Well here's an idea anyway... I wouldn't slurp the whole file at once, I would process line by line assuming all pertinent info per entry is a 1-liner as shown.
You could first split on | which gives portionA, portionB. For portionA, split on white space; the second field is the time. For portion B, use a regex to grab the "thread: number" as a hash key. There must be a module or unix tool for converting colon delimited time to minutes/seconds, and back again. By hook or by crook, do this conversion and += the hash with the resulting value. ?? Cheers, Nathanael -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
