Depends on D10586
Adds a new `type` param to the `change` object passed from server to the client to describe the change type. For changes to rules, the client marks the whole rule as either added or removed and styles it accordingly in the Changes panel.
Change types for declarations are not used at this time, but are put in for consistency and future-proofing.
Differential Revision: https://phabricator.services.mozilla.com/D11116
--HG--
extra : moz-landing-system : lando
Depends on D10585
Renames the logChange() to logDeclarationChange() to distinguish it from the newly introduced logSelectorChange() method which tracks selector rename by logging two changes: a whole rule remove using the old selector and a whole rule insertion add using the new selector.
MozReview-Commit-ID: 9VoVMHYXumE
Differential Revision: https://phabricator.services.mozilla.com/D10586
--HG--
extra : moz-landing-system : lando
Depends on D10584
Other methods for tracking changes need to make use of the CSS rule
metadata (selector, rule index, ancestors, etc). This extracts that
logic into an accessor on the StyleRuleActor to facilitate reuse.
The CSS rule metadata will be augmented with information about a CSS
change (declarations added or removed) before being tracked.
MozReview-Commit-ID: xXec1XgUhk
Differential Revision: https://phabricator.services.mozilla.com/D10585
--HG--
extra : moz-landing-system : lando
Depends on D10582
Improves the reducer for tracking CSS changes to handle more than one
CSS declaration changed in one operation. This is a requirement for
tracking whole rule removal or whole rule addition, like it happens
when renaming a CSS selector in the Rules view.
MozReview-Commit-ID: 25pf2GRiH4D
Differential Revision: https://phabricator.services.mozilla.com/D10584
--HG--
extra : moz-landing-system : lando
This isn't strictly necessary for this patch series, but it adds an
optimization to improve performance for React rendering and solves
warnings thrown while using React in dev mode.
MozReview-Commit-ID: ujqOa9qUsd
Differential Revision: https://phabricator.services.mozilla.com/D10582
--HG--
extra : moz-landing-system : lando
In many places we were using bounds to calculate an objects dimensions but that obviously doesn't work when an object is rotated e.g.
{F588303}
Also, we were using `getBoxQuads()`, which gives the co-ordinates of the translated object. We were then applying the transform matrix to the canvas even though the coordinates came from the object **after** it was already transformed.
Anyhow, now we get the dimensions of objects as if they are not transformed and then apply the transformation matrix, which gives a great result every time.
Differential Revision: https://phabricator.services.mozilla.com/D9805
--HG--
extra : moz-landing-system : lando
When Selection is NOT collapsed, we remove selected content. Therefore,
web apps don't need to know range information of user operation. However, web
apps may want to know direction of the operation (backward or forward). E.g.,
web apps may just mark selected range as "deleted" and move caret before or
after the range.
Therefore, when computed EditAction is eDeleteWordBackward or
eDeleteToBeginningOfSoftLine, we should use eDeleteBackward instead. When it
is eDeleteWordForward or eDeleteToEndOfSoftLine, we should use eDeleteForward
instead.
Note that only on Windows, we follow behavior of richtext control (and Word).
That is, Ctrl + Backspace/Delete collapse from start of selected range to
start/end of current word. I.e., collapsing Selection to start first and
removing to start or end of current word is Windows's standard behavior.
Currently, we do this in DeleteSelectionAsSubAction() but every caller
specifies eNone to aDirection except DeleteSelectionAsAction(). So, we can
move this before re-computing EditAction in DeleteSelectionAsAction().
Differential Revision: https://phabricator.services.mozilla.com/D10992
--HG--
extra : moz-landing-system : lando
So that we don't generate anymore unnecessary change hints in
RestyleManager::AddLayerChangesForAnimation for the layer has no corresponding
animations.
Depends on D11105
Differential Revision: https://phabricator.services.mozilla.com/D11106
--HG--
extra : moz-landing-system : lando
This makes it easier to read these flags from JIT code. The patch also splits
them in MutableFlags and ImmutableFlags, this should let us simplify XDR and
CopyScript in the future.
Differential Revision: https://phabricator.services.mozilla.com/D10735
--HG--
extra : moz-landing-system : lando
Also uses JS_DefinePropertyById instead of JS_SetPropertyById because it's more natural.
Differential Revision: https://phabricator.services.mozilla.com/D11094
--HG--
extra : moz-landing-system : lando
This patch re-enables the new behavior of bug 1479964, to set keyCode or
charCode of keypress event whose value is zero to the other's non-zero value.
However, some web apps are still broken with the new behavior. Therefore,
this patch adds a blacklist to keep using our legacy behavior in some specific
web apps.
Note that Google Docs, Gmail and Remember The Milk are reported as broken.
However, I don't see any broken shortcut with Gmail. Therefore, this patch
adds only Google Docs and Remeber The Milk into the blacklist.
Differential Revision: https://phabricator.services.mozilla.com/D10322
--HG--
extra : moz-landing-system : lando
At first, I wanted it is better to call DebuggerServer.destroy() when
`devtools.debugger.remote-enabled` will set to false. Likewise, call
DebuggerServer.init() in case of true. But, because DebuggerServer.init() is
called by various context such as RemoteDebugger, restoring is difficult after
destroying. In this patch, simply, we make avoiding to call
DebuggerServer.closeAllListener() from inside of ServerSocketConnection.
Depends on D10864
Differential Revision: https://phabricator.services.mozilla.com/D10865
--HG--
extra : moz-landing-system : lando
Yet another approach. In the previous D9889 approach, both DebuggerSerever and
SocketServerListener retain the connections duplicately, it will be possible to
be difficult for the management. In this approach, only DebuggerSerever retains
the connections, then when ServerSocketListener was removed, removes the
connections as well that were linked the ServerSocketListener.
Differential Revision: https://phabricator.services.mozilla.com/D10864
--HG--
extra : moz-landing-system : lando
This gets rid of an unnecessary allocation.
Differential Revision: https://phabricator.services.mozilla.com/D11066
--HG--
extra : rebase_source : 138c808d78e666e55242bf142269d77fb065d7a6
extra : histedit_source : de8f5f4566cb886e155bce1a22e36f40af7b26f7
All remaining code paths hardcode true here, we can remove this argument.
It wasn't clear to me how this could ever be null before this series. The
only case I could find is that DoFakeShow might be called with a null
RenderFrameChild, but that code is gone now.
Differential Revision: https://phabricator.services.mozilla.com/D11063
--HG--
extra : rebase_source : 872d78ae1b46b0651bc4e109786f4d0d332859c2
extra : histedit_source : 40719e865f2243f90618deba5261124a260ba974
This commit removes all destruction code for RenderFrameParent to be handled by
TabParent. It's important that we remove the layer mapping in ActorDestroy to
prevent a race condition where the TabChild isn't fully destroyed yet and
sends a LayerTransaction constructor to the compositor and hits an
assertion.
Differential Revision: https://phabricator.services.mozilla.com/D11062
--HG--
extra : rebase_source : 61df2cfa3867617c39805883c06649624fffa518
extra : histedit_source : a810303ec335b4ab064dbd56b4652ea4c66deaad
SetRenderFrame() can be implemented in terms of InitRenderFrame(). I'm not sure if
the call to MaybeShow() is necessary, but to be conservative I've moved it into
the window.open path which might need it. BrowserElementParent shouldn't need it
because nsFrameLoader::SetRemoteFrame will call Show().
Differential Revision: https://phabricator.services.mozilla.com/D11061
--HG--
extra : rebase_source : 7d5defc113d107bf77296f1a0a4f7e7dad910db6
extra : histedit_source : 397121af3a86ed3820f055292a8622d3e0bea2b5
We should just have the parent handle initialization here. This lets us
cut down on the information we have to pipe around and simplifies our
amount of code paths.
Differential Revision: https://phabricator.services.mozilla.com/D11060
--HG--
extra : rebase_source : d6c52cb81c7deceb34ad6813d2a55f779d3a1a7d
extra : histedit_source : 4e5a24206e4b8142f2ee788a18c2e186969acfd7
Dão Gottwald
Ciure Andrei
Razvan Caliman
Jack Smith
Jim Lennon
Michael Ratcliffe
Mike Conley
Martin Stransky
Csoregi Natalia
Masayuki Nakano
Tooru Fujisawa
Ryan Hunt
byron jones
Dorel Luca
WR Updater Bot
Hiroyuki Ikezoe
Gabriele Svelto
Marcos Cáceres
Arnaud Bienner
Johann Hofmann
Jan de Mooij
Patrick Brosset
Daisuke Akatsuka