I think that Chuck's calculation is best conceptualized as a two-step one. First, there is UTC-to-local-standard time conversion and vice versa; and Chuck has these two problems well in hand.
Second, there is standard-to-daylight time conversion and vice versa. The standard-to-daylight correction is applied iff the standard time value falls within the local daylight interval, which may be empty. If it is non-empty, o it begins at the STANDARD date-time at which daylight time begins, call if SB, and o it ends at the DAYLIGHT date-time at which daylight time ends, call it DE. Now, fortunately, SE, the standard date-time at which daylight time ends, is just SE = DE -1. The time component of these date-times is all but invariant. Its value is usually 0200, 2 AM for Americans. Their calendar-date components vary widely. John Gilmore, Ashland, MA 01721 - USA ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
