Sean Madsen [Sun, 22 Apr 2018 23:06:46 +0000 (19:06 -0400)]
security/core#14 Add output encoding within js
Sean Madsen [Sun, 22 Apr 2018 21:50:47 +0000 (17:50 -0400)]
security/core#14 Validate "context" inputs
When "context" is passed as a GET parameter, ensure that its values is
a valid "Alphanumeric" type. This helps prevent XSS when the "context"
value finds its way into templates that lack HTML output encoding.
Replace...
CRM_Utils_Request::retrieve\((['"])context\1,(\s*)(['"])String\3
...with...
CRM_Utils_Request::retrieve\($1context$1,$3Alphanumeric$3
Also search for the following and manually fix:
\$_GET\[(['"])context\1\]
\$_POST\[(['"])context\1\]
\$_REQUEST\[(['"])context\1\]
Eileen McNaughton [Wed, 18 Jul 2018 21:54:15 +0000 (09:54 +1200)]
Merge pull request #12492 from eileenmcnaughton/msg_fix
dev/core/#/258 Fix message update routine.
Eileen McNaughton [Wed, 18 Jul 2018 12:57:58 +0000 (00:57 +1200)]
Merge pull request #12502 from eileenmcnaughton/no_pay
Prevent financial transactions from being saved with no payment instr…
eileen [Tue, 17 Jul 2018 21:33:42 +0000 (09:33 +1200)]
Prevent financial transactions from being saved with no payment instrument
eileen [Tue, 17 Jul 2018 07:59:27 +0000 (19:59 +1200)]
Fix message update routine.
Turns out this was seeming to pass but not because the test wasn't running.
Seamus Lee [Tue, 10 Jul 2018 20:36:35 +0000 (06:36 +1000)]
Merge pull request #12447 from mattwire/allcontactexport_regression
dev/core#249 Fix crash when performing export all contacts from search
Matthew Wire [Mon, 9 Jul 2018 22:38:10 +0000 (23:38 +0100)]
Fix crash when performing export all contacts from search
Seamus Lee [Fri, 6 Jul 2018 05:46:41 +0000 (15:46 +1000)]
Merge pull request #12417 from aydun/core_124_template_upgrade
Follow up to PR 12160 to update the DB version of the templates
Seamus Lee [Fri, 6 Jul 2018 05:46:13 +0000 (15:46 +1000)]
Merge pull request #12426 from totten/master-container-cache
Change default for CIVICRM_CONTAINER_CACHE to simplify admin/developer experience
Seamus Lee [Fri, 6 Jul 2018 01:13:05 +0000 (11:13 +1000)]
Merge pull request #12425 from seamuslee001/5_3_alpha1_fix
Ensure db updates are run for 5.3.alpha1
Tim Otten [Thu, 5 Jul 2018 21:12:39 +0000 (14:12 -0700)]
Change default for CIVICRM_CONTAINER_CACHE to simplify admin/developer experience
In discussion of cache-related PRs for Civi 5.3, there were a few
reports/issues from developers observing `ServiceNotFoundException`. This
is because there's not much awareness about how service-definitions are
cached. It shouldn't be a significant issue for production systems running
vanilla code, but for admins and developers (who juggle patches/branches),
it can cause confusion/support-issues/false-failures. This PR aims to
reduce those.
(This is a follow-up/substitute for #12401.)
Before
------
* The default value of `CIVICRM_CONTAINER_CACHE` is `always`.
After
-----
* The default value of `CIVICRM_CONTAINER_CACHE` is `auto`.
Technical Details
-----------------
The Symfony container essentially stores a list of "services". In some
Symfony-based apps, building the list of services can be time consuming, so
it's common to cache this.
In Civi, this cache physically appears as
`files/civicrm/templates_c/CachedCiviContainer.*.php`. The constant
`CIVICRM_CONTAINER_CACHE` determines how Civi manages the cache. There are
three available policies:
* `never`: This means we never use the cache. You never have to worry about
staleness. But it means we have to fire `hook_civicrm_container` on every
page-request, and we go through any container-compilation tasks.
This would have the fewest support-issues for devs and advanced admins.
* `always`: This means we always use whatever is in the cache. It never
(automatically) rebuilds the cache or checks freshness... if you make
change, then you must flush the cache explicitly. This should be the
fastest in production.
* `auto`: This means we typically use the cache (avoiding
hooks/recompilation), but it has to `stat()` a half-dozen files to check
the modification time. (To wit: if the timestamp on
`Civi/Core/Container.php` changes, then we discard the cache.)
Since performance is a consideration, I did some light benchmarking on my
laptop (fetching a basic public Civi page 100 times across 3 threads).
https://gist.github.com/totten/
ec4bffd723afb7967aec56a3040b9ca3
In these results, the `never` policy appears to be ~15-20ms slower than
`auto` or `always`. `auto` is only ~2ms slower than `always`.
The other consideration is accuracy -- `auto` will usually re-compile if you
make a change, but there are some edge-cases where you must still flush
manually. (In particular -- when you first implement
`hook_civicrm_container` in a new extension, it might not be aware of the
new extension. And extensions need to call `$container->addResource()`.)
However, overall, `auto` is a pretty good compromise that's almost as fast
you can get and works out-of-the-box for many dev/admin scenarios.
Seamus Lee [Thu, 5 Jul 2018 21:24:27 +0000 (07:24 +1000)]
Ensure db updates are run for 5.3.alpha1
Aidan Saunders [Wed, 4 Jul 2018 14:06:17 +0000 (15:06 +0100)]
Follow up to PR 12160 to update the DB version of the templates
CiviCRM [Wed, 4 Jul 2018 19:53:35 +0000 (19:53 +0000)]
Set version to 5.4.beta1
colemanw [Wed, 4 Jul 2018 14:35:37 +0000 (10:35 -0400)]
Merge pull request #11998 from compucorp/PCHR-3410-add-default-assignee-when-creating-cases
dev/core#107: Add default assignee when creating cases
René Olivo [Tue, 22 May 2018 20:51:23 +0000 (16:51 -0400)]
dev/core#107 Refactor selection of default assignee by relationship type
René Olivo [Thu, 15 Mar 2018 16:00:05 +0000 (12:00 -0400)]
dev/core#107 Automatically add default assignees when creating new cases
This feature allows users to define default assignees for each activity in a case type.
When creating a new case contacts are assigned to activities automatically by following
one of these rules:
* By relationship to case client
* The user creating the case
* A specific contact
* None (default)
Eileen McNaughton [Wed, 4 Jul 2018 07:52:01 +0000 (19:52 +1200)]
Merge pull request #12411 from eileenmcnaughton/matt_opt
Fix sample custom data is_reserved=0
Eileen McNaughton [Wed, 4 Jul 2018 06:11:11 +0000 (18:11 +1200)]
Merge pull request #12403 from civicrm/5.3
5.3
Eileen McNaughton [Wed, 4 Jul 2018 06:10:52 +0000 (18:10 +1200)]
Merge pull request #12409 from eileenmcnaughton/paydata
dev/core/issues/211 Fix mis-allocation of financial transactions when editing payment method on a completed payment
Eileen McNaughton [Wed, 4 Jul 2018 06:10:35 +0000 (18:10 +1200)]
Add late changes
eileen [Tue, 3 Jul 2018 23:52:22 +0000 (11:52 +1200)]
Fix mis-allocation of financial transactions
Per
https://lab.civicrm.org/dev/core/issues/211 there seems to be an issue with assigning entity_financial_trxn
entries the wrong financial_trxn_id - I cannot find any recent change that would have caused this
(unless it's simply the greater use of the payment edit block) - it has been reported as working in
5.0.2.
Matthew Wire [Tue, 29 May 2018 21:00:39 +0000 (22:00 +0100)]
Fix sample custom data is_reserved=0
Eileen McNaughton [Wed, 4 Jul 2018 04:38:18 +0000 (16:38 +1200)]
Merge pull request #12233 from mattwire/optiongroup_form
dev/core#155 Improve Option Groups form
Seamus Lee [Wed, 4 Jul 2018 02:22:45 +0000 (12:22 +1000)]
Merge pull request #12296 from eileenmcnaughton/greeting
Use email_greeting where possible in receipts
Eileen McNaughton [Wed, 4 Jul 2018 02:14:08 +0000 (14:14 +1200)]
Merge pull request #12408 from eileenmcnaughton/location_id
Add location_id if missing CRM-20711.
Seamus Lee [Tue, 3 Jul 2018 23:15:04 +0000 (09:15 +1000)]
Merge pull request #12399 from eileenmcnaughton/jamie
dev/core/#/229 Fix fatal error on send test mail
eileen [Tue, 3 Jul 2018 22:29:40 +0000 (10:29 +1200)]
Add location_id if missing CRM-20711.
On a small number of installs it seems location_id is missing from the group_contact table. This adds if missing
Eileen McNaughton [Tue, 3 Jul 2018 22:24:06 +0000 (10:24 +1200)]
Merge pull request #12398 from agh1/5.3.0-releasenotes
5.3.0 release notes
Eileen McNaughton [Tue, 3 Jul 2018 22:20:13 +0000 (10:20 +1200)]
Merge pull request #12362 from totten/master-cache-session
(dev/core#174) Forms/Sessions - Store state via Civi::cache('session')
Eileen McNaughton [Tue, 3 Jul 2018 22:02:58 +0000 (10:02 +1200)]
Merge pull request #12391 from eileenmcnaughton/matt
Revert change to BillingBlock to support hidden fields.
eileen [Tue, 3 Jul 2018 02:21:30 +0000 (14:21 +1200)]
Fix fatal error on send test mail
Seamus Lee [Tue, 3 Jul 2018 07:53:27 +0000 (17:53 +1000)]
Merge pull request #12400 from seamuslee001/dev_core_227
dev/core#227 Fix issue where on some extened groups the multiple reco…
Eileen McNaughton [Tue, 3 Jul 2018 04:11:53 +0000 (16:11 +1200)]
Merge pull request #12394 from yashodha/dev-223
(dev/core#223) Expose prefix and suffix in reports
CiviCRM [Wed, 20 Jun 2018 06:00:44 +0000 (06:00 +0000)]
Add release notes for 5.2.2
Tim Otten [Fri, 8 Jun 2018 23:39:40 +0000 (23:39 +0000)]
Add release notes for 5.2.1
Andrew Hunt [Tue, 3 Jul 2018 02:15:20 +0000 (22:15 -0400)]
5.3.0 release notes: combined, organized, added contributors, boilerplate
Also removed redundant issues from 5.2.1 and 5.2.2
Andrew Hunt [Tue, 3 Jul 2018 01:50:06 +0000 (21:50 -0400)]
5.3.0 release notes: my half of the issues
Alice Frumin [Mon, 2 Jul 2018 20:37:32 +0000 (16:37 -0400)]
5.3.0 release notes: Alice part of release notes
Seamus Lee [Tue, 3 Jul 2018 02:24:09 +0000 (12:24 +1000)]
dev/core#227 Fix issue where on some extened groups the multiple records checkbox doesn't show this causes empty string to be passed in which was causing problems when working out if we needed to make changes to the custom value table
Eileen McNaughton [Mon, 2 Jul 2018 20:26:01 +0000 (08:26 +1200)]
Merge pull request #12393 from scardinius/core-222
dev/core#222 - Fix importing contacts with dedupe rule
Tim Otten [Mon, 2 Jul 2018 18:58:36 +0000 (11:58 -0700)]
Merge pull request #12389 from totten/master-psr16-xtest
(dev/core#174) Add test to prevent hidden interactions among caches
Andrew Hunt [Mon, 2 Jul 2018 17:37:55 +0000 (13:37 -0400)]
5.3.0 release notes: raw from script
yashodha [Mon, 2 Jul 2018 10:47:38 +0000 (16:17 +0530)]
(dev/core#223) Expose prefix and suffix in reports
Eileen McNaughton [Mon, 2 Jul 2018 09:33:41 +0000 (21:33 +1200)]
Merge pull request #12340 from eileenmcnaughton/merge_cleanup
Move DAO function to DAO class, call it from Merge class
scardinius [Mon, 2 Jul 2018 08:55:41 +0000 (10:55 +0200)]
dev/core#222 - Fix importing contacts with dedupe rule
Eileen McNaughton [Mon, 2 Jul 2018 08:36:49 +0000 (20:36 +1200)]
Merge pull request #12390 from yashodha/dev-220
(dev/core#220) State province/country doesn't show properly in the…
eileen [Mon, 2 Jul 2018 05:50:01 +0000 (17:50 +1200)]
Revert change to BillingBlock to support hidden fields.
Per discussion with Matt Wire this is actually unnecessary, if we define 'id' in the payment fields
metadata. This is a bit too much like spooky magic but it ... erm .. works
yashodha [Mon, 2 Jul 2018 05:32:51 +0000 (11:02 +0530)]
core/issues#220 - State province/country doesn't show properly in the history report
Tim Otten [Thu, 28 Jun 2018 03:19:58 +0000 (20:19 -0700)]
(dev/core#174) Forms/Sessions - Clear in roughly the same ways as before
In the past, if one clears the `civicrm_cache` table (i.e. by calling
`CRM_Core_Config::clearDBCache()` ==> `TRUNCATE TABLE civicrm_cache`), then it
has the effect of destroying any active sessions/forms.
Now, in allowing sessions to be stored elsewhere, we lose that side-effect.
If we want strictly equivalent behavior (from a business-logic perspective),
then we'd want the patch to go a bit further -- calling
Civi::cache('session')->clear() at the same time that it does
`clearDBCache()`.
This revision adds `clear()` calls to various spots discussed here:
* https://docs.google.com/spreadsheets/d/1FxuIvr2noelBvhu5eja9_ps3YUWnkmGhqijBO3gH8Po/edit?usp=sharing
* https://github.com/civicrm/civicrm-core/pull/12362#issuecomment-
400897782
Tim Otten [Thu, 29 Mar 2018 01:52:57 +0000 (18:52 -0700)]
(dev/core#174) Forms/Sessions - Store state via Civi::cache('session')
Overview
----------------------------------------
When using forms based on CiviQuickForm (specifically `CRM_Core_Controller`), `CRM_Core_Session`
stores form-state via `CRM_Core_BAO_Cache::storeSessionToCache` and `::restoreSessionFromCache`,
which in turn calls `CRM_Core_BAO_Cache::setItem()` and `::getItem()`.
However, using `CRM_Core_BAO_Cache::setItem()` and `::getItem()` means that all session state
**must** be written to MySQL. For dev/core#174, we seek the **option** to store via
Redis/Memcache.
Before
----------------------------------------
* (a) Form/session state is always stored via `CRM_Core_BAO_Cache::setItem()` and `civicrm_cache` table.
* (b) To ensure that old sessions are periodically purged, there is special purpose logic that accesses `civicrm_cache`
(roughly `delete where group_name=Sessions and created_date < now()-ttl`).
* (c) On Memcache/Redis-enabled systems, the cache server functions as an extra tier. The DB provides canonical storage for form/session state.
After
----------------------------------------
* (a) Form/session state is stored via `CRM_Utils_CacheInterface`.
* On a typical server, this defaults to `CRM_Utils_Cache_SqlGroup` and `civicrm_cache` table.
* (b) To ensure that old sessions are periodically purged, the call to `CRM_Utils_CacheInterface::set()` specifies a TTL.
* It is the responsibility of the cache driver to handle TTLs. With #12360, TTL's are supported in `ArrayCache`, `SqlGroup`, and `Redis`.
* (c) On Memcache/Redis-enabled systems, the cache server provides canonical storage for form/session state.
Tim Otten [Thu, 28 Jun 2018 17:35:55 +0000 (10:35 -0700)]
(dev/core#174) Add test to prevent hidden interactions among caches
Eileen McNaughton [Mon, 2 Jul 2018 03:13:27 +0000 (15:13 +1200)]
Merge pull request #12379 from totten/master-psr16-sqlgroup
(dev/core#174) Full PSR-16 compliance for SqlGroup
Eileen McNaughton [Mon, 2 Jul 2018 00:11:58 +0000 (12:11 +1200)]
Merge pull request #12386 from mattwire/IPN_NFC
NFC code cleanup for AuthNet, Paypal, PaypalPro IPNs
Monish Deb [Sun, 1 Jul 2018 18:56:54 +0000 (00:26 +0530)]
Merge pull request #12305 from eileenmcnaughton/dedupe_limit
Re-instate Dedupe limit functionality & fix select toggle functionality
Matthew Wire [Sun, 1 Jul 2018 18:53:27 +0000 (19:53 +0100)]
NFC code cleanup for AuthNet, Paypal, PaypalPro IPNs
Eileen McNaughton [Sun, 1 Jul 2018 07:39:31 +0000 (00:39 -0700)]
Merge pull request #12345 from MiyaNoctem/CRM-195-add-counts-to-contribution-sub-tabs
dev/core#195 Add Contribution Counts to Sub-tabs
Eileen McNaughton [Sun, 1 Jul 2018 07:19:01 +0000 (00:19 -0700)]
Merge pull request #12383 from civicrm/5.3
5.3 to master
Eileen McNaughton [Sun, 1 Jul 2018 07:17:47 +0000 (00:17 -0700)]
Merge pull request #12382 from eileenmcnaughton/jaap
Fix mis-reporting of false booleans in reports
eileen [Sun, 1 Jul 2018 07:10:12 +0000 (19:10 +1200)]
Fix mis-reporting of false booleans in reports
Pradeep Nayak [Sun, 1 Jul 2018 06:48:30 +0000 (12:18 +0530)]
dev/financial/issues/24, Add unit test for Link payment method to financial account when created using api (#12365)
* Added api test to create new payment method
* updated api test
Eileen McNaughton [Sun, 1 Jul 2018 06:39:59 +0000 (23:39 -0700)]
Merge pull request #12356 from pradpnayak/issue-202
dev/core/issues/202, check if default currency is set
Eileen McNaughton [Sun, 1 Jul 2018 05:07:25 +0000 (22:07 -0700)]
Merge pull request #12381 from totten/master-psr16-apc
(dev/core#174) APCcache - Updates to comply with PSR-16
Eileen McNaughton [Sun, 1 Jul 2018 04:50:33 +0000 (21:50 -0700)]
Merge pull request #12380 from totten/master-psr16-memcache
(dev/core#174) Memcache(d) - Updates to comply with PSR-16
Seamus Lee [Sat, 30 Jun 2018 23:26:09 +0000 (09:26 +1000)]
Merge pull request #12378 from eileenmcnaughton/totten
(dev/core#174) Full PSR-16 compliance for ArrayCache, Redis
Tim Otten [Wed, 27 Jun 2018 04:49:54 +0000 (21:49 -0700)]
(dev/core#174) APCcache - Updates to comply with PSR-16
One noteable quirk is that APC retains expired (per TTL) records until the
following page-request. To comply with the test suite, we have to double-check
the expiration time.
Tim Otten [Tue, 26 Jun 2018 23:10:27 +0000 (16:10 -0700)]
(dev/core#174) Memcache(d) - Updates to comply with PSR-16
There are two drivers, `CRM_Utils_Memcache` and `CRM_Utils_Memcached`. It's
nice to update them in tandem (with similar design decisions). If an admin
admin is experimenting/debugging, this consistency makes it easier to switch
between drivers. (Cache data written by one driver can be read by the other
driver.)
In addition to the standard PSR-16-style changes, there are a couple changes
in how data is formatted when written to memcache:
* To allow support for targetted `flush()`ing (one prefix at a time), we update
the naming convention per https://github.com/memcached/memcached/wiki/ProgrammingTricks#deleting-by-namespace
This means that a typical key includes a bucket-revision code:
* BEFORE: `<site-prefix>/<bucket-prefix>/<item-key>` (`dmaster/default/mykey`)
* AFTER: `<site-prefix>/<bucket-prefix>/<bucket-revision>/<item-key>` (`dmaster/default/
5b33011fea555/mykey`)
* Values are `serialize()`d. This resolves an ambiguity where `Memcache::get()`
does not let us know if it returns `FALSE` because there's an error because
that's the stored value. By serializing, those scenarios can be distinguished.
* `get(...) === FALSE` means "item was not found"
* `get(...) === serialize(FALSE)` means "item was found with value FALSE"
Tim Otten [Mon, 25 Jun 2018 23:20:16 +0000 (16:20 -0700)]
(dev/core#174) CRM_Core_BAO_Cache - Periodically flush expired records
Tim Otten [Thu, 21 Jun 2018 01:15:34 +0000 (18:15 -0700)]
(dev/core#174) SqlGroup - Rewrite with PSR-16 and TTL support. Bypass BAO multitier cache.
Tim Otten [Sat, 23 Jun 2018 02:22:34 +0000 (19:22 -0700)]
(dev/core#174) Redis - Updates to comply with PSR-16
Tim Otten [Thu, 21 Jun 2018 00:35:14 +0000 (17:35 -0700)]
(dev/core#174) ArrayCache - Updates to comply with PSR-16
colemanw [Sat, 30 Jun 2018 16:42:42 +0000 (12:42 -0400)]
Merge pull request #12236 from agileware/CIVICRM-874
CiviCase, added reference of activities from different timelines.
Eileen McNaughton [Sat, 30 Jun 2018 02:52:21 +0000 (19:52 -0700)]
Merge pull request #12376 from totten/master-psr16-util
(dev/core#174) Add various utilities to support PSR-16
Tim Otten [Fri, 29 Jun 2018 23:31:11 +0000 (16:31 -0700)]
(dev/core#174) NaiveMultipleTrait - Fix validation of oddball iterables
Tim Otten [Tue, 26 Jun 2018 23:08:50 +0000 (16:08 -0700)]
(dev/core#174) CRM_Utils_Date - Add helpers for handling PSR-16 TTL's
Tim Otten [Thu, 21 Jun 2018 00:51:55 +0000 (17:51 -0700)]
(dev/core#174) Implement CRM_Utils_Cache::assertValidKey
Tim Otten [Thu, 21 Jun 2018 00:33:27 +0000 (17:33 -0700)]
(dev/core#174) NaiveMultipleTrait - Throw errors for bad inputs
Tim Otten [Thu, 21 Jun 2018 00:07:39 +0000 (17:07 -0700)]
(dev/core#174) Add E2E_Cache_CacheTestCase
Tim Otten [Thu, 21 Jun 2018 00:05:43 +0000 (17:05 -0700)]
(dev/core#174) Add concrete classes for cache exceptions
PSR-16 specifies cache exceptions using interfaces. This requires that we
implement the interfaces.
Eileen McNaughton [Fri, 29 Jun 2018 21:20:17 +0000 (14:20 -0700)]
Merge pull request #12368 from totten/master-cache-schema
(dev/core#174) civicrm_cache - Finish transition from DATETIME to TIMESTAMP
Eileen McNaughton [Fri, 29 Jun 2018 18:51:23 +0000 (11:51 -0700)]
Merge pull request #12374 from yashodha/report-fixes
core/issues#96 - Expose source column in booking report
yashodha [Fri, 29 Jun 2018 09:38:47 +0000 (15:08 +0530)]
core/issues#96 - Expose source column in booking report
Eileen McNaughton [Thu, 28 Jun 2018 16:15:14 +0000 (09:15 -0700)]
Merge pull request #12342 from totten/master-psr16-min
Caching - Comply with PSR-16 interfaces
Monish Deb [Thu, 28 Jun 2018 07:31:50 +0000 (13:01 +0530)]
Merge pull request #12355 from mattwire/CRM-21574_disable_contact_email_tellafriend_nfc
CRM-21574 NFC code cleanup of CRM_Friend_BAO_Friend class
Tim Otten [Tue, 19 Jun 2018 23:54:25 +0000 (16:54 -0700)]
(dev/core#174) Update top-level interfaces to show PSR-16 support
Tim Otten [Tue, 19 Jun 2018 23:52:47 +0000 (16:52 -0700)]
composer.json - Import PSR-16 interfaces (psr/simple-cache ~1.0.1)
Tim Otten [Tue, 19 Jun 2018 23:43:21 +0000 (16:43 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface - Add `has()` (PSR-16)
Tim Otten [Tue, 19 Jun 2018 23:22:28 +0000 (16:22 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface::clear() is an alias for flush()
flush() is specified by Civi's traditional interface, and clear() is
specified by PSR-16. The semantics should be the same (i.e. remove
any records managed by this cache-object).
This treats flush() and clear() as aliases.
Tim Otten [Tue, 19 Jun 2018 23:18:48 +0000 (16:18 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface::delete() - Return bool like PSR-16
The function signature for `delete()` did not specify a return value, and
actual behaviors varied depending on the driver (`bool` for memcache,
`int` for redis, and `void` for others).
This standardizes on returning a bool -- which makes the signature align
with PSR-16's `delete()`.
Tim Otten [Tue, 19 Jun 2018 23:06:26 +0000 (16:06 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface::flush() - Return bool like PSR-16
The function signature for `flush()` did not specify a return value, and
actual behaviors varied depending on the driver (`bool` for memcache,
`int` for redis, and `void` for others).
This standardizes on returning a bool -- which makes the signature align
with PSR-16's `clear()`.
Tim Otten [Tue, 19 Jun 2018 22:53:46 +0000 (15:53 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface - Add getMultiple/setMultiple (PSR-16)
Tim Otten [Tue, 19 Jun 2018 22:19:01 +0000 (15:19 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface::set() should match PSR-16
Comparing `CRM_Utils_Cache_Interface::set()` and `Psr\SimpleCache\CacheInterface::set()`,
they agree on key details:
1. The first value is a string key.
2. The second value is a mixed field (string or number or array...).
3. The return value should be a boolean indicating success/failure.
There are two differences:
1. Our old interface didn't actually enforce the return-value in all cases. Should be better now.
2. PSR-16 accepts a third argument, `$ttl`, which defaults to `NULL`. In the NULL/default case,
the semantics are the same (i.e. up to the discretion of the driver).
1. Since no existing callers actually use this, we can (for moment) throw an
error if someone tries to pass `$ttl` to a driver that doesn't support it.
3. The `$value` is pass-by-reference. However, this probably doesn't mean anything.
1. In older Civi code, there was a tendenacy to pass-by-reference without any real
need -- perhaps just speculation about performance.
2. Most cache drivers serialize the value and send it elsewhere. References will
not survive the process. If there's some code/caller/use-case that requires the
references to work, then it's already broken for most drivers (except maybe `ArrayCache`).
3. The old signature meant that callers had to say `$bar=123; $cache->set('foo', $bar)` rather than
`$cache->set('foo', 123)`. This change relaxes the contract -- so that you can pass
either `$bar` or `123.
See also: https://www.php-fig.org/psr/psr-16/
Tim Otten [Tue, 19 Jun 2018 22:09:08 +0000 (15:09 -0700)]
(dev/core#174) CRM_Utils_Cache_Interface::get() should match PSR-16
Comparing `CRM_Utils_Cache_Interface::get()` and `Psr\SimpleCache\CacheInterface::get()`,
they agree on key details:
1. They return values are `mixed` (strings, arrays, numbers,, etc)
2. The default representation of a cache-miss is NULL.
They differ in that PSR-16 allows the caller to optionally specify a `$default`.
Since no existing callers actually use this, we can (for moment) throw an
error if someone tries to pass `$default` to a driver that doesn't support
it.
See also: https://www.php-fig.org/psr/psr-16/
Seamus Lee [Wed, 27 Jun 2018 21:15:44 +0000 (07:15 +1000)]
Merge pull request #12348 from totten/master-clean-keys
(dev/core#174) Cache-keys for CRM_Utils_Cache_* should avoid reserved chars
Tim Otten [Wed, 27 Jun 2018 20:20:36 +0000 (13:20 -0700)]
(dev/core#174) civicrm_cache - Mandate change from DATETIME to TIMESTAMP
v4.7.20 updated the definition of `created_date` and `expired_date` so that
new installs would default to TIMESTAMP instead of DATETIME. Status-checks
and DoctorWhen have been encouraging a transition, but it wasn't mandated,
and there was little urgency... because `expired_date` was ignored, and
adhoc TTLs on `created_date` had generally long windows. Now that we're
using `expired_date` in more important ways for 5.4 (dev/core#174), we want
to ensure that these values are handled precisely and consistently.
Before
------------------------------
* On new deployments, `created_date` and `expired_date` are TIMESTAMP.
* On upgraded deployments, `created_date` and `expired_date` may
be using DATETIME or TIMESTAMP (depending on the particular history
and administrative decisions).
After
------------------------------
* On all deployments, `created_date` and `expired_date` are TIMESTAMP.
Tim Otten [Tue, 26 Jun 2018 19:09:11 +0000 (12:09 -0700)]
Civi\Angular\Manager - Use PSR-16 compliant cache key
Tim Otten [Mon, 25 Jun 2018 23:19:41 +0000 (16:19 -0700)]
CRM_Contact_BAO_ContactType - Use PSR-16 compliant cache key
Tim Otten [Thu, 21 Jun 2018 23:19:37 +0000 (16:19 -0700)]
CRM_Core_BAO_CacheTest - Unit test for CRM_Core_BAO_Cache::cleanKey
Tim Otten [Thu, 21 Jun 2018 08:22:57 +0000 (01:22 -0700)]
CRM_ACL_BAO_ACL - Use PSR-16 compliant cache key