There's currently no way to get, e.g., all categories except the
hidden ones just as text. The OutputPage::getCategories() method
always returns all categories as an array of strings (titles) and
the getCategoryLinks() method returns the result of Linker::link
but with the distinction between "normal" and "hidden" categories.
This change adds a new parameter to OutputPage::getCategories(),
$type, which can be used to define, what categories should be
returned. The default value is "all", which means, that all categories
are returned (the current result of the method). With the value
"normal" and "hidden", the method will return the respective values.
This could be used in I97d7de723fe72da26c7dbde0a559a13704c7099a to
remove the stupid Linker::link() and isset workaround.
Change-Id: Iadda9ae362a21fbee770240234b8f55326219932
This change resulted in unreasonable feature loss (human-readable
limit report was gone). Three months and multiple followups later,
the functionality is still not completely restored. Given lack
of response from the original author, I think it is time to revert
and reconsider, especially since the 1.28 release is soon.
A machine-readable limit report would be a very useful feature,
but not at the cost of losing human-readable limit report.
This reverts the following commits:
* Move NewPP limit report HTML comments to JS variables
b7c4c8717f
* Only pretty-print the parser report JS vars
28adc4d7ee
* Show wgPageParseReport on page previews too
1255654ed5
* Re-add human readable parser limit report
0051f108b9
* Restore hooks.txt for ParserLimitReportFormat
4663e7a737
Resolved minor merge conflicts in OutputPage (with 80e5b160)
and release notes.
Bug: T110763
Bug: T142210
Change-Id: Id88c8066fae3f369e8977b4b7488f67071bdeeb7
Use HTTPS instead of HTTP where the HTTP link is a redirect to the HTTPS link.
Also update some defect links.
Change-Id: Ic3a5eac910d098ed5c2a21e9f47c9b6ee06b2643
It looks like there is something missing after the last statement
Also remove some other empty lines at begin of functions, ifs or loops
while at these files
Change-Id: Ib00b5cfd31ca4dcd0c32ce33754d3c80bae70641
This makes frequently changed pages be less likely to be
seen in stale forms if purges are delayed or lost. Pages
that have not been edited for somewhat longer than the
nominal CDN cache TTL will retain the nominal TTL.
Category pages will adapt based on page_touched rather
than the last revision, given the possibility of constaintly
changing membership. With their lesser overall usage,
this is less risky than for Article, and also more useful.
Change-Id: If621aca2fb68e9f87a50c891dac8dc6ec7641f5c
* Remove filter between $moduleStyles and preloadTitleInfo().
Previously, this was filtering out wiki modules created by
gadgets and extensions. Still causing a couple of direct queries
from getTitleInfo().
* Store __METHOD__ in $fname outside getWithSetCallback.
This way queries are logged as either getTitleInfo or preloadTitleInfo.
This regressed in 0852a000a5, after which the latter was logged
as "::{closure}".
Change-Id: I454e43f43f8ad3270b86f4cfdbd68192c305cc07
This mostly reverts commit 1255654ed5.
This re-adds the human readable parser limit report, and makes a few
adjustments necessary for it to work properly.
* In EditPage::getPreviewLimitReport(), only generate the HTML report,
the JS variable will be added by OutputPage
* If there are multiple calls to OutputPage::addParserOutputMetadata(),
only use the limit report data from the first one.
* Only add the wgPageParseReport variable if limit report data is
available.
Bug: T142210
Change-Id: Iad2646acde79b8a59710bb9fd5fbbfea5a39c341
The Config interface (and it's implementation(s)) was never thought
to be an access method for objects saved in the global state, even
if it works with the current implementation GlobalVarConfig.
Imagine, that MediaWiki core switches to another file-based configuratiion
storage or a a database based one, we wouldn't be able to provide
access to global objects anymore, without weird hacks in the new
config-backend implementation or serializing objects to store in
the database or something else. This all isn't the idea with the Config
interface, as far as I know, so don't use it at all.
This commit changes the access to wgContLang to use the global keyword,
instead of accessing it through Config.
Follow up: Ice4f40911c3761c2542430935bc1898bc4e7a4d4
Follow up: I46f376a82205a5c99b98c9e971f9e9d7868ce9fb
Change-Id: I7a08b3bb649898abd445317a523051b07420b211
It's primarily meant for nullable timestamps in the database.
And for the mere purpose of an is_null() check, it's more confusion
than useful to cause this function since in these two cases its return
value is not used as-is but further conditionalized.
* Also fix fragile link between getRegistration()'s false return
value never being passed to wfTimestamp (to mean "now") because
of the isLoggedIn() guard. The ternary would now make those
follow the 'else' branch toward null.
Change-Id: I9a1ee2a56e1767bfb750e27b1f37bdaeb5e6378d
On pages where the 'user' module is filtered out (e.g.
on Special:Preferences) it would export state 'loading' (or state
'ready') eventhough the actual call to makeResourceLoader() later
in getBottomScripts() would be a no-op due to filtering.
This would cause either an indefinite state of "loading" or a
state "ready" that wasn't true.
This restores status quo as it was before 80e5b160 and 3e7a50d5f.
Test plan:
* Logged-in with non-empty user page common.js.
* View Special:Preferences.
* Verify in <head> source code, or via mw.loader.getState('user')
that it has state "registered" (the default initial state) and
not state "loading" or "ready".
Change-Id: I9b360d7e12703bddb80793aef47296fd63032c3d
* Use OutputPage::output() as the method to mask latency, since it
takes a good while to run. By the time it runs, cache replication
should have caught up, so the reap call will likely not block.
* For redirects emitted after changes in POST, instead of masking
with OutputPage, add a parameter to the redirect and block on
the positions appearing. This uses the redirection RTT to mask
the replication latency.
Change-Id: Ib23690c302e8033610fef9a0ef451dafe8a5803e
This is more consistent with LoadBalancer, modern, and inclusive
of master/master mysql, NDB cluster, and MariaDB galera cluster.
The old constant is an alias now.
Change-Id: I0b37299ecb439cc446ffbe8c341365d1eef45849
Follows-up 80e5b160e0, which had to move this call out of the
headElement() and buildCssLinks() methods as it was no longer
allowed to modify the module queue after it was created.
It was moved to OutputPage::output(), right before Skin::outputPage()
is called, which ends up calling headElement().
The point in time was effectively unchanged for page views.
However for the caller in ApiParse() this meant setupSkinUserCss()
no longer got called at all as it never calls output(), but instead
calls headElement() directly.
Move it to getRlClient(), which is where we set all other
OutputPage-specific things relating to module loading already.
* For page views this has no impact.
* For ApiParse it means headElement(), which calls getRlClient(),
will once again include skin stylesheets.
Bug: T144301
Change-Id: I5fd4a27fb2d70b98ce9161dc050788d8ac364110
Before 93ed259cf and 7bb7315d4, 'site' and 'site.styles' module were one module
loaded in two parts. One part via the general queue (not in a separate request),
and the other part in the stylesheet queue (in a separate request).
This was achieved by hacking the stylesheet queue handler to hardcode the group
to be 'site' for a module named 'site' (later 'site.styles' in c20a76ee)
eventhough the module as a whole was not actually in that group.
Now that these modules are separated, this hack is no longer needed
and the module can be registered as being in group 'site' properly.
Change-Id: If193f556be211749b115f3808ee46536b578fb22
Anon users have predictable edit tokens, hence someone could
force an anon to execute arbitrary CSS by means of a CSRF.
Bug: T133147
Change-Id: I442b2b46cadb967aaa1f35648eff183fc7eaa475
Follows-up 80e5b160e which moved queue formatting out of OutputPage into a
a separate ResourceLoaderClientHtml class.
The special handling for 'user' and 'user.styles' modules, and the exempt
module groups was kept in OutputPage. However the handling for it was
hardcoded for the modules in that group by default. It did not account for
modules with a group of 'user' loaded by an extension (e.g. GlobalCssJs).
GlobalCssJs modules were wrongly loaded in the regular style queue
(still in a separate request group, but not in the right cascading order
below the DynamicSyles marker).
This commit generalises the handling previously put in buildExemptModules
and moves it to getRlClient() so that it may apply to all style modules.
This commit should be a no-op besides the moving of any <link rel=stylesheet>
for non-core modules in group 'site' or 'user' now being one line lower
in the <head> HTML (after the DynamicStyles marker).
Bug: T143357
Change-Id: I1d6ea10b42293acfc535578172ad7ab2369f6299
Follows-up 80e5b160e, which moved filter logic to getRlClient()
so that ResourceLoaderClientHtml doesn't have to deal with this.
While that worked fine for most modules (including 'site'),
it no longer applied the filter for exempt/hardcoded modules
(such as 'site.styles' and 'startup').
Bug: T143066
Change-Id: Iec924003873bc47484a0dc2f1a215f87aa4afdfb
HTML formatting of the queue was distributed over several OutputPage methods.
Each method demanding a snippet of HTML by calling makeResourceLoaderLink()
with a limited amount of information. As such, makeResourceLoaderLink() was
unable to provide the client with the proper state information.
Centralising it also allows it to better reduce duplication in HTML output
and maintain a more accurate state.
Problems fixed by centralising:
1. The 'user' module is special (due to per-user 'version' and 'user' params).
It is manually requested via script-src. To avoid a separate (and wrong)
request from something that requires it, we set state=loading directly.
However, because the module is in the bottom, the old HTML formatter could
only put state=loading in the bottom also. This sometimes caused a wrong
request to be fired for modules=user if something in the top queue
triggered a requirement for it.
2. Since a464d1d4 (T87871) we track states of page-style modules, with purpose
of allowing dependencies on style modules without risking duplicate loading
on pages where the styles are loaded already. This didn't work, because the
state information about page-style modules is output near the stylesheet,
which is after the script tag with mw.loader.load(). That runs first, and
mw.loader would still make a duplicate request before it learns the state.
Changes:
* Document reasons for style/script tag order in getHeadHtml (per 09537e83).
* Pass $type from getModuleStyles() to getAllowedModules(). This wasn't needed
before since a duplicate check in makeResourceLoaderLink() verified the
origin a second time.
* Declare explicit position 'top' on 'user.options' and 'user.tokens' module.
Previously, OutputPage hardcoded them in the top. The new formatter doesn't.
* Remove getHeadScripts().
* Remove getInlineHeadScripts().
* Remove getExternalHeadScripts().
* Remove buildCssLinks().
* Remove getScriptsForBottomQueue().
* Change where Skin::setupSkinUserCss() is called. This methods lets the skin
add modules to the queue. Previously it was called from buildCssLinks(),
via headElement(), via prepareQuickTemplate(), via OutputPage::output().
It's now in OutputPage::output() directly (slightly earlier). This is needed
because prepareQuickTemplate() calls bottomScripts() before headElement().
And bottomScript() would lazy-initialise the queue and lock it before
setupSkinUserCss() is called from headElement().
This makes execution order more predictable instead of being dependent on
the arbitrary order of data extraction in prepareQuickTemplate (which varies
from one skin to another).
* Compute isUserModulePreview() and isKnownEmpty() for the 'user' module early
on so. This avoids wrongful loading and fixes problem 1.
Effective changes in output:
* mw.loader.state() is now before mw.loader.load(). This fixes problem 2.
* mw.loader.state() now sets 'user.options' and 'user.tokens' to "loading".
* mw.loader.state() now sets 'user' (as "loading" or "ready"). Fixes problem 1.
* The <script async src> tag for 'startup' changed position (slightly).
Previously it was after all inline scripts and stylesheets. It's still after
all inline scripts and after most stylesheets, but before any user styles.
Since the queue is now formatted outside OutputPage, it can't inject the
meta-ResourceLoaderDynamicStyles tag and user-stylesheet hack in the middle
of existing output. This shouldn't have any noticable impact.
Bug: T87871
Change-Id: I605b8cd1e1fc009b4662a0edbc54d09dd65ee1df
Also make EditPage::getPreviewLimitReport return the same
output instead of using copy-pasted code from the old hook.
Bug: T142210
Bug: T142214
Change-Id: I16c3c2da067fd24bf7ae564cb1d61beaa3be58c1
Fixes addModuleStyles() violation from T92459 (a464d1d41).
Similar to 93ed259cf and Id2342454b for the 'site' module.
Doesn't need to be in separate commits since per-user HTML isn't
cached the same way.
Bug: T92459
Bug: T108590
Change-Id: I195f67d061de1306c97413aada7919e9f1b1d12c
* Instead of having messy code to create a hidden HTML
comment of English strings at the bottom of the page,
expose the structured data of the parse information
to JS so tools can use it.
* Make makeConfigSetScript() use pretty output so these
variables are also easy to read in "view source".
* Remove ParserLimitReportFormat hook, since the data
is not formatted to HTML anymore.
Bug: T110763
Change-Id: I2783c46c6d80f828f9ecf5e71fc8f35910454582
Right now OutputPage requires head items to have a name when using
the addHeadItem() method.
This in contrary to most of the helper methods, which don't require
a name. ParserOutput::addHeadItem() also doesn't require a name.
Since the existing signature isn't easily changed without adding
confusion, add a separate method that allow numerical indexes
to be pushed.
This ability was already exposed via addParserOutputMetadata()
and its array_merge. This method exposes it directly.
Change-Id: I433afd124b57526caa13a540cda48ba2b99a9bde
No uses of 'modulemessages', getModuleMessages() or addModuleMessages()
anywhere in Wikimedia Git.
Change-Id: I59420880f3545d1aabf9bcbea1e34b1475697d26
We originally imagined rolling out the display of empty elements
simultaneously with the Html5Depurate, but now we have added support for
marking empty elements to Html5Depurate and plan on having some sort of
longer migration period. So, move the relevant CSS to content.css, and
remove the concept of CSS dependant on tidy driver.
Add a body class which will allow the effect to be toggled in a gadget or
extension. Actual toggling in the CSS will be in the stage 2 patch, to be
deployed after the varnish cache and parser cache have expired.
I originally imagined that there would be a gadget that overrides the
rule with an !important selector, but that method does not allow you to
recover the original display property, which is often overridden by the
style attribute or site CSS to be "inline".
Also, in RaggettWrapper, switch to the new class mw-empty-elt, following
Html5Depurate, instead of mw-empty-li. The old class will be removed in
the stage 2 patch.
Change-Id: Ic0f432c43a006629ca5a1a7c2dda3552ceb4dc4f
This allows dynamically loaded modules to depend on page-style modules
without it causing the page-style module to be loaded a second time.
* New method Module::getType() indicates whether a module is
a page-style module or supposed to be dynamically loaded.
* Emit warning from addModuleStyles() when given a module that is
not a page-style module (to be enforced later)
Bug: T92459
Bug: T87871
Change-Id: I8b6c6a10d965e73965f877c42e995d04202524f3
Right now, getInlineHeadScripts(), buildCssLinks() and getExternalHeadScripts()
all return WrappedString::join(). But because they don't know about each other
and because they need to return strings (not arrays), headElement() has no way
of merging them.
WrappedStringList allows this array to be kept, whilst still being backward-compatible
with code that calls these methods and assumig a string (since it will lazy-join
the array if the object is treated like a string).
To be used by I8b6c6a10d965e7396. Output is not changed in this commit.
Merely refactoring.
Change-Id: Iae08345473bd93cc0948d51b62c48aeb1ea460a3
