Re: Removal of depth



Philip Van Hoof wrote:
This is a new version of the refactor patch that makes cancelled items's
callback and destroyer happen exactly the same as for non-cancelled
items.

This is a different strategy from the original, but avoids us from
having to do dirty tricks like calculating the mainloop's depth in order
to know whether or not to throw the execution to the mainloop.

Now the execution would happen in the queue's thread, so we always throw
it to the mainloop. And now we can also securely wait for it just like
any other normal non-cancelled operation.

So this patch *does* change the behaviour, unlike the previous version
which didn't change behaviour itself, just merged reusable code
together, so please test this thoroughly!

Looks nice to me. The code is now easier.

Br


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]