To improve the overall performance of cookie banner clicking mechanism. This
patch makes the cookie banner clicking mechanism only executes once per
browsing session.
Differential Revision: https://phabricator.services.mozilla.com/D190763
This patch also does some cosmetic changes for when parent controlled
loads should be used, but more importantly introduces the pref
'browser.tabs.documentchannel.speculative-load' with the default value
false.
It turns out that speculatively loading in the parent when a parent
controlled load isn't possible is the cause for performance
regressions.
Differential Revision: https://phabricator.services.mozilla.com/D192489
Removes the browser.translations.languageIdentification.useFastText
pref, making CLD2 the only option for language identification
for Translations.
Differential Revision: https://phabricator.services.mozilla.com/D192656
Removes the browser.translations.languageIdentification.useFastText
pref, making CLD2 the only option for language identification
for Translations.
Differential Revision: https://phabricator.services.mozilla.com/D192656
This adds support for the experimental `invoketarget` and `invokeaction`
attributes, as specified in the open-ui "Invokers" explainer.
(https://open-ui.org/components/invokers.explainer/)
The `invoketarget` attribute maps to the IDL `invokeTargetElement`,
similar to `popoverTargetElement`, and the `invokeaction` is a freeform
string.
The Button behaviour checks for `invokeTargetElement` in its activation
behaviour, and dispatches an `InvokeEvent` if there is one.
This also adds some basic scaffolding for `handleInvokeInternal` which
will allow elements to provide their own invocation action algorithms.
Differential Revision: https://phabricator.services.mozilla.com/D190449
To improve the overall performance of cookie banner clicking mechanism. This
patch makes the cookie banner clicking mechanism only executes once per
browsing session.
Differential Revision: https://phabricator.services.mozilla.com/D190763
This patch also does some cosmetic changes for when parent controlled
loads should be used, but more importantly introduces the pref
'browser.tabs.documentchannel.speculative-load' with the default value
false.
It turns out that speculatively loading in the parent when a parent
controlled load isn't possible is the cause for performance
regressions.
Differential Revision: https://phabricator.services.mozilla.com/D192489
Removes the browser.translations.languageIdentification.useFastText
pref, making CLD2 the only option for language identification
for Translations.
Differential Revision: https://phabricator.services.mozilla.com/D192656
Let's ship lazy loading iframes in Fx121 if no more regressions are
found in Nightly.
If agreed and nothing else is found, I'll land this patch ~Nov 6.
Differential Revision: https://phabricator.services.mozilla.com/D192245
Instead, use forced-color-adjust: none to disable HCM forced colors in
DevTools, for now. It's a more straight-forward way of doing it.
Differential Revision: https://phabricator.services.mozilla.com/D192669
When I landed this initially, it looked to me like the spec PR was about to land, but now there is an ongoing discussion about the behavior in uncomposed documents and even concerns with the previous <input> showPicker method.
Differential Revision: https://phabricator.services.mozilla.com/D192747
The spec explicitly requires that we "must ensure that tab stops continue to line up"
when applying justification to content with preserved white-space that includes tabs.
So when we're computing justification spacing adjustments, we must not apply adjustments
if there is a preserved tab later on the same line; only text after the last tab is to
be justified.
Differential Revision: https://phabricator.services.mozilla.com/D191996
Change the pref to "0". Nightly builds (not Beta or Release, even once
this patch rides the trains) will interpret this as
remote-with-fallback.
Differential Revision: https://phabricator.services.mozilla.com/D192611
This patch also does some cosmetic changes for when parent controlled
loads should be used, but more importantly introduces the pref
'browser.tabs.documentchannel.speculative-load' with the default value
false.
It turns out that speculatively loading in the parent when a parent
controlled load isn't possible is the cause for performance
regressions.
Differential Revision: https://phabricator.services.mozilla.com/D192489
When colors are converted to sRGB to render onto the display, make sure
that they are within sRGB gamut limits.
Gamut mapping is implemented according to:
https://drafts.csswg.org/css-color-4/#gamut-mapping
The color-mix-non-srgb-001 test is checking the expected result in
sRGB, which happens to be out of gamut limits, but because the test
is for color-mix and not gamut mapping, I changed the expected
results to the color space of the mix.
The svg reftest now has an increased fuzzy to allow for the final colors
to be gamut mapped. Gamut mapping is dependent of available hardware, so
we can't pin down exact colors for the result.
Differential Revision: https://phabricator.services.mozilla.com/D191083
