[p4] Efficiency of protect

Robert Cowham robert at vaccaperna.co.uk
Mon Feb 26 04:22:52 PST 2007


Hi Sushree

These sorts of things are a mixture of education so that people do the right
thing rather than the wrong thing, and enforcement (which is not always
easy). Equally some organisations will make certain actions a disciplinary
offence...

Another aspect of this is organising your codelines etc to make it really
clear to everyone what should go where, and also to explain why things are
being done (e.g. for export restrictions or licensing restrictions). People
usually act better when they understand the reason why.

In large organisations the communication issues can be significant!!

Note that the technical approach is to use triggers to control things more
finely, but there are probably going to be ways around this for the creative
developer, so I would back it up with the education etc.

Robert

> I have two groups say
> 
> bangalore_users
> newyork_users
> 
> I have the following protections:
> read group bangalore_users * //depot/main/...
> write group bangalore_users * //depot/main/jam/...
> read group bangalore_users * -//depot/main/dev/...
> wirte group newyork_users * //depot/main/...
> 
> But what if someone from newyork_users group branches the 
> files under //depot/main/dev/... to //depot/main/jam/dev/...
> 
> Then bangalore_users get write access to files under 
> //depot/main/dev/... which they should not have access.
> 
> I do understand that the above example can be stated as a 
> mischief but my point is that are the protections that 
> efficient w.r.t files/folders?
> 
> Also can we implement some triggers so that we restrict 
> branching paths by which files become visible to restricted users?
> 
> Please suggest....
> 
> 
> Regards,
> Sushree.


More information about the perforce-user mailing list