A reference to a document (and the act of saving it) can outlive the
browser tab it was originally loaded in, and this needs to be reflected
in IPC in order to avoid MsgRouteError crashes; see bug for more info.
Note that we still need to pass the PBrowser when starting persistence,
because that's the only handle the parent has on the top-level document;
see comments in this patch for more info.
Also makes TabChildBase::GetDocument public, because it's just a wrapper
around WebNavigation() which is already public, to avoid code duplication.
This avoids a real event being dropped in favour of a synthesized event via IPDL compression, which is important because synthesized events don't generate 'mousemove' DOM events.
--HG--
extra : rebase_source : 57b771d2d7de686148a29b91b2b7bf514e541246
extra : source : 0becd45c08952628961dc9bdccb0e11dc5b6f910
This avoids a real event being dropped in favour of a synthesized event via IPDL compression, which is important because synthesized events don't generate 'mousemove' DOM events.
--HG--
extra : rebase_source : 711341f7ae50583498854993028bbd9e1b1299cc
extra : source : c0e8553911e54232f060bbb284f936d81f73f411
This alters nsIWebBrowserPersistable so that startPersistence takes an
outerWindowID. This allows us to target a particular subframe from
beneath an nsFrameLoader, which is useful when attempting to Save
Frame As a remote browser.
--HG--
extra : commitid : KWck2rfRjXw
extra : rebase_source : d2390a8cf1b830e4e1c0fa52574455fbc0b05c14
The creation of the surrogate native window in the child NPAPI process was
failing when then sandbox was at low integrity, because the parent is from the
chrome process, so at medium integrity.
Instead of making an IPC call to get the parent, we now create the window upfront
and send it in an IPC message to be parented in the chrome process.
This is done with asynchronous messaging.
The high-level overview is that the parts of nsWebBrowserPersist which
access the DOM have been factored out (as WebBrowserPersistLocalDocument)
and abstracted (nsIWebBrowserPersistDocument) such that they can be
implemented in the cross-process case using IPC.
The code that deals with the features is looking for char *, and we were going through
the hassle of converting that char * to a UTF16 string, and then back down to UTF8,
and then to char *. Really, we just need to send up a nsCString, which simplifies things.
--HG--
extra : commitid : HgzV7FFsQCv
extra : rebase_source : feabba83d86d732fbac5d1d26c6bf362928c0b9f
extra : histedit_source : 5a0cdbbbbd05632d2071de2ecc7771ed71abcbe2
We were returning false from TabParent::RecvCreateWindow whenever anything went wrong.
Returning false in response to an IPC message results in the content process being
killed, which is a terrible user experience. With this patch, instead of returning
false, we more often return the nsresult of operations occurring within TabParent.
--HG--
extra : commitid : FVk1R6eSp5M
extra : rebase_source : 1f86739a7159205cb3467316d87e51cb113d9085
extra : amend_source : 151568149d4c59d67bb26facab6f54b90a2d3965
When connecting a lowdpi external monitor on hidpi mac, TabChild gets an
UpdateDimensions call, followed by a UIResolutionChanged call. After
the UpdateDimensions call, the content process is in an incorrect state
where it has the dimensions of the new display and scale of the old one.
After the UIResolutionChanged message, the values are again consistent.
In the interim, reflow resizes layers based on the incorrect (old) scale
and subsequently uses those incorrect values when the new scale comes
in. This patch normalizes the message parameters by dividing by scale
(the result is what OS X calls point coordinates) so that this doesn't happen.
When connecting a lowdpi external monitor on hidpi mac, TabChild gets an
UpdateDimensions call, followed by a UIResolutionChanged call. After
the UpdateDimensions call, the content process is in an incorrect state
where it has the dimensions of the new display and scale of the old one.
After the UIResolutionChanged message, the values are again consistent.
In the interim, reflow resizes layers based on the incorrect (old) scale
and subsequently uses those incorrect values when the new scale comes
in. This patch normalizes the message parameters by dividing by scale
(the result is what OS X calls point coordinates) so that this doesn't happen.
Catalin Badea
Benoit Girard
Jed Davis
Jim Mathies [:jimm]
Nicholas Nethercote
Neil Deakin
Andrea Marchesini
Markus Stange
Xidorn Quan
Botond Ballo
Phil Ringnalda
Masayuki Nakano
William Chen
Mike Conley
Gabor Krizsanits
Bob Owen
Carsten "Tomcat" Book
Kartikaya Gupta
David Parks