I find most uses of array_filter(), array_reduce(), etc. to be
excessively clever, i.e. they are used to prove how smart the
developer is, at the expense of readability and performance. So I am
pleased to have a defensible reason to remove these instances, which
broke PHPStorm's type propagation.
Change-Id: I03dcd6c3c80f19f90e7b39448b5508713da63806
Done:
* Replace LanguageConverter::newConverter by LanguageConverterFactory::getLanguageConverter
* Remove LanguageConverter::newConverter from all subclasses
* Add LanguageConverterFactory integration tests which covers all languages by their code.
* Caching of LanguageConverters in factory
* Make all tests running (hope that's would be enough)
* Uncomment the deprecated functions.
* Rename FakeConverter to TrivialLanguageConverter
* Create ILanguageConverter to have shared ancestor
* Make the LanguageConverter class abstract.
* Create table with mapping between lang code and converter instead of using name convention
* ILanguageConverter @internal
* Clean up code
Change-Id: I0e4d77de0f44e18c19956a1ffd69d30e63cf51bf
Bug: T226833, T243332
Released just now.
Many old suppressions can now be removed. Enabling the issue for
undeclared variables is left to do later, given that there are
roughly 200 warning.
Change-Id: I99462a1e9232d6e75022912e2df82bc2038476ef
Scalar casts are still allowed (for now), because there's a huge amount
of false positives. Ditto for invalid array offsets.
Thoughts about the rest: luckily, many false positives with array offsets
have gone. Moreover, since *Internal issues are suppressed in the base
config, we can remove inline suppressions.
Unfortunately, there are a couple of new issues about array additions
with only false positives, because apparently they don't take
branches into account.
Change-Id: I5a3913c6e762f77bfdae55051a395fae95d1f841
Before this, authentication error messages are custom built and do not
contain as much information as block error messages for other actions.
They also assume the block target is either an IP or an IP range, and
have no customisation for different types of block.
Instead, this uses the BlockErrorFormatter to choose the most
appropriate and informative message for the block.
Bug: T227110
Change-Id: I942ac605075b6c2174682c7e75fe1213f82ebea2
This reverts commit 5f06efb318, which
reverted 9335363789, which makes
the deprecated property AbstractBlock::mReason private.
After 9335363789, AbstractBlock::mReason is obsolete, since the block
reason is now stored as a CommentStoreComment, AbstractBlock::reason.
Change-Id: Ica0a74be90383689ca8e4cfe6d0fb25c9a5942c5
This reverts commit 9335363789.
Reason for revert: It's full of code accessing AbstractBlock::mReason
out there, see [1]. Also, it was never hard deprecated. While that may
be acceptable under some circumstances, it's definitely not OK to remove
code when there are consumers around. I'd have fixed it right now without
reverting if it were a single repo, but there's just too many.
[1] - https://codesearch.wmflabs.org/search/?q=-%3EmReason&i=nope&files=&repos=
Change-Id: I8669f502b50cff89e28dada0f65fe2b130ae9b37
AbstractBlock::setReason now accepts a string, Message or
CommentStoreComment. The CommentStoreComment is accessed via
AbstractBlock::getReasonComment.
AbstractBlock::getReason returns the reason as a string, with
the language and format consistent with how block reasons were
built before this commit. This method is deprecated, since it
makes assumptions about the language and format needed. The
deprecated mReason property is no longer public.
Doing this (and T227005) will remove the implicit dependency of
BlockManager::getUserBlock on language, which causes a recursion
error if the block is checked before the user has loaded. It also
provides a mechanism for getting the block reason in a language
specified by the caller. (This does not apply to DatabaseBlock
reasons entered via the Special:Block form, which were not and
are still not translatable.)
This commit also updates authentication classes to return the
translated reason.
Bug: T227007
Change-Id: Iec36876e930dff96a256aebbdc39cbfb331c244e
Phan is clever enough to understand when the return value depends on the
name of the string passed in. This avoids lots of @phan-var annotations
in all subclasses.
Note that PHPCS rejects @template, ref T232256.
Change-Id: I7b7c7fae9c80320cad41bedc2972a3a5e22e19b0
This allows us to remove many suppressions for phan false positives.
Bug: T231636
Depends-On: I82a279e1f7b0fdefd3bb712e46c7d0665429d065
Change-Id: I5c251e9584a1ae9fb1577afcafb5001e0dcd41c7
Keep Block as a deprecated class alias for DatabaseBlock.
Update calls to the Block constructor and Block static
methods from external classes.
Also update documentation in several places that refer to
blocks as Blocks.
Bug: T222737
Change-Id: I6d96b63ca0a84bee19486471e0a16a53a79d768a
This introduces a minimal BlockManager service, for getting blocks
that apply to a User.
Move the part of User::getBlockedStatus that checks for the blocks
into BlockManager::getUserBlock, and move the related helper
methods from User to BlockManager.
Hard deprecate or remove these helper methods, and move to private
methods in the BlockManager:
* User::getBlockFromCookieValue
* User::isLocallyBlockedProxy
* User::inDnsBlacklist
Soft deprecate these helper methods, and move to public methods in
the BlockManager:
* User::isDnsBlacklisted
Add tests to cover the methods moved to BlockManager.
Bug: T219441
Change-Id: I0af658d71288376735cebe541215383b56bb72e5
The method User::isBlocked() attempts to answer two questions:
(1) Does the user have a block?
(2) Is the user prevented from performing this action?
The method can answer #1, but it cannot answer #2. Since User::getBlock() can
also answer #1, this method is redundant. The method cannot answer #2 because
there is not enough context in order to answer that question.
If access is being checked against a Title object, all access checks can be
performed with PermissionManager:userCan() which will also check the user's
blocks.
If performing all access checks is not desirable, using
PermissionManager::isBlockedFrom() is also acceptable for only checking if the
user is blocked. This method does *not* determine if the action is allowed,
only that the user's block applies to that Title.
If access is being checked without an existing Title, User::getBlock() can be
used to get the user's block. Then Block::appliesToRight() can be used to
determine if the block applies explicitly to a right (or returns null if
it is unknown or false if explicitly allowed). If the user is creating a new
Title, but the text of the title is not yet known (as in the case of Wikibase),
access should be checked with Block::appliesToNamespace().
Bug: T209004
Change-Id: Ic0ad1b92e957797fee8dcd00bd1092fe69fa58f1
Its return value was already not used. By using runWithoutAbort()
it means that if a hook accidentally returns false, it will throw
an exception instead of silently skipping other hook handlers.
This type of mistake has repeatedly caused problems in WMF prod
over the years.
Change-Id: I6af2261957eb20f18fa5f0d50c4c4835d3850cfe
Use getters and setters for $mReason, $mTimestamp, $mExpiry and
$mHideName; use Block::getType to check if a block is an autoblock
instead of checking $mAuto; no change needed for $mParentBlockId,
which is not accessed externally.
Change-Id: I767ed44ce4c2e21f53962d75fb86891add2282f6
Password policy checks that fail and have `suggestChangeOnLogin` set to true will
prompt for a password change on login.
Below are some rules that apply to this setting in different scenarios:
- If only one policy fails and has `suggestChangeOnLogin = false`, a password change will
not be requested
- If more than one policy fails and one or more have `suggestChangeOnLogin` set to true`,
a password change will be requested
- If `forceChange` is present in any of the failing policies, `suggestChangeOnLogin` value
will be ignored and password change will be enforced
- if $wgInvalidPasswordReset is set to false `suggestChangeOnLogin` is ignored
IMPORTANT**
Before this patch, suggesting a password change was the default behavior (depending on
$wgInvalidPasswordReset), which means that the necessary changes to $wgPasswordPolicy
need to be in place before this patch is merged and gets to production.
Bug: T211621
Change-Id: I7a4a0a06273fa4e8bd0da3dac54cf5a1b78bb3fd
AuthManager::autoCreateUser() causes createAndPromote.php to give error
"Automatic account creation is not allowed." when
$wgGroupPermissions['*']['createaccount']=false is set. Anonymous user
checks should be skipped for maintenance scripts.
Change-Id: Ib61889a758e542abe991707d8b7853a25cfed8e9
So far, our key derivation code assumed that it has control over
the salt used by the derivation routines, however I want to add Argon2
support and it doesn't work this way: password_hash() generates the
salt itself, and the only way to verify a password is by using
password_verify(). Current way the things are done doesn't support it
because it relies on the result of password hashing with parameters we
provide to be deterministic.
Therefore, I'm deprecating Password::equals(), as well as whole concept
of comparing Password objects - it's used only in tests anyway. It's
getting replaced with verify() that only accepts password strings.
Uses of old function are fixed with exception of a few calls in tests
that will be addressed in my Argon2 patch.
Change-Id: I2b2be9a422ee0f773490eac316ad81505c3f8571
