CODEWITHSUNDEEP
sedawk
firoz
firoz

Reputation: 91

how to write repeat patterns

A file has list of unique "tags" and "values" separated by tab. I want to repeat the tag according to the given value. Example of input file:

tag  value

AAAAA   2
BBBBB   1
CCCCC   3
DDDDD   5

Expected output File

AAAAA 
AAAAA 
BBBBB 
CCCCC 
CCCCC 
CCCCC 
DDDDD 
DDDDD 
DDDDD
DDDDD
DDDDD

Could you please tell me the awk/sed command. Thanks a lot.

Upvotes: 2

Views: 97

Answers (5)

potong
potong

Reputation: 58478

This might work for you (GNU sed):

sed -r 's/(\S+)\s+(\S+)/seq \2 | sed c\1/e' file

Split the line into arguments for seq and sed commands and evaluate.

Upvotes: 1

Thor
Thor

Reputation: 47189

Doing arithmetic in sed is a pain, so I would avoid that. awk and perl are good choices, you can also straightforwardly do it with bash:

while read tag value; do
  while ((value--)); do 
    printf "%s\n" "$tag"
  done
done < infile

Or as a one-liner:

while read tag value; do while ((value--)); do printf "%s\n" "$tag"; done; done < infile

Output:

AAAAA
AAAAA
BBBBB
CCCCC
CCCCC
CCCCC
DDDDD
DDDDD
DDDDD
DDDDD
DDDDD

Upvotes: 0

jaypal singh
jaypal singh

Reputation: 77155

Here is a perl alternate:

$ perl -ane 'print "$F[0]\n"x$F[1]' file
AAAAA
AAAAA
BBBBB
CCCCC
CCCCC
CCCCC
DDDDD
DDDDD
DDDDD
DDDDD
DDDDD

Upvotes: 2

perreal
perreal

Reputation: 98088

An alternative version for GNU awk:

awk '{while($2--) print $1}'

This is not a good problem to solve with sed. You need to replace a number n with n 1's (for example 3 with 111) and print the word as you consume these 1's.

Upvotes: 6

Jotne
Jotne

Reputation: 41460

This awk should do:

awk '{for (i=1;i<=$2;i++) print $1}' file
AAAAA
AAAAA
BBBBB
CCCCC
CCCCC
CCCCC
DDDDD
DDDDD
DDDDD
DDDDD
DDDDD

It loops the number of times found in column #2, then print the word in column #1

Upvotes: 3

Related Questions