Mon 04 Jun 2012 11:11:51 AM UTC, comment #7:
In the revised patch (datevec-new.patch), I added support for the format 21,22, 25, 26 29 and 31.
I tested the code and I got the same results as in matlab R2010a (test_datevec.m).
However, formats 12,24,25 and 28 give quite strange (bogus?) result in matlab:
datevec('Mar00',12) % mmmyy
ans = 2012 2 29 0 0 0
datevec('01/03/2000',24) %dd/mm/yyyy,
ans = 2000 1 3 0 0 0
datevec('00/03/01',25) %yy/mm/dd
ans = 101 1 3 0 0 0
datevec('Mar2000',28) %mmmyyyy
ans = 2017 8 21 0 0 0
Can somebody with a newer version of matlab confirm this?
Format 24 gives the same results in octave 3.6.1 (with or without patch) than in matlab. However, day and month are reversed according to documentation.
(file #25981, file #25982)
|