Matju writes:

# > What about Java?
# 
# java.util.Date (1.0) does it.
# 
# "In all cases, arguments given to methods for ....

Thanks for the info.

# > It's intended use for enterprise applications
# 
# You mean Ruby/Perl aren't intended for enterprise applications?

I mean that Ruby/Perl were not originally designed (AFAIK) with a specific 
goal of suitability for enterprise applications as such. (Which might or 
might not make any difference in how carefully dates are handled; hence my 
question about Java. Sometimes such details matter; sometimes they don't.)

Just so there is no misunderstanding, I think that language-wise, the rise 
of Ruby should turn out to be the best thing for enterprise applications 
(among lots of other types of applications) since the rise of Java. (Maybe 
Perl should be included with Java here because of its very widespread use 
behind the scenes as infrastructure glue and so on.)

... To follow up on related notes, it seems like no separate check is made 
to distinguish when leap seconds are specified in error versus when they 
are legitimate. OTOH, since AFAIK, no other major relevant language 
handles better, it's probably not worthwhile (relative to other things 
that could be done for Ruby) to clean it up this "possible misfeature" (as 
Matz put it), which might in retrospect be more favorably characterized as 
a "pragmatic quirk", which seems to violate the principle of least 
surprise (for most typical non-fully-informed people), but not the 
principle of least meta-surprise (for people fully-informed about a given 
issue).

Conrad Schneiker
(This note is unofficial and subject to improvement without notice.)