* Add ini setting redis.session.early_refresh to allow for session TTL updates on session start ( requires redis server version 6.2 or greater )
* Enable cluster session support for strict mode sessions ( via PS_VALIDATE_SID_FUNC )
* Cluster sessions used to write on every session, now we only write if the session has been modified.
* Send EXPIRE instead of SETEX if sessioh has not been changed
* If early refresh is enabled use GETEX for initial session read
* When strict sessions are enabled, check whether the session exists first, validate sid and regenerate if necessary
This commit removes support for PHP 5 by getting rid of all of our Zend
compatibility layer methods, as well as any call checking against
PHP_MAJOR_VERSION or ZEND_MODULE_API_NO.
Unit tests are all passing for Redis, RedisCluster, and RedisArray but
this should still be considered a work in progress until more testing
can be done.
Addresses issue #1448
Newer session handler API(PS_MOD_UPDATE_TIMESTAMP) supports 2 more callbacks:
* validate_sid() is used by session.use_strict_mode, which provides better security.
* update_timestamp() is used by session.lazy_write, which can improve performance in some situations.
* Use a precalculated SHA1 of our lock release script and first
attempt to EVALSHA the script while falling back to EVAL if that
fails. This means that the very first time we attempt to release
a lock after server restart or script cache will fail but subsequent
attempts will succeed as Redis will cache the script automatically.
* Reorganized the new locking prototypes by making them static and removing
them from the header file as they never need to be called from outside
of redis_session.c anyway. Did the same for the lock_secret structure
as we don't need to expose the structure externally.
* Consolidated allocation and deallocation of lock pointers such that our
redis_pool "constructor" and "desctructor" handle that as well.
* Updating how we release the lock means we no longer need the new REDIS_G
module globals, so those were removed.
* HOST_NAME_MAX doesn't exist on OSX so added some preprocessor logic around that
Still a WIP as it needs more testing as I'm sure I broke *something* :-)
