ACLTransport and Other ACL Thoughts

[Alexander Belchenko]

I comment some of your thoughts. But I could be wrong in some aspects.

Jeff Abbott пишет:
> Folks,
> 
> In doing more thinking about ACLs in Bazaar, I was looking over the 
> ACLTransport spec (http://bazaar-vcs.org/Specs/ACLTransport) and digging 
> through source seeing what I could learn, and a thought occurred to me:
> 
>   Could this be done with a TransportDecorator?

I'm not expert in bzr+http, but may be (just may be as an idea) it's
better to implement as subclass (or maybe mixin) of RemoteTransport?
So it will be deeper pluggable into all bzr+*:// combinations?

> My inspiration came largely from ReadonlyTransportDecorator, except 
> instead of universally saying "no" to some operations it would check a 
> config file before doing so.  I was figuring that 
> ..bzr/branch/branch.conf could be used for this purpose with sections like:
> 
>   [read]
>   ** = *
> 
>   [write]
>   ** = alice
>   docs/* = bob, charlie
>   src/** = david
> 
>   [deny]
>   ** = edward
> 
> In that example, everyone except edward would have "pull" and "branch" 
> access, and the other people listed in the [write] section would have 
> "push" access to the respective portions of the branch (i.e. alice can 
> push to anywhere, bob and charlie can only push to the docs directory 
> but none of its subdirectories, and david can push to src and anywhere 
> below).

One problem here is on the server you don't touch actually working tree
for push/pull/branch/commit operations. So it seems like provide fine
grained access within working tree of one branch will be very hard.
Because now bzr works with branch as a single whole.

But splitting big project into subprojects will works fine. And it's easiest
path IMO. Then next step is using NestedTrees. Unfortunately bzr still
don't have them.

> I'm not sure if we'd need, or even want, that level of granularity, but 
> it seems like it'd be better to implement that now rather than do it 
> more coarsely and have to re-think things again in the future when 
> someone finds a good case for it being more precise.  I'm also not 
> convinced that .bzr/branch/branch.conf is the right place for this 
> information; perhaps .bzr/branch/access.conf would work better?

As I said above it's better to place on repository level and control
access to particular branches.

> I've started fiddling with some code but I was wondering what other 
> people thought of the idea of, say, ACLTransportDecorator, before I go 
> too far along.  If it turns out that doing it as a full-fledged 
> ACLTransport would be the better choice then, John, if you're around, 
> I'd be keen on working on that with you.  Otherwise... where do we go 
> from here?
> 
> Thanks,
> Jeff
> 
>