What about PREV? (was Re: Summary: URL rev proposals)

[Scott Collins <sc...@ScottCollins.net>]

I don't recall seeing anyone mention that

   http://host/.../repos@PREV/x/y/z

   http://host/.../repos/x/y@PREV/z

   http://host/.../repos/x/y/z@PREV

May all logically and intuitively refer to valid and different 
resources, at least if my limited understanding (gained with |svn diff 
-rPREV| on various paths in a working copy constructed to test this 
theory) is correct.  I think this only applies to 'PREV', but that 
doesn't stop it from impacting the whole of the problem.
__________
Scott Collins <http://ScottCollins.net/>

Re: What about PREV? (was Re: Summary: URL rev proposals)

[cm...@collab.net]

Scott Collins <sc...@ScottCollins.net> writes:

> I don't recall seeing anyone mention that
> 
>    http://host/.../repos@PREV/x/y/z
> 
>    http://host/.../repos/x/y@PREV/z
> 
>    http://host/.../repos/x/y/z@PREV
> 
> May all logically and intuitively refer to valid and different
> resources, at least if my limited understanding (gained with |svn diff
> -rPREV| on various paths in a working copy constructed to test this
> theory) is correct.  I think this only applies to 'PREV', but that
> doesn't stop it from impacting the whole of the problem.

I would apply also to COMMITTED (which is logical, since COMMITTED is
PREV + 1).  Interesting point you raise.  I'm not sure if the folks
who are dragging this topic on were intending support for all the
magic revision names, but this is definitely food for thought if they
were.


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