On Fri, Nov 7, 2014 at 10:06:26AM -0500, Tom Lane wrote:
> 5. A side issue that's probably pretty relevant for the OP's application
> is that not every country adopted the Gregorian calendar at the same time.
> Really if he wants any data integrity he's going to need to store an
> explicit indication of whether a date is in Julian or Gregorian calendar.
> (In a world of infinite manpower you could imagine extending the zic
> database to provide information about which calendar was used when in
> every region, but I don't see us going there.)
Oh, it is worse than the date of the switch to Gregorian --- Russia had
some really odd calendars:
http://novaonline.nvcc.edu/eli/evans/his241/Notes/Calendar.html
So, roughly speaking, in the nineteenth century, the Russian calendar
was 12 days behind the Western European calendar.
http://en.wikipedia.org/wiki/Soviet_calendar
The Soviet calendar added five- and six-day work weeks between 1929 and
1940 to the Gregorian calendar adopted by Russia in 1918.
Many Russian events, even those in the 20th century, have multiple dates
for a single day/event, based on the calendar system used.
We could figure out a way to relax our code to allow for leap year
handling for pre-1582 dates, but the larger issue is that date
calculations would also be affected by the existance of such Julian-only leap
days, whether you stored those dates or not, meaning this change would
affect everyone, not just those who want to store such dates. For
example, should:
SELECT '1500-03-01'::date - '1500-02-28'::date;
return 1 or 2?
--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://enterprisedb.com
+ Everyone has their own god. +
