[MERGE/RFC] Add transport jail to bzr smart server; add ignore_fallbacks option to BzrDir.open_branch

[Robert Collins]

On Thu, 2009-03-26 at 09:17 +0100, Vincent Ladeuil wrote:
> 
> What I still don't get is why you use a per thread data instead
> of just a global one ? You have a use case where different jails
> are needed inside the same process ? Outside the test suite ?

The server is multithreaded. Each thread gets jailed separately.

> The other disturbing thing is having a *permanent* hook, why not
> just make it part of the appropriate class then ?

Hooks are a clear separation of policy and functionality; they allow
other servers to provide different checks at the same point, and for
entirely different uses to be found, much better than hard coding a
single server specific check (which would still need to be thread local)
inline-into the BzrDir.open code.

> Aren't we abusing the hook mechanisms here ?

Nope :).

-Rob
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20090326/c9a2d29a/attachment.pgp