If the paction=serializeforcache request fails, we were erroneously
converting it to a successful result with no value, which later causes
an exception, because since 381b58585c
other code expects the result to an an object.
The bug was introduced in 2015 in 07001001be,
but until that recent change it would only cause a 'badcachekey'
error, which was handled correctly later.
Change-Id: Ie1ffc8c3e616a7d296f2186fb17eaf039971a44f
This means the multiple lines of html (e.g. '<p>a</p><p>b</p>') are
treated as plain text and only separated by one linebreak.
Bug: T201561
Change-Id: I83b98e41bfd92d1848557b58f961abdd0db26294
New changes:
18e32c6ed Simplify paste test runner
7d71154bd Properly support middle click paste
Local changes to use simplified paste tests runner
Bug: T157956
Depends-On: Id66bff4e41a36ed967a8cba2f6653bb26e7b4ea1
Change-Id: I0101e8bc079cd050bfbc65577a10e98213d5f00c
With this patch, the toolbar slides into place place nicely after
scrolling again, but it still occasionally flickers during the scroll.
* window.innerHeight is now smaller or something, and we have to
twiddle the scroll position by a larger value.
* document.body.scrollTop no longer works for setting or getting the
scroll position, so use different methods.
* requestAnimationFrame() now generates an insufficient delay to make
scrolling happen, so use setTimeout() instead. We actually have to
add a nonzero delay there, otherwise the toolbar sometimes doesn't
animate like it should or flashes in random places on the screen.
This delay is bad because the user can't start scrolling again
during that time, but I think we can live with that.
Bug: T233470
Change-Id: I6c40ee8ce5994e12eadb085bbffd120ef160d4ee
New changes:
4af3f84f7 Mark surface as "showAsDeactivated" when opening a window
79eb0e4e5 ve.ce.Surface: Guard against focusing a un-initialized surface
4124c275e [BREAKING CHANGE] ve.ui.TargetWidget: Construct a real target inside the widget
Local changes:
* Use new target widget
* Remove calls to deprecated methods
* 'surfaceReady' event was upstreamed
Bug: T236400
Change-Id: I765d657c172d96c3b2e2ae5998083e4926a31f15
As of commit c65ed0e7a8ac5f32a3a6e4cb2760dae03e4fca22 in MobileFrontend,
it uses errorformat=html queries (the same as we do), so we no longer
need to massage the responses to make it happy. The same commit also
turned parseSaveError() into a no-op, so we can remove that as well.
Change-Id: I4f0109ce120ebf94e5709d47d775a8178ce216fa
Something is causing the 'ETag' headers produced by the "public"
RESTBase (queried directly from the client) to be mangled or lost.
My theory is that some proxy or browser extension is doing that.
When we detect a bad etag when fetching the page contents, discard
the result and try querying the "private" RESTBase via the MediaWiki
API (similar to what we do on private wikis, except there we talk
directly to Parsoid instead of RESTBase). After I463a84de63, that
returns the etag as part of the payload rather than HTTP headers,
and should pass unharmed through whatever is mangling the data.
Also compare and log the two etags.
Bug: T233320
Change-Id: I2ef0ca872597566f74b650aea71bf3f15747a6d7
Previously, the ve-mw/init/ directory contained two kinds of files:
those that were used when initializing VE, and those that may be
loaded even if VE is not going to be initialized at all. The latter
kind must not use the `ve` global variable.
After moving those files to ve-mw/preinit/ we can enforce this with
.eslintrc.json in that directory. This would have prevented T228684.
(Technically they merely must not use `ve.init`, and may use `ve`,
but that's harder to enforce. We should instead move the few non-init
methods out of `ve`: now, track, trackSubscribe, trackSubscribeAll).
Also, group some files under ve-mw/init/: targets/ now (only)
contains ve.init.mw.Target and its subclasses, apiresponsecache/
now contains ve.init.mw.ApiResponseCache and its subclasses.
Bug: T228684
Change-Id: I945249a27f6a0fa10a432d5c5dc57bc7e0461fd8
The ...target.wt property contains the wikitext used to generate
the template name. It can contain trailing newlines (T234817) and
all kinds of funny wikitext syntax. Instead, use ...target.href,
which is the title of the page that is actually transcluded. Compare
the new code to ve.dm.MWTransclusionNode.prototype.getPartsList.
Additionally, fix some confusion about namespaces (treating template
names as titles in the main namespace). The template names in the
configuration page (visualeditor-template-tools-definition.json)
now support overriding namespaces in the same way as in wikitext.
Bug: T234817
Change-Id: I7c557d28e961d0b9117fc0380c65cdd42035ae96
If you had an image thumbnail for a file 'Foo?.png' on the page,
ve.ui.MWMediaContextItem and ve.ui.MWMediaDialog did not escape
the '?' when linking to it, which resulted in incorrect links.
Similarly, if you had an internal link to the page 'Foo?',
ve.ui.MWInternalLinkContextItem did not escape it.
Additionally, the links were always generated as if the wiki was
using short URLs, even when it is not (T233628).
The approach using mw.Title is copied from ve.ui.MWGalleryDialog.
Bug: T233628
Change-Id: I10256ed6883dae0ea216de4c0719f03d7fd19ae4
* In normal images, parse relative 'href' attributes instead of
expanding them to absolute, and parse 'resource' to keep it
identical to 'href' if they refer to the same page (including
same percent-encoding and space/underscore). This resolves Parsoid
generating |link= options for copy-pasted images (T193253).
* In gallery images stuff, prefix the 'resource' attribute with './',
same as normal images do. This causes no functional changes, but it
makes updating tests easier, and the consistency is probably good.
* Update test examples to also prefix 'resource' and relative 'href'
attributes with './', like the real Parsoid does.
Bug: T193253
Change-Id: If2d7f080d9d693568054f8311c1e1b15ca27ea5c
* Handle mw:MediaLinks pointing to to non-existent files, which come
with typeof="mw:Error" (similar to image nodes).
* Fix regression from c66f8e0547, which
caused all mw:MediaLinks to be treated as plain external links again.
* Add test cases.
Bug: T232754
Change-Id: I9ae5bcfc4e24e8c0d22ef77d6a4d03f817fc9768
In order to get the desired padding, the `em` values were pre-multiplied
by the parent font-size, which was assumed to be `0.875em` for normal
surfaces and `(13.3333/16)em` for wikitext surfaces. (That should have
been `(13/16em)`, by the way, but that's a negligible difference.)
Unfortunately, the font-size for wikitext surfaces is actually `13px`.
Unlike `em`, `px` values are not affected by custom browser font-size,
so when one was set, the final padding was all out of whack.
Use `rem` units instead to respect the custom browser font-size,
without getting problems due to the parent element font-size.
We don't need separate rules for .ve-ui-mwWikitextSurface now.
Bug: T222217
Change-Id: Ib7ffbf09d5aa23fddb894aa3b081ec993ddcee2d
This meant we were returning invalid documents with bare content outside
content branch nodes. Instead let ve.ce.Surface in VE core do the unwrapping.
Bug: T232944
Depends-On: I8799d51958b966c99307f4c70546ea326e67385c
Change-Id: I0479116b51cc3135a992fdf36b8edfb2c44916ba
