[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: New log/blame -g option and authz rules

From: Hyrum K. Wright <hyrum_wright_at_mail.utexas.edu>
Date: 2007-08-28 22:15:00 CEST

Mark Phippard wrote:
> Hyrum,
> Was having an email discussion internally at CollabNet and the
> question was asked whether the new log -g and blame -g options respect
> authz rules? I imagine we have some rules/standards as to whether you
> are allowed to know the kind of details about a revision these options
> expose. So not sure if there is an issue here or not.
> I can file an issue in the issue tracker if you have either not
> thought about it, or would like a place to record an answer for
> posterity.
> Of course, I do not mean to direct this question to just you. Anyone
> is free to jump in with opinions or answers.

My feeling is that we should honor authz rules when tracing ancestry for
'log -g' and 'blame -g'. I believe that the current implementation does
do appropriate authz checking, but I haven't tested it thoroughly. We
should probably file an issue to ensure that this gets done before 1.5
is released. Additional tests are welcome.

Also, I'm in the process of reevaluating the way that we iterate over
merge history (think something like svn_repos__walk_ancestry() ), and
may have some additional ideas about this topic as I get deeper into it.
 If people have additional insight, I'm very much open to it.


To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Aug 28 22:12:16 2007

This is an archived mail posted to the Subversion Dev mailing list.