CODEWITHSUNDEEP
bashawk
KunalC
KunalC

Reputation: 544

Filtering data in a text file in bash

I am trying to filter the data in a text file. There are 2 fields in the text file. The first one is text while 2nd one has 3 parts seperated by _. The first part in the second file is date in yyyyMMdd format and the next 2 are string:

xyz  yyyyMMdd_abc_lmn

Now I want to filter the lines in the file based on the date in the second field. I have come up with the following awk command but it doesn't seems to work as it is outputting the entire file definitely I am missing something.

Awk command:

awk -F'\t' -v ldate='20140101' '{cdate=substr($2, 1, 8); if( cdate <= ldate) {print $1'\t\t'$2}}' label

Upvotes: 1

Views: 1337

Answers (2)

Ed Morton
Ed Morton

Reputation: 204468

Try this:

awk -v ldate='20140101' 'substr($NF,1,8) <= ldate' label

Upvotes: 1

jaypal singh
jaypal singh

Reputation: 77175

Try:

awk -v ldate='20140101' '{split($2,fld,/_/); if(fld[1]<=ldate) print $1,$2}' file

Note:

  • We are using split function which basically splits the field based on regex provided as the third element and stores the fields in the array defined as second element.
  • You don't need to set -F'\t unless your input file is tab-delimited. The default value of FS is space, so defining it to tab might throw it off in interpreting $2.

  • To output with two tabs you can set the OFS variable like:

    awk -F'\t' -v OFS='\t\t' -v ldate='20140101' '{split($2,fld,/_/); if(fld[1]<=ldate) print $1,$2}' file

Upvotes: 3

Related Questions