Add a finalizer SmartSubtransportRegistration

[nulltoken]

This would allow us to actually get rid of UnregisterSmartSubtransport

Cf. #854 (comment)

[nulltoken]

/cc @ethomson @jamill potential member of the stabilization gang if making any sense

[ethomson]

Would it? Here's what I wrote in #528 :

I don't think so; in this case, the Dispose method should be called explicitly by libgit2, and never by any caller. I think that decorating with IDisposable suggests to the caller that they could (or worse, should) dispose it when in fact they should not.

[nulltoken]

@ethomson I'm not sure I'm following you here. Actually, UnregisterSmartSubtransport has to be called by the client and not by libgit2.

However, I agree that exposing a Dispose() method may not be that useful. Indeed, there's very little chance this kind of method will be used in the context of a using statement.

How would you feel about adding a finalizer though?

[ethomson]

I think that makes sense.

On Thu, Apr 9, 2015 at 2:26 PM, nulltoken notifications@github.com wrote:

@ethomson https://github.com/ethomson I'm not sure I'm following you
here. Actually, UnregisterSmartSubtransport has to be called by the
client and not by libgit2.

However, I agree that exposing a Dispose() method may not be that useful.
Indeed, there's very little chance this kind of method will be used in the
context of a using statement.

How would you feel about adding a finalizer though?

—
Reply to this email directly or view it on GitHub
#1017 (comment)
.

[nulltoken]

Renamed and labeled.

[nulltoken]

/cc @ammeep I think that may also apply to FilterRegistration in #854