This module provides partial-update, lockless sessions as well as additional named storage space that may optionally haev an expiration date. Both sessions and named data stores are implemented as a basic key-value store, where the key must be a string but the value can be any pickle-able python object.
Download the project source and run python setup.py install to install. This project is not entirely python 3.X compatible until PyMongo gains 3.X compatibility.
This distribution is supported and tested with python 2.6 and 3.1.
Depends on:
- CherryPy 3 - the web framework that this module is designed for
- (optional) PyMongo - for connectivity to a mongodb database for storage
Set the following in your CherryPy 3 config file:
tools.lg_slates.on: True
tools.lg_slates.session_timeout: 120 #Use a 2-hour timeout instead of
#default 1hr
tools.lg_slates.storage_type: 'pymongo' #For using mongodb
tools.lg_slates.storage_conf: {
'host': None #pymongo host address (None for localhost)
,'port': None #pymongo host port (None for default)
,'db': 'test' #pymongo db to connect to
,'collection': 'slates' #pymongo collection to store slates in
}
Then make sure that you call import lg_slates at some point in your python code before engine.start(), and you should be good to go.
If you'd like to make use of a named slate, for instance to associate user data with an account, do the above setup, and then do something like this in your code:
import lg_slates
s = lg_slates.Slate(cherrypy.session['username'])
s['permanentData'] = True
val = s.pop('permanentData')
The default behavior for named slates (non-sessions) is to never expire. However, either a second argument may be passed to Slate.__init__, or tools.lg_slates.timeout may be set to the desired timeout in minutes.
Run python setup.py test. Depends on the unittest module.