Is the iOS Do Not Disturb Bug Causing this Bug?

Question

I am trying to find out if the recent Do Not Disturb bug (Ars Technica link) is affecting my code. I am trying to get the date and timestamp for midnight the previous Monday. Since I am writing this on a Monday, I would expect to get today's date (January 7, 2013), yet I am January 4, 2013.

I am following the guide posted by Apple, but I am trying to modify it to find the previous Monday, instead of Sunday.

+(NSTimeInterval)StartOfWeekMonday {
    NSTimeInterval finalTime = -1;

    NSInteger weekday;

    NSDate *monday  = 0;

    NSCalendar *greg = [[NSCalendar alloc] initWithCalendarIdentifier:NSGregorianCalendar];

    NSDateComponents *weekdays = [greg components:NSWeekdayCalendarUnit fromDate:[NSDate date]];
    NSDateComponents *subtract = [[NSDateComponents alloc] init];
    NSDateComponents *final    = 0;

    weekday = ( weekdays.weekday == 1 ) ? 6 : weekdays.weekday;

    [subtract setDay:(0 - weekday - 1)];

    monday = [greg dateByAddingComponents:subtract toDate:[NSDate date] options:0];
    final  = [greg components:(NSYearCalendarUnit|NSMonthCalendarUnit|NSDayCalendarUnit) fromDate:monday];
    monday = [greg dateFromComponents:final];

#ifdef DEBUG
    NSLog( @"%s - %d -> Weekday: %d", __FILE__, __LINE__, weekday );
    NSLog( @"%s - %d -> Monday: %@", __FILE__, __LINE__, [monday descriptionWithLocale:[NSLocale currentLocale]] );
#endif

    finalTime = [monday timeIntervalSince1970];


    return finalTime;
}

My log out put below, the weekday is correct (I am writing this on a Monday), yet the date is obviously wrong, it should be Monday: Monday, January 7, 2013, 12:00 AM

2013-01-07 15:07:33.792 MY-APP[5524:14c03] Weekday: 2
2013-01-07 15:07:36.757 MY-APP[5524:14c03] Monday: Friday, January 4, 2013, 12:00:00 AM Eastern Standard Time

This is being in the simulator right now, but all my other date calculations are giving the expected values.

Is it likely that I am being affected by the recent Do Not Disturb Bug and will have to wait for tomorrow (January 8, 2013) to see the expected results, or am I missing something entirely different?

That article is probably not correct about the DND bug; the discrepancy beetween the ISO calendar and Gregorian calendar that the explanation relies on doesn't exist: http://www.tuaw.com/2013/01/03/a-possible-explanation-for-the-ios-new-years-do-not-disturb-bug/ (see Update second paragraph). Have you tried doing your calculations from a dummy date, or moving your system clock around? That should give you some clues. — jscs, Jan 07 '13 at 20:26
@JoshCaswell Heh, these are the consequences of spreading FUD... *"The bug could be related to the ISO week calculation, or it might not; however the working out in this article is definitely flawed in several ways. The blogger responsible has been taken out the back and shot."* — jtbandes, Jan 07 '13 at 21:34

score 1 · Accepted Answer · answered Jan 07 '13 at 22:20

The issue was with how to calculate the subtraction. Either I completely mis understood the example provided by Apple, or it was not clear enough. The solution for calculating the subtraction was as follows:

weekday = ( weekdays.weekday == 1 ) ? 8 : weekdays.weekday;
toSubtract = ( 2 - weekday );

[subtract setDay:toSubtract];

Running various tests produced the proper "Monday." In the end, this had nothing to do with the date bug affecting Do Not Disturb.

Is the iOS Do Not Disturb Bug Causing this Bug?

1 Answers1