3 
bash-3.2$ cat sample.log sample.log.1 sample.log.2
ID COL1 COL2 COL4
1  col1 col2 col4
2  c1   c2   c4
3  co1  co2  co4

ID COL3 COL1
1  col3 col1
2  c3   c1
3  co3  co1

ID COL1 COL2 COL3
1  col1 col2 col3
2  c1  c2   c3
3  co1  co2  co3

I need to write an awk script such that it gives me the values of the columns for a particular id like a select query on multiple tables in db.

give me col1 col2 and col3 fields for id 1 and should not duplicate result. meaning the result should be like

The result should be
ID COL1 COL2 COL3
1  col1 col2 col3

but not

The result should be
ID COL1 COL2 COL3 COL3
1  col1 col2 col3 col3

Even a suggestion is also good.

awk ' BEGIN { while ( (getline line < "sample.log") > 0 ) {ids[substr(line,1,index(line," ")-1)];} } { // get the column values here based on the stored id's .. } ' sample.log sample.log.1 sample.log.2

I am trying to do something like that mentioned above. I am not sure if it is a good idea.

Improve this question
1
  • 1
    You could just drop the data into SQLite and then run actual SQL queries against it. Commented Feb 23, 2012 at 13:10

1 Answer 1

Reset to default
3

You can use the join command to perform this task:

join -1 1 -2 1 sample.log sample.log.1 -o 1.1,1.2,1.3,2.2

The output will be 'single space' separated, but you can use awk to reformat it to be column aligned.

Note that the join input files must be sorted.

Improve this answer

You must log in to answer this question.