[RFC] Proof-of-Concept suggesting correct remote break-lock url (bug #250451)

[Wouter van Heyst]

Trying to send this again, hopes it works now.

There are lots of things wrong with the attached patch, but it succeeds
in not spewing the wait_lock notice on stdout (serverside), and
suggesting an url to the user that can actually work with break-lock.

Afaics:

 * the detection of wether we're running remote or not is very iffy
 * it requires a new protocol version
 * the message format is different (misses some things now)

Related, the wait_lock lock polling doesn't work if on remote branches,
it does locally.


What do people think, totally wrong approach, or salvagable?

Wouter van Heyst
-------------- next part --------------
A non-text attachment was scrubbed...
Name: break-lock.patch
Type: text/x-diff
Size: 10215 bytes
Desc: not available
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20090704/92a9aa58/attachment.bin