Re: svn commit: rev 5949 - in trunk: doc/book/book subversion/clients/cmdline subversion/include subversion/libsvn_wc

[Greg Stein <gs...@lyra.org>]

On Thu, May 15, 2003 at 11:20:59AM -0500, sussman@tigris.org wrote:
>...
> Mauve Bikeshed: show a "!" for incomplete dirs, just like missing
> items, so that users realize that 'svn up' needs to be run.
> 
> [ People are still arguming, but let's try this out for a little
>   while.  If we come up with a different consensus, or if this causes
>   disaster/confusion, we can always change it. ]

I think Jay's points about overloading this are *very* good. We should
choose a different symbol.

I have no opinion on *which* symbol, but overloading is Badness, IMO.

>...
> * svn_wc.h (enum svn_wc_status_kind): new 'incomplete' status enum.

To me, the fact that we can best model "incomplete" as a status leads me to
think it is a first-column marker, rather than sharing a column with the
"switched" state as Brane suggested.

>...
> * book/ch03.xml, ch08.xml: update book's description of '!', and what
>   to do when you see it.

I don't think you should have done this, given the discussion. Assuming we
remove the overloading of the symbol, this stuff will completely change.
*shrug*

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

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

Re: svn commit: rev 5949 - in trunk: doc/book/book subversion/clients/cmdline subversion/include subversion/libsvn_wc

[Branko Čibej <br...@xbc.nu>]

Greg Stein wrote:

>On Thu, May 15, 2003 at 11:20:59AM -0500, sussman@tigris.org wrote:
>  
>
>>...
>>* svn_wc.h (enum svn_wc_status_kind): new 'incomplete' status enum.
>>    
>>
>
>To me, the fact that we can best model "incomplete" as a status leads me to
>think it is a first-column marker, rather than sharing a column with the
>"switched" state as Brane suggested.
>

Ah, but is this really the best way we can model it? Consider Karls
suggestion in another thread about marking directories as conflicted. A
directory can be incomplete _and_ conflicted _and_ switched, all at the
same time. Incompleteness is orthogonal to modification That's why I
think we shouldn't overload the first-column marker for this.

-- 
Brane Čibej   <br...@xbc.nu>   http://www.xbc.nu/brane/


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