find one field into another

lcvs · 06-28-2012, 09:19 PM

Hi guys !

Using awk, I try to read a column and see if a string from a field in this column is present into a longer string from another field in the same column.
If it is the case, returns the sub-string length (RLENGTH) in a new field.
input:

Code:

111|abcd1234efgh|zzzzz
222|1234|yyyyyyyyyy
333|ij1234k|xxx
444|ui987654g|ww

I tried using match function:

Code:

awk 'BEGIN{FS=OFS="|"} {match(a[$2],$2); print $4 = RLENGTH}1' input

to obtain:

Code:

111|abcd1234efgh|zzzzz|8
222|1234|yyyyyyyyyy|4
333|ij1234k|xxx|5
444|ui987654g|ww|-1

But I don't manage to write a for loop to scan the entire column by taking every field one by one as a reference for match function. It returns $4 = "-1" for every record.

chrism01 · 06-28-2012, 09:35 PM

I reckon you can extend this one liner to match first, then sort by length http://unstableme.blogspot.com.au/20...g-awk-and.html & only print last record

lcvs · 06-28-2012, 09:45 PM

Hi Chrism01, thanks for your help !

The problem is that it doesn't tell me if it found the shorter string within another field ($4 = -1 for all records).

lcvs · 06-28-2012, 10:24 PM

edited first post

schneidz · 06-30-2012, 07:50 PM

not sure what you are trying to do (shouldnt your 3rd row = 6 ?).

anyways, this is my stab at it:

Code:

[schneidz@hyper ~]$ cat lcvs.txt 
111|abcd1234efgh|zzzzz
222|1234|yyyyyyyyyy
333|ij1234k|xxx
444|ui987654g|ww
[schneidz@hyper ~]$ awk 'BEGIN{FS=OFS="|"}  {print  index($2,"1234")}' lcvs.txt 
5
1
3
0