Re: GregorianCalendar doesn't work with large +ve/-ve time values

[Noa Resare]

On ons, 2004-10-13 at 11:02 +1000, Stephen Crawley wrote:
> Folks,
> 
> There has been a fair bit of discussion about Date-related bugs
> recently.  I'd like to draw your attention to the long-standing #9495
> bug against java.util.GregorianCalendar which fails when you try
> to convert a large +ve or -ve millisecond value to a Date.
> 
> There is a new(-ish) Mauve testcase that exercises GregorianCalendar
> millisecond conversion across its range of input values:
> 
>     gnu.testlet.java.util.GregorianCalendar.conversion.java
> 
> The root cause of the bug is that the conversion calculations are done
> using 'int' instead of 'long'.  For large millisecond values, some of
> the calculations are wrapping around.

I think this is on my  "mauve regressions to fix for kaffe 1.1.5"
hitlist so I will probably look at it later this week.

/noa