lovely.session 0.3.0

memcache-based session storage

===========================
Session Data Using memcache
===========================

This package provides a session data manager which stores it's data in
memcache. The package uses lovely.memcached to store it's data.

IMPORTANT:

This test expects a memcache server running on local port 11211 which
is the default port for memcached.

This test runs in level 2 because it needs external resources to work. If you
want to run this test you need to use --all as parameter to your test.

Start a memcache instance with : memcached <optional options>


Once memcached is running, we can start testing:

>>> from zope import component
>>> from lovely.memcached.interfaces import IMemcachedClient
>>> from lovely.memcached.utility import MemcachedClient
>>> util = MemcachedClient()
>>> component.provideUtility(util, IMemcachedClient, name='session')
>>> util.invalidateAll()


Import the container we will use for caching tests.

>>> from lovely.session.memcached import MemCachedSessionDataContainer

Timeout behavior
~~~~~~~~~~~~~~~~

We need to test the timeout capability of the container. We can do this by simulating the passage of time while minimizing the timeout period.

Create a new session data that we will cause to timeout.

>>> timeoutSessionData = MemCachedSessionDataContainer()
>>> timeoutSessionData.cacheName = u'session'
>>> timeoutSessionData.__name__ = 'MemCacheSession'

>>> timeoutSession = timeoutSessionData['mySessionId']
>>> timeoutSession
{}

So we expect it is empty at this point. Get a new session from it.

>>> timeoutData = timeoutSession['myData1']
>>> timeoutData
{}

Okay, so now add some data to that session.

>>> timeoutData['info'] = 'stored in memcache'
>>> timeoutData
{'info': 'stored in memcache'}

Now get that sessionData from the session. It should just give it to us and the sessionData should have data.

>>> timeoutData = timeoutSession['myData1']
>>> timeoutData
{'info': 'stored in memcache'}

Now simulate the effect of a timeout by forcing one.

>>> timeoutSessionData.timeout=1
>>> timeoutSessionData.lastAccessTime=0

Now ask the sessionData for the session again. If the timeout worked, the session will be empty.

>>> timeoutSession = timeoutSessionData['mySessionId']
>>> timeoutSession
{}

Attempt to get the data from the session anyway and it will also be empty.

>>> timeoutData = timeoutSession['myData1']
>>> timeoutData
{}


Normal memcache access
~~~~~~~~~~~~~~~~~~~~~~

Now we create a memcache session and connect it to the memcached client.

>>> sessionData = MemCachedSessionDataContainer()
>>> sessionData.cacheName = u'session'

We need to provide a name for the session data manager because it is used to
identify the cache entry in memcache.

>>> sessionData.__name__ = 'MemCacheSession'

>>> session = sessionData['mySessionId']
>>> session
{}
>>> type(session)
<class 'lovely.session.memcached.MemCacheSessionData'>

We can now get data from the session.

>>> data = session['myData']
>>> data
{}
>>> type(data)
<class 'lovely.session.memcached.MemCachePkgData'>

>>> data['info'] = 'stored in memcache'
>>> data
{'info': 'stored in memcache'}


Transaction support
~~~~~~~~~~~~~~~~~~~

Because the MemCacheSession is transaction aware we need to commit the
transaction to store data in the memcache.

>>> import transaction

>>> transaction.commit()

If we now read session data it is read back from the memcache.

>>> session = sessionData['mySessionId']
>>> session['myData']
{'info': 'stored in memcache'}

>>> sessionData.items()
[('mySessionId', <lovely.session.memcached.DataManager object at ...>)]


MemCacheSession is now also savepoint aware, let's check how that works:

We first set some data:

>>> session = sessionData['mySessionId']
>>> data = session['myData']
>>> data['info'] = 'we want to keep this'

Set a savepoint:

>>> savepoint = transaction.savepoint()

Change the data:

>>> data['info'] = 'this should be dumped'

Rollback to the previous value:

>>> savepoint.rollback()

And here it is, the before value:

>>> data['info']
'we want to keep this'

Newly added data must also go away:

We add a new data:

>>> data['newinfo'] = 'go away'

And a new container:

>>> newdata = session['myNewData']
>>> newdata['foo'] = 'bar'

Roll it back to the previous savepoint:

>>> savepoint.rollback()

The data is gone:

>>> data['newinfo']
Traceback (most recent call last):
...
KeyError: 'newinfo'

The container is empty, because it gets always created on retrieval:

>>> session['myNewData']
{}

Let's see what happens on commit:

>>> transaction.commit()

If we now read session data it is read back from the memcache.

>>> session = sessionData['mySessionId']
>>> session['myData']
{'info': 'we want to keep this'}

The data is not present:

>>> data['newinfo']
Traceback (most recent call last):
...
KeyError: 'newinfo'

The container is empty, because it gets always created on retrieval:

>>> session['myNewData']
{}


=======
CHANGES
=======

0.3.0 (12-11-2009)
------------------

- Use zope.container instead of zope.app.container

- Removed unused dependencies


0.2.2 (2009-08-14)
------------------

- Add 'lastAccessTime' class attribute (fix old instances)


0.2.1 (2009-08-14)
------------------

- Correctly implemented timeout behavior of session data.

- Remove duplicate extras_require (python2.4 compatibility)


0.2.0 (2008-09-25)
------------------

- Made the DataManager savepoint aware.


0.1.4 (2008-07-31)
------------------

- Fixed ZCML to avoid deprecation warnings, since the session API was moved to
``zope.session``. *Sigh*


0.1.3 (2008-07-31)
------------------

- Fixed `setup.py` to be on par with the latest layout.

- Fixed deprecation warnings, since the session API was moved to
``zope.session``.


0.1.2 (2007-08-13)
------------------

- Move source to svn.zope.org.


0.1.1 (2007-08-13)
------------------

- Fixed dependency on `lovely.memcached`.


========
Download
========