RE: [Pyatcron-devel-list] Some comments on Brian code

[Childers, Matthew]

> > >
> > > > Actually, they expect strings. My reasoning behind that was that
> > cron
> > > > accepts ranges (e.g. 2-5), lists (e.g. 1,2,3), step values (e.g.
> > */5)
> > > > and so on. So, are we not going to have those kinds of
> > functionality?
> > > > And what about the default value of '*'? If implemented as an
> > integer,
> > > > would we just use 0 for that?
> > >
> > > I guess it should be possible to "emulate" all that
functionalities
> > > using string arrays, isn't it ?
> > >
> >
> > That is a good point Brian.  I would imagine we would just need to
> > accept strings.  Not necessarily string arrays, since we should be
able
> > to set all the options for a value with one string.
> 
> Really, I don't see how it is harder to use int arrays instead of "*",
> or "2-5", as it is just a "human" representation of an array of
> integers, nothing more. For example, if you want to say "all the days
of
> the week except from the 5th one", you could write "*/5" or
> "1,2,3,4,6,7". For a computer program, it's easier to handle this
> information if it stored as an array containing {1,2,3,4,6,7} than in
a
> string containing "*/5", IMO at least.
> 

I see what you are saying now.  That would work, however, I have a
question more so with cron than how you are saying to do it.  If you
have a minute expressed as */10 that would be run every 10 minutes which
could be re-written (10,20,30,40,50,60).  Now if you have a minute
expressed as */45, will this actually run every 45 minutes, or only on
the 45th minute of the hour?  If it runs every 45 minutes, then I don't
know how you would express it in an array, as the minutes would change
each hour.  The first hour it would run on the 45th minute, the second
hour, on the 30th minute, the third on the 15th minute and so on.  I
looked around and couldn't find any information on this, so if it does
not actually run every 45 minutes, then Julien's way would work fine,
otherwise, we would have to look at something else.