Question

How to get rows from one file, that contains in another file

Example, i have

// first
foo
bar

// second
foo;1;3;p1
bar;1;3;p2
foobar;1;3;p2

This files are big, first file contain ~ 500 000 records, and second ~ 20-15 millions

I need to get this result

// attention there is no "p1" or "p2" for example
foo;1;3
bar;1;3

Answer 1

This looks like it wants the join command, possibly with sorting. But with millions of records, it’s time to think seriously about a real DBMS.

join -t\; -o 0,2.2,2.3 <(sort -t\; -k 1,1 first) <(sort -t\; -k 1,1 second)

(This requires bash or zsh for the <(command) syntax; portably, you would need to sort into temporary files or keep the input files sorted.)