Does a day start at 1:00 AM everwhere?

[Ken Ray]

On 10/27/06 10:39 AM, "Mark Powell" <mark_powell at symantec.com> wrote:

> My problem is that I have two inputs.  One is the user inputting a date.
> Another is the creation date in seconds that has previously been
> extracted via a 'detailed files' call, which has been concatenated in
> the format below.
> 
> C:/RevolutionStacks/SortTest/!?!Sept_009K.jpg!/9040/1157431138

Ah, now I understand...
 
> What I wanted to do is to express the input date in seconds and test it
> against item -1 of each line in the container (there are potentially >
> 100,000 such lines in a container). Using dateItems or 'word 1 of
> tFileDateTime' as you suggest would require modification to the original
> 'detailed files' extraction algorithm, which doubtless will have a
> ripple effect elsewhere.  So, is there not any reliable way to compare
> seconds to seconds?  Or do I have to go the route of approximation.
> 
> Also Ken, the DST et. al. wrinkles you describe:  does that affect the
> interpretation of a creation date extracted from the file?  Or is the
> complication confined to how a particular user's computer calculates a
> time query?  For example, if a user's OS displays a creation date for
> "foo.txt" as 11/21/05, will Rev not always interpret it as 11/21/05?  Or
> can fencepost error arise where Rev misinterprets the static date?

I really don't know, Mark... I would *assume* that the seconds is relative
to the computer that is doing the extracting of the seconds from 'detailed
files', but I have no conclusive proof...

It seems like you may have to approximate unless someone else can provide
another solution.

Ken Ray
Sons of Thunder Software, Inc.
Web site: http://www.sonsothunder.com/
Email: kray at sonsothunder.com