Subject: Re: kernel filesystem code move
To: Luke Mewburn <lukem@netbsd.org>
From: Andrew Brown <atatat@atatdot.net>
List: tech-kern
Date: 12/30/2002 17:37:59
>  | >  how about a slightly different marking: "DEAD" vs "dead". with
>  | > "DEAD" meaning consider dead for checkout but not for diff. CVS
>  | > hack required, but only on the server side (I think, it's been a
>  | > while since I've used this). Patch attached.
>  | 
>  | That's more or less what I suggested, and I think it is a good
>  | idea. It allows diffs and imports to continue to work without breaking
>  | date based checkouts.
>
>Would this require that anoncvs & cvsweb repository mirrors need to
>run the "netbsd modified cvs(1)" in this case to function correctly?
>If so, we need to consider the ramifications of this, even if it's
>just "you need to run a netbsd modified cvs(1)" ...

off the top of my head i'd say "yes" and "no".

yes, the anoncvs mirrors would almost certainly need a modified cvs
binary.

no, i think cvsweb actually grovels the repository by using rcs
commands directly on the ,v files.  it would probably need a modified
rcs co binary.  :)

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
werdna@squooshy.com       * "information is power -- share the wealth."