Re: JULDAY 5.4 not same as 5.3?

[Chris Torrence <chris(at)yang-sprouse.com>]

Hi Don,

The problem is with using an unsigned long integer. Inside JULDAY.PRO,
it subtracts 12 off of the hours before dividing by 24. This will cause
wrap-around for unsigned longs and unsigned long64s.

So you should be able to fix (no pun!) your problem by:

   in_jday = julday( umm, udd, uyear, FIX(uhr), umin, usec )

This bug has been logged with RSI, and will be fixed in the next IDL
version.

Cheers,
Chris Torrence
Research Systems, Inc.


Don Woodraska wrote:
> 
> Has anyone else noticed a bug in JULDAY that appeared in 5.4?
> 
> I tried this in IDL 5.3:
> 
> IDL> help,umm,udd,uyear,uhr,umin,usec
> UMM             LONG      =            2
> UDD             LONG      =           16
> UYEAR           LONG      =         2001
> UHR             ULONG     =            0
> UMIN            ULONG     =            0
> USEC            ULONG     =            0
> IDL> in_jday = julday( umm, udd, uyear, uhr, umin, usec )
> IDL> gps0_jday = julday(1,6,1980,0,0,0)
> 
...
> IDL> in_jday = julday( umm, udd, uyear, uhr, umin, usec )
> IDL> gps0_jday = julday(1,6,1980,0,0,0)
> IDL> help,in_jday,gps0_jday
> IN_JDAY         DOUBLE    =    1.8140893e+08
> GPS0_JDAY       DOUBLE    =        2444244.5
>