UNIX for Dummies Questions & Answers

Hello,

I often have this problem:

I have a file with a column of unique identifiers

e.g. file1 below has an id column and data column/columns with p rows:

Code:

 
cat data1
dog data2
cow data3
.
.
.
elephant datap-1
horse datap

and I have a conversion file,file2, with n<p rows and 2 columns: a column of the original ids and a column with the converted ids:
e.g.

Code:

 
horse earl
cat maggi
.
.
.
elephant jim

I want to out put the following output file, file3, with n rows:

Code:

 
earl datap
maggi data1
.
.
.
jim datap-1

Note that the n rows in the conversion file, file2, are not in the same order as in file3

I.E., I want file3 to be a subset of file1 where the first column of unique ids are replaced with converted unique ids using the conversion file.

Thanks much!

Try:

Code:

awk 'NR==FNR{A[$1]=$2;next}{print $2,A[$1]}' file1 file2

Thanks, this is really helpful! What if file1 had several fields after the first one, and I wanted to print out all of the fields (not just $2)?

Something like this:

Code:

awk 'NR==FNR{p=$1;sub($1 FS,x);A[p]=$0;next}{print $2,A[$1]}' file1 file2

It works very well thanks!