1705 Commits

Author SHA1 Message Date
Loz Calver
e829ad78b1
Merge pull request #8429 from sminnee/fix-2950
FIX: Throw deprecation on limit=0
2018-10-08 15:11:52 +02:00
Robbie Averill
c4788803ee FIX Remove unused cacheData prop from #8451 2018-10-08 12:05:29 +02:00
Maxime Rainville
36b1066413
Merge pull request #8451 from creative-commoners/pulls/4.3/memory-cache-findtemplate
NEW Add memory cache to ThemeResourceLoader::findTemplate()
2018-10-08 10:38:07 +13:00
Sam Minnee
e5d3b28a4d FIX: Don’t break validation on selects without a source.
Fixes cases where there are no valid values or an empty-string
value is manually added rather than using setEmptyString()

Fixes #4849
Fixes #7159
2018-10-06 11:53:17 +13:00
Sam Minnée
7915fe0c0f
Merge pull request #8303 from hchokshi/pulls/4.2/viewable-data-customised-isset
Implement __isset for ViewableData_Customised
2018-10-06 09:37:19 +13:00
Robbie Averill
98568262f2 Fixed phpcs violations 2018-10-05 16:07:33 +02:00
Robbie Averill
024762f03b Move default themes lookup before cache check 2018-10-05 15:00:32 +02:00
Robbie Averill
da9301f241 Use SilverStripe cache API instead of memory cache, add tests and fix cache config indentation 2018-10-05 14:58:48 +02:00
Robbie Averill
172d8915b7 Clear cache between ThemeResourceLoader tests 2018-10-05 14:12:26 +02:00
Robbie Averill
9ea7b58a8f NEW Add memory cache to ThemeResourceLoader::findTemplate()
For large sites, this reduces the number of file_exists, Path::join, getThemePaths etc
calls and can contribute to 26% reduction in TTFB load time
2018-10-05 14:12:26 +02:00
Sam Minnee
0fc06e51e5 FIX: Drop seconds from DBDatetime::Nice() to restore SS3 behaviour.
Note that the medium date format depends on locale, with en_NZ being
resolutely numeric. I’ve updated the test to verify a couple of locales
to make this more obvious.

Fixes #8121
2018-10-04 14:51:24 +13:00
Sam Minnee
638e6ec281 FIX: Throw deprecation notice on limit=0
The SS4 behaviour of limit=0 is unlikely to be the SS5 behaviour.
To clear the limit limit=null is recommended.

In addition, there’s a bit tighter type maintenance in the internal
limit data (ensure things are int).

Fixes #2950.
2018-10-04 10:38:15 +13:00
Robbie Averill
ee24413c30 Merge branch '4.2' into 4 2018-10-03 15:28:05 +02:00
Robbie Averill
b0871ebd75
Merge pull request #8417 from creative-commoners/pulls/4.3/many-many-through-sorting-tests
Add more list sorting tests and fix many many through sorting
2018-10-03 13:30:20 +02:00
Robbie Averill
d2b646cc13
Merge pull request #8286 from wilr/patches/mysql-create-table-issue
Check database has table before altering.
2018-10-03 13:24:59 +02:00
Robbie Averill
d74fed554a
Merge pull request #8427 from sminnee/fix-1397
FIX: Append any fields that don’t match name in insertBefore/insertAfter
2018-10-03 13:22:23 +02:00
Robbie Averill
6a5fb4999c
Merge pull request #8416 from lekoala/patch-14
raw2att accepts array, but sprintf doesn't
2018-10-03 13:21:43 +02:00
Robbie Averill
7f685b0bea
Merge pull request #8423 from creative-commoners/pulls/4.2/default-locale-for-members
FIX New members now receive the configured default locale, not the current locale
2018-10-03 13:18:05 +02:00
Robbie Averill
79f9d6b733
Merge pull request #8428 from sminnee/fix-2636
FIX: Don’t change state in ArrayList::getIterator()
2018-10-03 13:11:39 +02:00
Raissa North
90afb2037a NEW TabSet react component is no longer structural 2018-10-03 16:36:17 +13:00
Robbie Averill
cb2b9498fb API Deprecate updateSearchContextCallback and updateSearchFormCallback 2018-10-03 13:47:40 +13:00
Luke Edwards
bf60a699a5 Add extension points for ModelAdmin search API 2018-10-03 13:47:40 +13:00
Sam Minnee
0cc72c91ad FIX: Use DELETE FROM instead of TRUNCATE for clearTable
clearTable is mainly used for clearing data between tests. In this case,
there are very few or zero records, and DELETE FROM is quicker than
TRUNCATE, which works by deleting and recreating the table.

This materially speeds up test execution, at least on MySQL.

Cherry-pick of SS3 ae9ab22a8ff1b48c90f7dfe2899c09efaa65b161
2018-10-03 13:40:00 +13:00
Guy Marriott
5970fc2417
FIX Moving test to correct director 2018-10-03 10:47:17 +13:00
Robbie Averill
5a20bb5e20 Revert "FIX: Use DELETE FROM instead of TRUNCATE for clearTable"
This reverts commit 79c2b5ad427f4e95c8fb51b46c4ba31cdf2997c1.
2018-10-02 14:35:10 +02:00
Robbie Averill
c8e7672ede
Merge pull request #8414 from creative-commoners/pulls/4/searching-for-a-name-conflict
FIX Avoid having search fields with the same names as form elements
2018-10-02 13:43:59 +02:00
Robbie Averill
2ca089bcce
Merge pull request #8433 from sminnee/faster-cleartable
FIX: Use DELETE FROM instead of TRUNCATE for clearTable
2018-10-02 10:43:54 +02:00
Sam Minnee
79c2b5ad42 FIX: Use DELETE FROM instead of TRUNCATE for clearTable
clearTable is mainly used for clearing data between tests. In this case,
there are very few or zero records, and DELETE FROM is quicker than
TRUNCATE, which works by deleting and recreating the table.

This materially speeds up test execution, at least on MySQL.

Implemented in SS3 at 815da76b056a716c8831d22bbf93528912cbcb28
2018-10-02 18:00:38 +13:00
Sam Minnee
f2cbc1dfbb FIX: Don’t use USE_FRM in MySQL repair. Fixes #6300.
USE_FRM is a sufficiently dangerous option that it should require
manual DBA activity to carry out.

See http://dev.mysql.com/doc/refman/5.7/en/repair-table.html for more
info.
2018-10-02 10:48:47 +13:00
Sam Minnee
71dad5f685 FIX: Append any fields that don’t match name in insertBefore/insertAfter
Previous behaviour was to return false, which has been described as
a confusing bug on https://github.com/silverstripe/silverstripe-framework/issues/1397
where the issue was identified.
2018-10-02 10:39:48 +13:00
Sam Minnee
ad87890b2e FIX: Don’t change state in ArrayList::getIterator()
This prevents the map-to-ArrayData conversion from changing object
state and making the result of toArray() non-deterministic.

Fixes #2636. Other solutions were suggested on that ticket, but there is
no way of putting special code in for a SSViewer-specific iterator, and
having object state pre-emptively converted to ArrayData would
be an SS5-level change.
2018-10-02 09:56:09 +13:00
Maxime Rainville
755907d117
Merge pull request #8409 from open-sausages/pulls/4/treedropdown-caching
MINOR Add caching to the TreeDropdownField tree action
2018-10-01 13:56:21 +13:00
Guy Marriott
c503dbd634
Prefixing search filters for the React search form context 2018-10-01 13:38:32 +13:00
Maxime Rainville
13c89d0ed0 Add a comment and only apply caching if using the standard NumChildrenMethod 2018-10-01 11:29:51 +13:00
Harsh Chokshi
60dd5f288b Implement __isset for ViewableData_Customised
Resolves #8302.
2018-10-01 09:40:12 +13:00
Robbie Averill
231d6d9a9f FIX New members now receive the configured default locale, not the current locale 2018-09-28 16:25:10 +02:00
Robbie Averill
ac1fe5e9d5 FIX joinClass's default_sort is used when nothing else has been set already 2018-09-28 15:39:24 +02:00
Robbie Averill
4d14e9b6b1
Merge pull request #8421 from creative-commoners/pulls/4.3/psr-5-deprecations
Update deprecation PHPDocs to be PSR-5 compliant
2018-09-28 14:18:54 +02:00
Loz Calver
10ff6c4f81
Merge pull request #8418 from creative-commoners/pulls/4.3/lower-complexity
MINOR Reduce some code complexity, update array syntax and injected SQLSelect etc
2018-09-28 12:51:08 +02:00
Robbie Averill
e4768e44b0 Use splat over call_user_func_array and reduce calls to func_get_args() 2018-09-28 12:32:20 +02:00
Robbie Averill
e57edcd060
Merge pull request #8387 from sminnee/fix-start-session-without-headers
FIX: Ensure existing session can be accessed if headers_sent()
2018-09-28 10:59:29 +02:00
Robbie Averill
f842ee2eec Update deprecation PHPDocs to be PSR-5 compliant
See: https://github.com/php-fig/fig-standards/blob/master/proposed/phpdoc-tags.md#55-deprecated
2018-09-28 10:49:14 +02:00
Loz Calver
aec562f048
Merge pull request #8419 from 3Dgoo/patch-1
minor i18nEntityProvider.php API documentation fix
2018-09-28 10:21:37 +02:00
Sam Minnee
b98c87a6c5 FIX: Ensure existing session can be accessed if headers_sent()
If a session already exists, and Session::start() isn’t called until
after a large enough block of content is output, then headers_sent()
will be false. The previous code prevented the session from being
started in this case. That might makes sense for the creation of a new
session, but it prevent legitimate access to an existing session.

This mostly manifested when running debugging tools such as showqueries,
which may output content before the session is started.
2018-09-28 13:25:13 +12:00
3Dgoo
785b0c69da
minor i18nEntityProvider.php API documentation fix
Fixing documentation typo.
2018-09-28 10:28:00 +10:00
Ingo Schommer
25759ffc5f NEW Show file path on PHP parser exceptions 2018-09-28 08:58:47 +12:00
Robbie Averill
1b0293d8c5 Replace array syntax, use some strict comparison and remove unnecessary elseif 2018-09-27 17:37:37 +02:00
Robbie Averill
594fa30b82 Remove redundant elseif complexity and switch to short array syntax 2018-09-27 17:29:52 +02:00
Robbie Averill
adb4d1f92d MINOR Reduce some code complexity, update array syntax and injected SQLSelect etc 2018-09-27 16:40:23 +02:00
Thomas Portelange
fdd5acff5d
raw2att accepts array, but sprintf doesn't
it's not very likely to happen (it did in my case :-) ) but if the value is an array, sprintf will fail (because raw2att accepts array, but sprintf doesn't). i suggest to json encode any array data to ensure it's safely included in the html. Or we should throw proper exceptions to make sure invalid values do not result in a php error.
2018-09-27 12:12:59 +02:00