Unfortunately, we can't simply use session_id() to regenerate the
session id in unit tests. Starting from PHP 7.2, it would trigger
"session_id(): Cannot change session id when headers already sent",
refer to MDL-60978 and PHP bug #75628 for more details.
As a workaround, we use a static property allowing us to inject the
value that we then use as a session identifier. This is reasonably
enough to make sure that the identifier is used as a part of the key
prefix.
It is now safe to cache a reference to a cache and expect consistent results.
Changing identifiers altered cache results where a reference was
held to the cache. Identifiers have been set to be cached with
identifiers included so the caches are separate.
As a consequence of this it was identified that invalidation events
and identifiers don't easily work together as an event can't determine
which identifiers should be used for cache invalidation. So invalidation
events have been made incompatible with identifiers being set. No core
code used this combination as it's not possible to understand any expected
behaviour.
Event invalidation for application and session caches was centralised to the same
location. The only difference was the name of the lastinvalidation variable. This
improves support and consistency of invalidation code.
purge_all() and purge_by_definition() look in the configuration
for which caches are available and then creates them to purge them.
The configuration stores the values used by initialise(), not
initialise_unit_test_instance() and would therefore fail to purge
all caches if they were not purged by another means.
In the case of filestore, it's purged by unit tests, in the case
of memcache(d), it purges the whole store when a single definition
is requested.
Therefore all configuration was moved into the configuration file
during unit tests and does not have any special override codes in
the unit test infrastructure.
When a cachable object is store in the static cache from
the backing store, it was incorrect serialised rather than
using the wake function. This has been resolved and tests added.
During the investigation into cacheable_object, it was discovered
that set_identifiers never removes identifiers when you call it,
so set_identifiers(array('a')) and set_identifiers(array('b')) really
resulted in array('a','b') as the identifiers rather than 'b'.
The fix for this issue depends on the set_identifiers fix and
they have been coupled together as a result.
When an alternative cache configuration is used, aka:
define('TEST_CACHE_USING_ALT_CACHE_CONFIG_PATH', true);
$CFG->altcacheconfigpath = '/tmp/xxxxx.php';
the get_expected_application_cache_store() method was not able of
retroffiting the default application store being used.
This commit just implements that functionality so, when
executing unit tests using alternative configuration, assertions
looking for that will pass without a problem.
This issue makes several fixes for the session loader and the session store.
* maxsize argument now works for session caches.
* fixed performance hole when interation occurs frequently.
* fixed cache purge bug occuring when multiple caches are defined before being used.
* improved lastaccess handling.
Big thanks to Marina who contributed the following commits:
* Always make sure the elements in cache are sorted so we need to remove only elements in the beginning of array
* Remove expired elements from session store to free memory
* Minor bug fixes
isset is much faster than array_key_exists. For caches we need to
ensure as much performance as possible. To all null to continue to
be stored and provide correct results, non ttl also now saves in
$this->store[$key][0]. As isset($this->store[$key]) will return
true in all cases.
time() as found in cache::now() can be slow on some platforms and
we shouldn't bother checking it unless needed. It also stops non-ttl
caches paying the ttl price.
Unit tests were adjusted to ensure null is still handled correctly
when we are only using isset.
Issues:
1. Fix setType calls
2. Changing from default to all and editing again, default is still selected.
3. Uncheck all options causes error.
4. Doesn't seem to be restricting based upon option.
5. Picked all and got coding error with the database thing.
6. Bad title: https://github.com/samhemelryk/moodle/commit/wip-MDL-37500-m25#L1R157
7. Amend comments - should only be removed once 2.5 is the minimum version for an upgrade.
8. Document the defaultsharing option.
Outcomes:
1. Fixed - copy paste error.
2. Fixed - mforms was applying the default value despite a value being provided. A quirk of elements with array names.
3. Fixed - validation now requires at least one option to be selected.
4. Fixed - issue rose from definitions not being re-parsed. cache/admin.php now reparses the first time a user visits the page.
5. Fixed - better purging of definitions when working with them anonymously. Unit test added.
6. Fixed - new string added and used.
7. Fixed - comments amended.
New issue to address parsing of definitions during upgrade.
New issue to add debugging notice if definition sets only one possible sharing option and that option is user input.
This change is a large change to the way sessions are handled
within MUC after it was discovered that session did not function
as expected when any store other than the default session store
was being used.
As part of this change the session loader has been largely
customised in order to consolidate session data for the loader.
The unit tests have also being greatly increased to provide
better coverage for sessions.
