Re^2: Date::Manip problem when creating arrays

aplogies - its a balance between editing the output to highlight the problem, and enough of the output to show that there is a problem. All the output appears fine, except the last days of March

2011032100:00:00
.....
 2011032600:00:00
 2011032700:00:00
 2011032801:00:00
 2011032901:00:00
 2011033001:00:00 
it goes from 2011-03-27 00:00:00
to  
2011-03-28 01:00:00
same on 29th and same on 30th 
and misses out completely 31st
[download]

Comment on Re^2: Date::Manip problem when creating arrays Download Code

Replies are listed 'Best First'.
Re^3: Date::Manip problem when creating arrays by Corion (Patriarch) on Aug 02, 2011 at 06:43 UTC
Have you read the documentation on Date::Manip::Recur? It explains how the recurrence pattern works. If you use it as you did, you get exact distances, and it seems that, somewhat naturally, Date::Manip declares a "day" to be "24 hours". It seems from the documentation that the following could work: `0:0:0:1*0:0:0` [download] ... but to be sure, I recommend you actually read the documentation, now that you are aware of the fact that not all "days" are 24 hours long.	[reply] [d/l]
Re^4: Date::Manip problem when creating arrays by SBECK (Chaplain) on Aug 02, 2011 at 15:58 UTC
I'll confirm that the recurrence (0:0:0:10:0:0) DOES work (assuming that what you are after is a list of ALL dates at time 00:00:00). Corion is correct about the source of the confusion... since a day is not always 24 hours long, applying a recurrence like this can give unexpected results. The only way to get around this is to make sure that the recurrence has an () in it which explicitly sets values in the date(s) returned.	[reply]