changes from rc4 to rc5

2 views
Skip to first unread message

Wichert Akkerman

unread,
Feb 12, 2009, 1:00:00 PM2/12/09
to pylons-...@googlegroups.com
Is there a summary of the changes between rc4 and rc5?

Wichert.

--
Wichert Akkerman <wic...@wiggy.net> It is simple to make things.
http://www.wiggy.net/ It is hard to make things simple.

Ben Bangert

unread,
Feb 12, 2009, 1:20:10 PM2/12/09
to pylons-...@googlegroups.com
On Feb 12, 2009, at 10:00 AM, Wichert Akkerman wrote:

> Is there a summary of the changes between rc4 and rc5?

- beaker_cache decorator is fixed so that it works again with
arbitrary functions, and uses the proper function name
- Beaker has been updated since the last release to fix several bugs
involving memcached, and how it stores session data
- WebOb has gotten several updates to increase efficiency in header
handling and instantiation
- Controller inspect is cached for increased speed
- Other minor tweaks for efficiency (in Routes, and Pylons)

That's really about it. It's only an RC instead of final because we
wanted to be sure the small bug fixes and performance tweaks didn't
cause any new breakage.

Cheers,
Ben

Wichert Akkerman

unread,
Feb 12, 2009, 2:49:55 PM2/12/09
to Ben Bangert, pylons-...@googlegroups.com
Previously Ben Bangert wrote:
> - Beaker has been updated since the last release to fix several bugs
> involving memcached, and how it stores session data

Does that mean that it is now possible to use memcache for beaker
sessions?

Ben Bangert

unread,
Feb 12, 2009, 4:11:54 PM2/12/09
to Wichert Akkerman, pylons-...@googlegroups.com
On Feb 12, 2009, at 11:49 AM, Wichert Akkerman wrote:

> Does that mean that it is now possible to use memcache for beaker
> sessions?

It was possible to use them before, except due to the way they stored
the session keys in a variety of memcached keys, it was easy to have a
session get corrupted when memcached wipes one or two keys. Now it
stores the entire session as a single key so its either there, or not.
Obviously under enough users, if memcached isn't set high enough, this
is expected behavior, rather than sessions just losing bits and parts
of them.

Cheers,
Ben

Reply all
Reply to author
Forward
0 new messages