Re: Revision Dates Really Inclusive?

Max Bowsher wrote:

> The problem is that svn interprets that as "the revision that was HEAD
> at midnight on date1" to "the revision that was HEAD at midnight on
> date2". Which is not a very intuitive interpretation. I think our date
> range matching needs redesigning.

That certainly explains the strange behavior I see. I've already found
that being more specific by including time doesn't help. I would much
rather see the date range matching redesigned than the book corrected.
I'm trying to use the log output for inclusion on customer account
service records and right now I must always compare prior records to
determine what has been billed and what has not when I should be able to
just rely on the date range.

I haven't look at the svn code at all, but understand much of it is in
Python. If this date range logic is in Python code and this change
isn't major, I might be willing to tackle the patch if someone can point
me in the right direction.

-- 
John Florian
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org