Hi Taylor, I don't understand the piece about needing to use a string data type? Here's what I do, assuming that your dates start out as a date and time like 2015-04-19 02:00, 2016-03-12 18:46, etc.
1) Create a myTime calc that is FLOAT([myDateTime]) - INT([myDateTime]). This gets only the time portion of the datetime.
2) Format the myTime calc to be h:mm ampm.
3) Put that in a view.
I am sorting by strings and its working perfectly fine, may be because I have formatted the string in the format 'HH:MM AM/PM' while you have formatted your strings as 'H:MM AM/PM'
'H:MM AM/PM' will show time-stamp as 6:01 AM
'HH:MM AM/PM' will show time-stamp as 06:01 AM
Also, possibly this HH and H may not be the case at all, but I have tima-stamps behaving crazy in Excel a lot of times because of formatting issues, so worth giving it a try.
And then sort your time-stamp in descending order by alphabets.
I hope that works for you.
The reason my time dimension was a string type was because I was told
earlier it was the only way to get rid of the random date that seems to be
obligatorily added to the time. It looked like this:
As you can see, I took up your advice of making a calculation which I named
Time of Offense. I may not have followed your instructions correctly,
however, as it seems to have just changed the date from the highest
numerical date to the lowest. Here's my calculation:
I am not sure what you mean by a myTime calculation, assuming that's a
calculation type and not just an example name you've given it.
On Wed, Apr 20, 2016 at 1:59 AM, Ashish Singh <firstname.lastname@example.org>