[rabbitmq-discuss] New Persister merged into default

[Matthew Sackman]

On Fri, Jul 23, 2010 at 09:19:58AM +0200, John Apps wrote:
> Congratulations to one and all!

Many thanks, for everyone's patience with this and for the early
adopters who bravely tested this code whilst it was in development and
helped us by reporting bugs as they found them.

> All in all, then, you only changed some 7679 lines of code? [?]

The final merge to default was just 5762 lines. However, large amounts
of refactorings and abstraction had already made it to default - for
example we now have an API for developing backing-queues against, which
(currently) allows you to select between the new persister and the old
(and allows others to experiment with writing further persisters), and
was originally developed as part of the new persister but the
infrastructure there made it onto default much sooner and has been part
of the last couple of releases. As a result, I've no idea really how to
work out the number of LoC that originated with the new persister which
have now made it to default.

> I am really looking forward to trying the new persister out, having had lots
> of fun and games with the old one...

Indeed! I'm looking forwards to your reports - I'm hoping you're going
to report it's suddenly become much much harder to make Rabbit crash ;)

Matthew