Re: [Monotone-devel] "branch" Parameter in get_netsync_write_permit()?

[Richard Levitte - VMS Whacker]

In message <address@hidden> on Thu, 8 Sep 2005 15:04:37 -0700 (PDT), Logan 
Sackette <address@hidden> said:

lsackette> My question is this, why did you guys remove the
lsackette> branch/collection parameter
lsackette> get_netsync_write_permit()?  I can see the reason for
lsackette> changing collection to branch for the *read*().  But I
lsackette> think both read and write should take the same
lsackette> parameters.

I have the same gripes, but have been told that the way netsync
currently works, it doesn't really matter, because there's no way for
the server to tell the client what branches are OK, so the client can
still throw a lot of data on the server.  Also, because consistency
requires that a database should contain all ancestors of any revision,
and a propagation can be done from any branch to any branch in any
potential participating database, there may still be revisions sent
over that belong to a different branch.

Personally, I think that it's OK to send over anything, but that
branch certs with branch names that aren't permitted according to the
lua hooks should simply be discarded.  The result is that you may end
up with revisions not belonging to any branch, until someone who's
authorised happens to send over some branch cert that gets applied to
"branch-less" revisions...

Cheers,
Richard

-----
Please consider sponsoring my work on free software.
See http://www.free.lp.se/sponsoring.html for details.

-- 
Richard Levitte                         address@hidden
                                        http://richard.levitte.org/

"When I became a man I put away childish things, including
 the fear of childishness and the desire to be very grown up."
                                                -- C.S. Lewis