Apache::SessionX - An extented persistence framework for session data

Apache::SessionX extents Apache::Session. It was initialy written to use Apache::Session from inside of HTML::Embperl, but is seems to be usefull outside of Embperl as well, so here is it as standalone module.

Apache::Session is a persistence framework which is particularly useful for tracking session data between httpd requests. Apache::Session is designed to work with Apache and mod_perl, but it should work under CGI and other web servers, and it also works outside of a web server altogether.

Apache::Session consists of five components: the interface, the object store, the lock manager, the ID generator, and the serializer. The interface is defined in SessionX.pm, which is meant to be easily subclassed. The object store can be the filesystem, a Berkeley DB, a MySQL DB, an Oracle DB, or a Postgres DB. Locking is done by lock files, semaphores, or the locking capabilities of MySQL and Postgres. Serialization is done via Storable, and optionally ASCII-fied via MIME or pack(). ID numbers are generated via MD5. The reader is encouraged to extend these capabilities to meet his own requirements.

The interface to Apache::SessionX is very simple: tie a hash to the desired class and use the hash as normal. The constructor takes two optional arguments. The first argument is the desired session ID number, or undef for a new session. The second argument is a hash of options that will be passed to the object store and locker classes.

lazy

By Specifing this attribute, you tell Apache::Session to not do any access to the object store, until the first read or write access to the tied hash. Otherwise the tie function will make sure the hash exist or creates a new one.

create_unknown

Setting this to one causes Apache::Session to create a new session with the given id (or a new id, depending on CWrecreate_id) when the specified session id does not exists. Otherwise it will die.

recreate_id

Setting this to one causes Apache::Session to create a new session id when the specified session id does not exists.

idfrom

instead of passing in a session id, you can pass in a string, from which Apache::SessionX generates the id in case it needs one. The main advantage from generating the id by yourself is, that in 'lazy' mode the id is only generated when the session is accessed.

newid

Setting this to one will cause Apache::SessionX to generate a new id every time the session is saved. If you call CWgetid or CWgetids it will return the new id that will be used to save the data.

config

Use predefiend config from Apache::SessionX::Config, which is configured at package installation time. On Debian, these may be altered by running CWdpkg-reconfigure libapache-sessionx-perl.

object_store

Specify the class for the object store. (The Apache::Session:: prefix is optional) Only for Apache::Session 1.00.

lock_manager

Specify the class for the lock manager. (The Apache::Session:: prefix is optional) Only for Apache::Session 1.00.

Store

Specify the class for the object store. (The Apache::Session::Store prefix is optional) Only for Apache::Session 1.5x.

Lock

Specify the class for the lock manager. (The Apache::Session::Lock prefix is optional) Only for Apache::Session 1.5x.

Generate

Specify the class for the id generator. (The Apache::Session::Generate prefix is optional) Only for Apache::Session 1.5x.

Serialize

Specify the class for the data serializer. (The Apache::Session::Serialize prefix is optional) Only for Apache::Session 1.5x.

Example using attrubtes to specfiy store and object classes instead of a derived class:

 use Apache::SessionX

 tie %session, 'Apache::SessionX', undef,
    { 
    object_store => 'DBIStore',
    lock_manager => 'SysVSemaphoreLocker',
    DataSource => 'dbi:Oracle:db' 
    };

NOTE: Apache::SessionX will CWrequire the nessecary additional perl modules for you.

setid ($id)

Set the session id for futher accesses.

setidfrom ($string)

Set the string that is passed to the generate function to compute the id.

getid

Get the session id. The difference to using CW$session{_session_id} is, that in lazy mode, getid will not create a new session id, if it doesn't exists.

getids ($init)

return the an array where the first element is the initial id, the second element is the current id and the third element is set to true, when the session data was modified. If the session was deleted, the initial id (first array value) will be set to '!DELETE'. If the optional parameter CW$init is set to true, getids will initialize the session (i.e. read from the store) when not already done.

cleanup

Writes any pending data, releases all locks and deletes all data from memory.

See documentation of Apache::Session for more informations about it's internals

Apache::SessionX::Generate::MD5

Apache::Session::Store::*

Apache::Session::Lock::*

Apache::Session::Serialize::*

Gerald Richter <richter@dev.ecos.de> is the current maintainer.

This class was written by Jeffrey Baker (jeffrey@kathyandjeffrey.net) but it is taken wholesale from a patch that Gerald Richter (richter@ecos.de) sent me against Apache::Session.