I have a csv file that has a date/time column and a value column and a name column. think a check in check out register at a receptionists desk using one integer in the value column to denote ins and a different integer for outs, always the same 2 integers used.[IE 6 always denotes a check in and 8 always denotes a check out]. I need to be able to take every register for a certain name and determine how long they were there each time they were there. basically,

name,time checked in,time checked in,time checked in...

with a new column for each checkin - checkout cycle. a large part of my problem is that I am not familiar with any command line program that can interpret date/timestamps and I can't seem to think of a proper search string to find one.

the date -d command can read timestamps. Try info date for more information.

not sure how that will take 2 time stamps and give me the amount of time that took place between the two. sorry if I wasn't clear enough in my first post.

If you want to do date/time calcs, you could try this page for ideas http://tldp.org/LDP/abs/html/interna...s.html#TIMEOUT, but it might be easier to use Perl instead...

Hi.
Suppose you had a fixed starting time, say sometime in 1970. If you used date to convert both dates to the seconds since that fixed time, then you could subtract the two, and that result should be the time between the two timestamps.

This may require some reading of info date as jim advised, as well as some experimentation, but could be rewarding ... cheers, makyo

maybe consider using a better tool than shell scripting for this?
like perl or python?

sorry for the delay, not working over the weekend. yeah I backed up a step and found unix time in the original log file. I need it to show up next to my name that I don't know perl, python even less. they get recommended for every question I ask. I'm sure I'll have more questions when I get a little farther in to this.