Manipulating Old Dates

[Rob Cozens]

Several months ago I had a conversation with Ken Ray about the 
shortcomings of Transcript's date manipulation commands when applied 
to dates more than half a century old (or in the future).  I was 
reminded of the issue when Douglas McKay wrote to tell me Serendipity 
Library's validDate function did not work for the date he entered.

So I created a stack, a button, and a handler:

        on mouseUp
                ask "Enter a date:"
                convert it to dateItems
                put "The converted date is"&&it
        end mouseUp
I took a day, Feb. 8, and ran the handler several times, changing the 
year by a decade each time.

        "2/8/44" yields "1944,8,2...
        "2/8/34" yields "2034,8,2...
        "2/8/1934" yields "2034,8,2...
The first two results are predictable, as the centuryCutoff property 
defaults to 35.  The third result suggests to me that all date 
conversion uses seconds as an intermediate conversion format (and 
probably that is why one can add to the seconds, days, and months and 
get a correct date WITHIN THE RANGE OF THE CENTURY CUTOFF).

So just how does one work around this shortcoming, especially when 
dealing with current & past dates concurrently?  For example, I have 
a stack that charts biorhythms from any date based on the subject's 
dob.  Normally, biorhythms are charted from a current date a few 
months into the future; so one would expect the starting date to work 
with a centuryCutoff of 35 (until 2036).  But what of birth dates? 
Again, biorhythms are normally charted for living persons, not 
historical figures; but even so there are people alive today that 
could have been born in the 19th, 20th, or 21st centuries.

How does one deal with these issues given Transcript's date handling commands?

And isn't the centuryCutoff property setting developers up for "baby 
Y2K" conversion issues as the default centuryCutoff approaches?

When Ken & I spoke, we discussed the need for Julian date handlers in 
Transcript.  I have Julian date routines in my old FORTRAN & PL/1 
libraries.  Is it time they were incorporated into Serendipity 
Library?

Rob Cozens
CCW, Serendipity Software Company
http://www.oenolog.net/who.htm
"And I, which was two fooles, do so grow three;
Who are a little wise, the best fooles bee."
from "The Triple Foole" by John Donne (1572-1631)
_______________________________________________
use-revolution mailing list
[EMAIL PROTECTED]
http://lists.runrev.com/mailman/listinfo/use-revolution