Re: Statistics in r12501 - still a small problem

Carsten Johnsen wrote:

> The only issue I can find now is that Simons commit on March 31st
> 00:32:02 (monday) is counted with week 13 instead of week 14.
> Show stats of revisions 12502 to 12517 of TSVN trunk to verify.
> But this doesn't really bother me - just consider it to be an
> observation. I guess it has something to do with the fact that the
> commit is made within the first hour of March 31st, and with 12 hour
> settings the time would then be related to March 30th. (please note
> that my local settings is set to 24 hour).

I finally found the reason for this:
our function which calculates the week of the year uses the CTimeSpan
object for its calculation. We calculate the number of days since
january first (midnight). The docs for GetDays() state:

"Note that Daylight Savings Time can cause GetDays to return a
potentially surprising result. For example, when DST is in effect,
GetDays reports the number of days between April 1 and May 1 as 29, not
30, because one day in April is shortened by an hour and therefore does
not count as a complete day."

Now I need to find a way to work around that issue...

Stefan

-- 
        ___
   oo  // \\      "De Chelonian Mobile"
  (_,\/ \_/ \     TortoiseSVN
    \ \_/_\_/>    The coolest Interface to (Sub)Version Control
    /_/   \_\     http://tortoisesvn.net