Split a file using awk command.

Shell Programming and Scripting
1 
awk 'FNR == 1 { c = 1 } { print > (f c) } !FNR%n { close(f c); ++c }' n=$files_per_stream f=$input_path/filename_ $input_file

$input_file with some records are splitted into files named filename_1,filename_2...etc according to $files_per_stream.

Plz help me know how and if anyone has better option plz let me know ??

thnx in advance.

2

Perhaps try the split command.

1 Like
3

but how this awk command is working..
plz help me out.

thnx.

4

FNR == 1 { c = 1 } - For the first record in each input file, set c to 1.
{ print > (f c) } - For all records, write the current record to file (f and c concatenated, e.g. $input_path/filename_1)
!FNR%n { close(f c); ++c }' - if the current record number is exactly divisible by n, close the current output file and increment variable c.
n=$files_per_stream f=$input_path/filename_ $input_file - set variables & input file.

5

thanks CarloM but i still have some doubts...

1.) untill current record number is exactly divisible by n ,the filename_1 will be appended ????
This is an implicit behaviour of awk or what part of code code this ??

2.) what is "!" used for ?????

6

@guptam: I don't understand, did you find this code somewhere, how did you come across this? What are your requirements, what did you try yourself?

7
  1. Yes.
    Which behaviour do you mean? The second statement explicitly writes to the named file, and the third one explicitly checks FNR & changes the variable used to generate the filename.
  2. Not/negate - if FNR%n is FALSE or zero, !FNR%n is TRUE.
8

pardon me for inconvinience...

actually my operating system got crashed today...so i am unable to try myself...

hope i will soon get my system right and then i will share some some relevant questions.

thanks once again CarloM.