[BUG] Pull command on Windows: we have 2 problems

[John A Meinel]

Robert Collins wrote:
> On Thu, 2005-10-27 at 09:49 -0400, Aaron Bentley wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Robert Collins wrote:
>>> On Wed, 2005-10-26 at 16:32 -0400, Aaron Bentley wrote:
>>>
>>>
>>>> I see those as workarounds.  I'd add
>>>> 3) change Branch so that read-locked operations will work when Branch is
>>>> write-locked.
>>>
>>> This is already the case:
>> What I mean is, it should work whether you reuse the same Branch or
>> create a new copy of it.
> 
> I think thats a really bad idea.
> 
> It implies a process wide identity map for the separate lock at a
> minimum, and raises IMO hairy questions - for instance, if I open a
> write lock in one instance, create a new branch instance and take a read
> lock, then close the first instances write lock and grab a new one -
> should it succeed or fail?

The alternative is what we have now. Which is if you have one branch
with a write (or read) lock, and a second branch object is created, and
attempts to lock, you block.
This was what was happening in the "bzr pull" code, because somewhere it
looked back at the path, and created a new Branch object.

We can try and track down all of those places.

As far as your above case, the lock would be maintained in a write-lock
state, because it would never have been completely dropped.

So within that process, you would have maintained a write-lock the whole
time.

I'm not sure what you want to happen.

John
=:->


> 
> Rob
> 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 249 bytes
Desc: OpenPGP digital signature
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20051027/70e732a9/attachment.pgp