MathGroup Archive 2004

[Date Index] [Thread Index] [Author Index]

Search the Archive

RE: Re: Bug in FromDate

  • To: mathgroup at smc.vnet.net
  • Subject: [mg49143] RE: [mg49130] Re: [mg48903] Bug in FromDate
  • From: "Wolf, Hartmut" <Hartmut.Wolf at t-systems.com>
  • Date: Mon, 5 Jul 2004 04:54:22 -0400 (EDT)
  • Sender: owner-wri-mathgroup at wolfram.com

>-----Original Message-----
>From: Sébastien de Menten [mailto:sdementen at hotmail.com]
To: mathgroup at smc.vnet.net
>Sent: Friday, July 02, 2004 8:02 AM
>To: mathgroup at smc.vnet.net
>Subject: [mg49143] [mg49130] Re: [mg48903] Bug in FromDate
>
>
>>>>I am using FromDate with arguments outside the usual scope but don't
>>>>get any sensible result.
>>
>>Why should you?
>>
>
>Because sometimes it may be quite handy to be able to say: the date 20 
>months after the day d with just ToDate[FromDate[d + {0,20,0,0,0,0}]].
>
>In fact, this is the behaviour of the mktime function of the 
>standard C 
>library (allow any range in the different time fields year, 
>month,etc and 
>normalise them in mktime).
>
>
[...]

Handy, perhaps, but you should know what you're doing then! So what is 

    {2004,3,30,17,15,58} - 1 Month  ?
    {2004,3,30,17,15,58} + 1 Month  ?
    {2004,3,30,17,15,58} - 1 Month + 1 Month ?
    {2004,3,30,17,15,58} - 13 Month + 13 Month ?

The desired results seem to be dependend on culture, e.g. of banking account bankers, payroll payers, insurance police insurers -- and all that would be completely OK! (They have their answers.) But what a mess, if they were dependend on C-libraries, Perl-functions, etc...?  Mathematica hopefully wouldn'd line up there.

After all, it's easy to write an application function, but specify it first!

--
Hartmut Wolf


  • Prev by Date: Solid plot
  • Next by Date: Re: kelvin functions ker and kei
  • Previous by thread: Re: Bug in FromDate
  • Next by thread: Re: RE: Re: Bug in FromDate