Donald writes:
> I want to compare file a(single value) to file b(multiple values) and
> extract the multiple values into one record.

Your last example command file would appear to be trying to read each record
from file A then compare it against every record in File B.  This is a fine
way to do it if you're not in a hurry, since, if your two 100,000 record
files happened to be full it would require ten billion iterations through
your inner loop and 100,000 FCOPYs of file B's input into its Message File.

To get this to work in a reasonable amount of time (like before the sun
burns out and it's too dark to read the output) you should probably sort
both files on your "key" value, then perform a "merge" of the two files to
create your output.  Basically you read a record from each file and based on
comparing the two key values you read the "lesser" file until the keys match
and then read the B file until the key changes while accumulating your
output.  There are various boundary conditions to deal with when you hit the
end of a file to make sure you don't lose the last record, etc.

The "merge" solution will run in O(N+M) time rather than the original
O(N*M), a factor of improvement similar to the number of records in the
larger file.

This is a complex enough task that most people would use a "real"
programming language rather than a CI script, but it's certainly possible to
implement using the CI if you insist.

G.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *