Q: finding distance between 2 time's

[martin at hvidberg.net]

On May 30, 4:10 pm, Steven D'Aprano <st... at REMOVE-THIS-
cybersource.com.au> wrote:
> On Sat, 30 May 2009 12:06:55 +0200, jkv wrote:
> > I added a few lines to your script, and now it ought to only print files
> > newer than 3601 seconds (3600 seconds is one hour).
> ...
> >     #if file newer than one hour print a line
> >     if time_difference < 3601:
>
> That's a potential off-by-one error. That may print files that are older
> than one hour. Admittedly, they'll be off by less than one second, but if
> you're going to write code, write correct code. The right test is:
>
>     if time_difference <= 3600:
>
> (and you may even want to deal with files that have a *negative* time
> difference, e.g. they were created apparently in the future).
>
> This is particularly necessary if you use time.time() to generate the
> current time, since that returns fractions of a second. But even if you
> don't, it's still the right thing to do: it's defensive programming.
> Rather than assume that all file systems store timestamps accurate only
> to a second, assume that some file system, somewhere, will be accurate to
> fractions of a second, and code accordingly.
>
> That way, no matter what the file system does, your code will still do
> the right thing, and (in this case) it doesn't even cost you anything.
>
> --
> Steven

Hi Steve thanks for your comment.
But I only need it to re-run some test scripts in case there seems to
be hick-up in our live download of satellite images. It's not critical
within seconds, not even within minutes...
But in principel the potential error could of cause be important - in
other context.

:-) M