline b; getline c; print a,b,c' OFS=, filename
A,B,C
D,E,F
G,H,I
J,K,L
M,N,O

How it works

• 
  

  -v RS='[,n]'

  
  
  

  This tells awk to use any occurrence of either a comma or a newline as a record separator.

  
  


• 
  

  a=$0; getline b; getline c

  
  
  

  This tells awk to save the current line in variable a, the next line in varaible b, and the next line after that in variable c.

  
  


• 
  

  print a,b,c

  
  
  

  This tells awk to print a, b, and c

  
  


• 
  

  OFS=,

  
  
  

  This tells awk to use a comma as the field separator on output.

  
  

Using tr and paste

$ tr , 'n' <filename | paste -d, - - -
A,B,C
D,E,F
G,H,I
J,K,L
M,N,O

How it works

• 
  

  tr , 'n' <filename

  
  
  

  This reads from filename while converting all commas to newlines.

  
  


• 
  

  paste -d, - - -

  
  
  

  This paste to read three lines from stdin (one for each -) and paste them together, each separated by a comma (-d,).

  
  

Alternate awk

$ awk -v RS='[,n]' 'printf "%s%s",$0,(NR%3?",":"n")' filename
A,B,C
D,E,F
G,H,I
J,K,L
M,N,O

How it works

• 
  

  -v RS='[,n]'

  
  
  

  This tells awk to use any occurrence of either a comma or a newline as a record separator.

  
  


• 
  

  printf "%s%s",$0,(NR%3?",":"n")

  
  
  

  This tells awk to print the current line followed by either a comma or a newline depending the value of the current line number, NR, modulo 3.

  
  

sed 's/(([^,]+,)3)/1n/g;s/,n/n/g' filename

I know that you asked for an awk solution, and I'll now try to submit that as an edit to this answer, but for me a sed solution was simpler... ... and user john1024 beat me to it, with a fine awk solution. See there. His paste and tr solution is probably the most proper classic unix-ish answer.

1. This solution uses the extended regex features of GNU sed.




2. (..) is a regex collection group. Note that the solution uses two, one nested within the other.




3. [^,]+, is any string that doesn't have a comma, followed by a comma. In your case, a column or field.




4. 3 is a regex multiplier, indicating to use the prior regex expression three times.




5. 1 is a regex back-reference. to the prior regex.




6. g means do it for all instances on the line.




7. s/,n/n/g removes the trailing comma. It's necessary to include the newline character here, because sed is still considering the input as a single line.