Follows-up b118bddc15, accidentally started it lowercase, which is
different from all other doc groups in core.
Change-Id: I51262f71bc24e1caf943f73e254f16b481860f27
Also improve the description a bit and explain at a high level
how this code is used, and which classes back those entry points.
Change-Id: Ia69531858e714d0b43a29a7d5b6dcc5c7edd8240
Various maintenance scripts assume reserved usernames like
"MediaWiki default" exist, but since they're reserved
User::isUsableName() returns false and therefore the actor migration
created them as anonymous actors. Which would then prevent those
maintenance scripts from using User::newSystemUser() to ensure they
actually exist.
This adjusts User::newSystemUser() to be able to create users for
those anonymous actors.
This also adjusts uses of "MediaWiki default" in core to create it as a
system user.
Bug: T236444
Change-Id: I59a646df36ff9343cc43c05aa20b2b69b2ee124a
This bug was fixed over 10 years ago and is impossible to encounter
on modern libxml versions. While libxml versions might vary slightly
for the same PHP version depending on how it's built, this gives a
general idea:
https://3v4l.org/Tk27c
Change-Id: I669abb3543180e6edd090297fcfcb811aa833b57
Since MediaWiki 1.18, $wgServer has been automatically set by the web installer
when it generates LocalSettings.php, so this shouldn't be an issue for most
wikis. The CLI installer now supports a --server optional parameter to
specify $wgServer, otherwise it'll be set to 'http://localhost' by default.
Users will see a fatal error pointing them to the on-wiki $wgServer
documentation that I've updated as well.
Originally this functionality was slated for removal in 1.20, but now is
just a good time as any. It also calls into other parts of MediaWiki before
most things are initialized, making it difficult to librarize some code.
Bug: T30798
Bug: T232931
Change-Id: Ia5d616e7fafbab01655067c24c5a3a073b254f21
There is no reason this should be necessary. Resetting services should
do the trick, together with Language::$mLangObjCache while it exists.
Resetting only a subset of services is fragile and shouldn't be done,
because one of the reset services might later be injected into a service
that's not reset, resulting in bugs.
Change-Id: I8c72b466ba7337649f9256ee6c078c76163f9785
Languages with variants no longer can override the parent's constructor
(which is now used for injecting services). Instead, they need to
override Language::newConverter().
Bug: T201405
Change-Id: I923400d61763cf1db88cb0c3f684c9c10e58032d
This removes Language::$dataCache without deprecation, because 1) I
don't know of a way to properly simulate it in the new paradigm, and 2)
I found no direct access to the member outside of the Language and
LanguageTest classes.
An earlier version of this patch (e4468a1d6b) had to be reverted
because of a massive slowdown on test runs. Based on some local testing,
this should fix the problem. Running all tests in languages is slowed
down by only around 20% instead of a factor of five, and memory usage is
actually reduced greatly (~350 MB -> ~200 MB). The slowdown is still not
great, but I assume it's par for the course for converting things to
services and is acceptable. If not, I can try to optimize further.
Bug: T231220
Bug: T231198
Bug: T231200
Bug: T201405
Change-Id: Ieadbd820379a006d8ad2d2e4a1e96241e172ec5a
This reverts commit cc7ec36a57.
It makes the following changes from the original patch:
* findExtensions() will not consider config-extension-not-found as an
error.
* Fixed a bug where extension info wouldn't actually be returned by
findExtensions().
* When an extension's dependency doesn't exist, wrap the
config-extension-not-found error in config-extension-dependency.
This makes it more clear, and prevents the error from being ignored
per the first bullet.
* maintenance/install.php will use ->text() rather than ->parse()
when printing thrown errors.
* Change the stuff done to make phan happy.
Bug: T225512
Change-Id: I7d29700e8b7e91841556847d669b350cbd306fe6
Move the default of 2000 to DefaultSettings.php, and promote
the installer warning to a hard value.
Recommend that the setting be disabled, or increased to 5000.
Change-Id: Ifb0aadf8b52fd2d5c2a32f55e38eaa9c2600dfb5
Changes the behavior of -—with-extensions from silently skipping
extensions that lack another extension dependency to throwing an
install exception.
Bug: T225512
Change-Id: Ia5b41841aaf55fa5c4971ae67218f8231f8adcb8
PHP doesn't care much but I think we humans do because we should
call methods by the name we give them. Method fixed are;
- isOk() -> isOK()
- setOk() -> setOK()
- teardown() -> tearDown()
Change-Id: I6b3f0cf3902887058efa426968da380803869e0b
This removes Language::$dataCache without deprecation, because 1) I
don't know of a way to properly simulate it in the new paradigm, and 2)
I found no direct access to the member outside of the Language and
LanguageTest classes.
Change-Id: Iaa86c48e7434f081a53f5bae8723c37c5a34f503
After approval of RFC T191231, we are going to drop oracle and mssql
and it will be possible to bring back the support using the abstract schema
Adding to release notes will be done in a follow-up
Bug: T230418
Change-Id: I90bd5cfcc3e18011b193c965fdb1fa54675040b5
Previously, if there was an error during CLI installation,
CliInstaller::showStatusMessage() exited the script directly. The exit
timing of the script should be given to the caller, not the callee.
So, I coding:
[1] Remove `exit()` from CliInstaller::showStatusMessage()
[2] Make the callee to return Status, the caller determine how to handle these Status
[3] Strictly check the key database type instead of just outputting message
Bug: T46511
Change-Id: I72ffd33fe5c592b9ea78f37bae5a9c081295c624
overrideConfig is a public static method and can be called multiple
times by other processes but does not check that the constant
`MW_NO_SESSION_HANDLER` is defined before attempting to set it.
Bug: T224287
Change-Id: I01bcbd36ec001a5c55c2acebb24c44df3d3d3277
The argument $callback is not a callable
The callable is in $callback['callable'], so change the type to array
Change-Id: I9b990f2729874b2743cbcc501b97758da5bf8154
The nice thing about explode() is that the resulting array is
guaranteed to contain at least one element. The array can not be
empty.
In some of these cases it might be possible to use strstr() instead,
but that returns an empty string when the needle character is not
found. explode() returns the original string in this case.
Change-Id: I6ad1f3273defeaf36e2305fd871eaaf9d3c1e134
All methods got moved to HttpRequestFactory or MWHttpRequest or dropped.
I made the return value of the new HttpRequestFactory::request/get/post
methods null on error instead of false, so that when we drop PHP 7
support, we can use a "?string" return value. This could theoretically
change behavior of code that was switched from the old Http methods, but
probably won't. I kept the old behavior for the deprecated methods.
I changed the default value of $wgHTTPProxy from false to ''. This way
it should be usable directly without a trivial wrapper method. For the
benefit of anyone who might have set it to false in LocalSettings.php, I
also recommend casting to string just in case.
Http::$httpEngine is deprecated. Eventually it will be removed along
with the curl and PHP engines, leaving only the Guzlle engine.
I also added deprecation of MWHttpRequest::factory, which occurred in
1.31, to the release notes for 1.34. Now hopefully we can hard-deprecate
it in another couple of versions.
Bug: T214390
Change-Id: I2a316a758d793857f248bd251b90f5e9a6440e3a
HHVM does not support variadic arguments with type hints. This is
mostly not a big problem, because we can just drop the type hint, but
for some reason PHPUnit adds a type hint of "array" when it creates
mocks, so a class with a variadic method can't be mocked (at least in
some cases). As such, I left alone all the classes that seem like
someone might like to mock them, like Title and User. If anyone wants
to mock them in the future, they'll have to switch back to
func_get_args(). Some of the changes are definitely safe, like
functions and test classes.
In most cases, func_get_args() (and/or func_get_arg(), func_num_args() )
were only present because the code was written before we required PHP
5.6, and writing them as variadic functions is strictly superior. In
some cases I left them alone, aside from HHVM compatibility:
* Forwarding all arguments to another function. It's useful to keep
func_get_args() here where we want to keep the list of expected
arguments and their meanings in the function signature line for
documentation purposes, but don't want to copy-paste a long line of
argument names.
* Handling deprecated calling conventions.
* One or two miscellaneous cases where we're basically using the
arguments individually but want to use them as an array as well for
some reason.
Change-Id: I066ec95a7beb7c0665146195a08e7cce1222c788
MediaWiki core now runs phan 1.2.6, bringing in nearly 2+ years of upstream
fixes.
Configuration was moved from `tests/phan` to `.phan/`. The legacy bash wrapper
script is still kept in the repository in its own location for any extensions
that are still using it. It should be removed before 1.33 is released.
Since there's a lot of new issues being flagged, all currently failing issues
are suppressed, and will be fixed in follow-up patches.
We're dropping the jetbrains/phpstorm-stubs repository in favor of just
the minimal stubs we need. Stubs for PHP extensions are kept in
the new `.phan/internal_stubs` directory, since they're in a slightly
different format than normal stubs.
Normal stubs are kept in `.phan/stubs`. wikidiff2 and excimer are kept with
these since we're also the upstream for them.
Change-Id: I3fe437befa17f4fbaf97aa6271f659b56021f396
global $wgParser was deprecated in 1.32 and services are used
instead to get the parser object via MediaWikiServices. Making
this replacement as we're avoiding globals.
Change-Id: I967a52c481bfcc83552053a03e88acb038810828
This code doesn’t use any MediaWiki-specific code, so rename
MediaWiki\Services to Wikimedia\Services and move it below libs/. (Of
course, this does not apply to the MediaWikiServices subclass.)
Class aliases are added to retain backwards compatibity for now.
Bug: T211608
Change-Id: Ic14ea28ef21c359695b309d4293dbaaf5deedc09
In the web installer this isn't the last step and there is a different
success message. Replace the existing config-install-success with a
much more generic config-install-db-success message.
Previously we were telling people to go to their wiki prior to
them downloading LocalSettings.php which would obviously not work.
Change-Id: I56e28157a1a4d55e4eeda5fdd536289960511938
Allow the extensions and skins installed by maintenance/install.php to
be customised using --skins= and --extensions=. If the argument is
am empty string then no extensions/skins are installed. For backwards
compatibility, the default is to install all skins, but to install all
extensions only if --with-extensions is given.
The new CLI options may be specified multiple times, but for
convenience, comma-separated lists can also be used.
Also:
* Rename $option to $options
* If an extension has a dependency error, propagate the very readable
error message generated by ExtensionRegistry back to the user.
* Split getExtensionInfo() from the loop body of findExtensionsByType(),
so that CliInstaller can use it to validate its parameters and get
error messages.
* I didn't like the idea of removing the "s" from the directory name in
order to construct the JSON file name, so I split
findExtensionsByType() from findExtensions(), with the former not
having this hack. In findExtensions(), make the previous assumption
that the directory name is always "extensions" or "skins" explicit,
throwing an exception if it is otherwise.
Change-Id: Id0fb63cd4e61a047ef3396ee1c38d6073dfc7fd1
This change parallels the new method added to OutputPage in
Ia58910164baaca608cea3b24333b7d13ed773339 and ensures that the
content added is always tidied. We leave the old alias in
place for now in accordance with our deprecation policy.
Change-Id: I89f3398cffa771afcd5a33cfd11eb8510af3e7f7
Clarify and simplify exception output by deprecating
$wgShowSQLErrors and wgShowDBErrorBacktrace.
$wgShowExceptionDetails will now control most related output.
$wgShowHostnames will now solely control output of
MWExceptionRenderer::reportOutageHTML.
Bug: T165768
Change-Id: Idead2c11c499463dfa6293c3d4b33be3bde92e1a
To limit the likelihood of incidents like T189665, the ability to
edit sitewide CSS/JSON/JS is split out from editinterface, into
separate 'editsitecss', 'editsitejson' and 'editsitejs' rights.
editsitecss, editsitejs, and the right to edit another (potentially
more privileged) user's personal CSS/JS is removed from sysops, and
a new user group, interface-admin, is created specifically for that
task (along with a new grant group 'editsiteconfig').
interface-admin is granted to the first user of a new wiki, along with
sysop and bureaucrat.
Bug: T120886
Bug: T190015
Depends-On: Ia9b2ea1450aff6121dc0f3777bf029292c8aaad9
Change-Id: Ifefd872640642441e26f8b2f144ffe4b88d2eb12
