From: Williams, Ken [mailto:[EMAIL PROTECTED]
That sounds convenient but potentially dangerous - does it understand leap years/seconds, for example? And various other pitfalls of date/time math? ------------------------------------ Regarding leap *years*, I agree that it should account for them, and I think it does for many decades. Regarding leap *seconds*: Maybe it's a leap of faith on my part ;-) but doesn't essentially all date/time software ignore leap seconds? I thought leap seconds were scheduled "as necessary" to keep things in synch with the star observations, so they can't be predicted accurately for years in advance. And if Sqlite *could* account for leap seconds, wouldn't it better if it did not? For example, most computer systems are not set up so as to implement leap seconds as they occur, so any software measuring durations would be incur an error if it made an adjustment "unilaterally," right? For those few systems where a second is quite significant, I wonder if it's better to simply correct the computer's clock at a time when you know you can account for this little quantum leap. [speaking only for myself and not on behalf of my employer] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
