Hi, khdubya.
If you are still interested in doing this, I have been working on a rudimentary perl code that may work.
Consider the two files:
Code:
File: data1
1 one
2 two
3 three
4 four
5 five
6 six
7 seven
8 eight
9 nine
10 ten
File: data2
1 alpha
2 beta
3 gamma
The perl code would be called with a series of arguments like:
filename:line-count (default:1) and produces output like so:
Code:
% ./shuffle data1:3 data2
one
two
three
alpha
four
five
six
beta
seven
eight
nine
gamma
ten
I timed it on sample files for your ratio
l1:20 l2 of (line count) size:
to get:
Code:
real 0m0.044s
user 0m0.031s
sys 0m0.006s
The drawback is that while the code seems to work, I have not finished it, there could be bugs in it, and so I provide no warranty.
Let me know if you'd like me to post it; about 80 lines ... cheers, makyo