890125 - Window position in virtual desktop not restored across browser sessions on MS Windows

Status: RESOLVED FIXED

(Firefox :: Session Restore, defect, P2)

Description

[islisis]

User Agent: Mozilla/5.0 (Windows NT 5.1; rv:22.0) Gecko/20100101 Firefox/22.0 (Beta/Release)
Build ID: 20130618035212

Steps to reproduce:

- Open a virtual desktop manager (in my case Litestep, WinXP)
- Open multiple Firefox windows in different desktops
- Exit Firefox, then restart


Actual results:

Firefox windows on all non-focused desktops will be moved to the current focused desktop, with positions cropped to the nearest screen edge.

It seems that Firefox is attempting to prevent any windows being created offscreen during startup, but needs to be disabled to prevent conflicts with VWMs in Windows.


Expected results:

Windows should have opened in the respective relative virtual desktop positions, like versions 21 and before.

Comment 1

[firefox24]

Needs to be CONFIRMED. Hope it gets fixed. Personal choice of sticking with older version for now.

Comment 2

[islisis]

The problem is that my original submission of this bug was in reference to the general Firefox handling of window borders for all Windows platforms - see https://support.mozilla.org/en-US/questions/962982#answer-452667

After that an admin read my anecdote and decided to relabel it is a LiteStep/XP specific bug, when it clearly is not just that. It just a way of sweeping it under the carpet.

This bug should probably be resubmitted by someone.

Comment 3

[Trek]

After upgrading from firefox 17 to firefox 24, the window position is no more remembered across sessions. I tried to remove localstore.rdf without luck and the stored screenX/Y values are correct. The window height and width are instead remembered correctly.

User agent: Mozilla/5.0 (X11; Linux i686; rv:24.0) Gecko/20140207 Firefox/24.0 Iceweasel/24.3.0

Comment 4

[firefox24]

(In reply to Trek from comment #3)
> After upgrading from firefox 17 to firefox 24, the window position is no
> more remembered across sessions. I tried to remove localstore.rdf without
> luck and the stored screenX/Y values are correct. The window height and
> width are instead remembered correctly.
> 
> User agent: Mozilla/5.0 (X11; Linux i686; rv:24.0) Gecko/20140207
> Firefox/24.0 Iceweasel/24.3.0

https://bugzilla.mozilla.org/show_bug.cgi?id=864107

It was a bug that got fixed.
I don't believe firefox was implemented with a setting to reverse the bug so I myself had found a solution. I use palemoon browser which a option was implemented to reverse this bug. Name of the setting in about:config browser.sessionstore.exactPos to true.
More detailed info.
http://forum.palemoon.org/viewtopic.php?p=19756#p19756
http://forum.palemoon.org/viewtopic.php?f=24&t=3357

Comment 5

[islisis]

Devs, it would be greatly appreciate it if you could add a config option similar to that implemented in Palemoon.

Comment 6

[Paul Silaghi, QA [:pauly]]

Can you reproduce the problem on the latest nightly ? http://nightly.mozilla.org/

Comment 7

[islisis]

Thanks for your response. Yes, I can confirm 30.0a1 still snaps all browser windows to the relative desktop resolution. There is still no upgrade path from 17.0 for users with this issue.

Comment 8

[islisis]

To clarify: 17.0 and previous would not always exactly restore sessions in identical VWM desktops, partly since FF cannot know how different VWM coordinate implementations. But that is a problem for the VWM user to deal with, since they will usually be able to perform a 'gather all windows' and at least have the relative positioning correctly restored. After 17, both absolute and relative positioning are disrupted due to nearest desktop edge cropping. VWM and other users at least require a config option to revert back to the 17.0 behaviour.

Comment 9

[islisis]

Sorry -- FF 21, not 17

Comment 10

[Boris Wirtz]

Since the bug is still present in the (just released) 29.0 :
Is there any progress on this ? 

It's really annoying shuffling around >50 browser windows on a daily basis.

Comment 11

[islisis]

Admins, kindly confirm this behaviour. Users with this problem would like to upgrade from FF21.

Comment 12

[Boris Wirtz]

What is the clean way forward to get rid of the platform restriction to "x86 Windows XP" and the restrictions to windows and litestep in the bug name ?

This bug is obviously crossplatform and does not depend on litestep as window manager.

Linux / vtwm here, btw.


And as a help for bug analysis : I see the browser windows placed on the correct (off screen) positions on the virtual desktop, and then finally all moved to the on screen part of the virtual desktop.

Comment 13

[Boris Wirtz]

Thanks. Getting closer :)
Since I actually see the windows at start up being restored to their correct position, and they are moved to the visual part of the desktop just before the window content starts loading, I'd guess the
positions _are_ stored correctly across sessions, but they are overwritten later on.
Very likely (as pointed out by my predecessors) by the code introduced as fix for

    https://bugzilla.mozilla.org/show_bug.cgi?id=864107

Since there are contradicting targets (restoring windows to visible screen in 864107 and keeping the original positions in a virtual desktop manager), probably the best way is indeed adding another configuration variable (defaulted to the 864107 behaviour).

Comment 14

[Tim Taubert [:ttaubert] (inactive)]

Yeah, I'm pretty sure this is caused by bug 864107.

Comment 15

[Boris Wirtz]

Still present in 30.0.

Comment 16

[Boris Wirtz]

Still present, still annoying in 31.0.

Comment 17

[Boris Wirtz]

Still present in 32.* and 33.0.

I guess the version tag (still pointing to the 30 Branch) needs to be updated. Is there sth. like a sticky tag ?

Just a side note : This probably will gain importance with MS Windows 10 supporting multiple desktops.

And another sidenote : I'm tempted writing a fix. However I need some advise, whether another configuration variable is the way to go, or autodetection. IMHO this is a worthy candidate for a configuration variable (although I also see that lots of configs are QA's nightmare). I have no clue about how to do autodetection, either.

Comment 18

[islisis]

A fix would be most welcome. I might recommend following the same about:config variable convention used in Palemoon for now. I'm not sure how this could be generally autodetected to make it worthwhile.

Comment 20

[Boris Wirtz]

The issue in (In reply to :Gavin Sharp [email: [e-mail]] from comment #19)
> 
> *** This bug has been marked as a duplicate of bug 372650 ***

Please note that the analysis of the problem is more evolved in this bug record.
Following 372650 only, the knowledge is lost that the original problem was already fixed (at least in general), but clearly reintroduced by the code change for 864107.

Comment 21

[islisis]

As Wirtz says, it is a different bug from a more recent code change than 372650.

Comment 22

[j]

I still have this problem. KDE 4.14.2 and Firefox Nightly 45.0a1 (2015-11-26). It never worked for me.

Comment 23

[Ben Bucksch (:BenB)]

This was just fixed for Linux and Mac.

Comment 24

[Ben Bucksch (:BenB)]

For developers:
In widget/windows/nsWindow.cpp, implement functions:

• int32_t nsWindow::GetWorkspaceID();
• void nsWindow::MoveToWorkspace(int32_t workspaceID);

See:

• Base implementation
• GTK/Linux implementation

Comment 25

[Masatoshi Kimura [:emk]]

(In reply to Ben Bucksch (:BenB) from comment #24)

In widget/windows/nsWindow.cpp, implement functions:

• int32_t nsWindow::GetWorkspaceID();
• void nsWindow::MoveToWorkspace(int32_t workspaceID);

IVirtualDesktopManager::GetWindowDesktopId and IVirtualDesktopManager::MoveWindowToDesktop would be usable to implement them.
https://docs.microsoft.com/en-us/windows/win32/api/shobjidl_core/nn-shobjidl_core-ivirtualdesktopmanager

But int32_t is insufficient to hold Guid (128-bit).

Comment 26

[Mike de Boer [:mikedeboer]]

(In reply to Masatoshi Kimura [:emk] from comment #25)

(In reply to Ben Bucksch (:BenB) from comment #24)

In widget/windows/nsWindow.cpp, implement functions:

• int32_t nsWindow::GetWorkspaceID();
• void nsWindow::MoveToWorkspace(int32_t workspaceID);

IVirtualDesktopManager::GetWindowDesktopId and IVirtualDesktopManager::MoveWindowToDesktop would be usable to implement them.
https://docs.microsoft.com/en-us/windows/win32/api/shobjidl_core/nn-shobjidl_core-ivirtualdesktopmanager

But int32_t is insufficient to hold Guid (128-bit).

Yeah, I'm looking into this too, but it indeed requires switching to strings to support the Windows style. Right now my patches look like:

void nsWindow::GetWorkspaceID(nsAString& workspaceID);
void nsWindow::MoveToWorkspace(const nsAString& workspaceID);

And I've changed the Linux & OSX code to convert the string to int's first, which seemed like the least messy solution.
But why MS needed to over-engineer this thing into using Guids is beyond me. Oh well.

Comment 27

[Boris Wirtz]

Thanx for working on this, and sorry for being obnoxious, but:

IMHO the bug report describes two related, but separate problems:
a) The workspace ID of a window is not preserved across session restores.
b) After session restore, all windows are pulled into the current viewport of the workspace (so in effect the window positions are not preserved across session restores).

AFAICS the code fixes a) but not b).
So this is a partial (though highly appreciated) fix only.

Just to make sure the virtual desktop concept is understood: In some virtual desktop managers (window managers) desktop extensions may be much larger than the part being displayed (similar to painting programs, in which you can zoom in to display parts of the canvas only). The arrangement of firefox windows on that large (virtual) desktop is exactly what is lost in b).

Comment 28

[Mike de Boer [:mikedeboer]]

Attachment: Bug 890125 - Part 1 - Change Window widget base classes to return and use strings for workspace IDs, rather than integers. r?nika!

Comment 29

[Mike de Boer [:mikedeboer]]

Attachment: Bug 890125 - Part 2 - Convert the OSX Spaces handling to accept string-type workspace IDs. r?mstange!

Depends on D67822

Comment 30

[Mike de Boer [:mikedeboer]]

Attachment: Bug 890125 - Part 3 - Convert the Linux virtual desktop handling to accept string-type workspace IDs. r?stransky!

Depends on D67823

Comment 31

[Mike de Boer [:mikedeboer]]

Attachment: Bug 890125 - Part 4 - Implement virtual desktop switching for Windows 10 and above. r?mhowell!,emk!

Depends on D67824

Comment 32

[Martin Stránský [:stransky] (ni? me) (PTO, back on Feb 17)]

Mike, you didn't specify reviewers correctly, phabricator sets them as empty. Do you want me to review the patch or are you going to do other changes there?

Comment 33

[Mike de Boer [:mikedeboer]]

Yeah, that was all intentional, sorry for not mentioning that explicitly! This is a set of WIP patches that I intend to submit for review whenever I get the chance to make them compile and work on all platforms. That's quite an essential bit right there 😉

Comment 34

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/08359730557d
Part 1 - Change Window widget base classes to return and use strings for workspace IDs, rather than integers. r=nika
https://hg.mozilla.org/integration/autoland/rev/cd850f6d0b0b
Part 2 - Convert the OSX Spaces handling to accept string-type workspace IDs. r=mstange,nika
https://hg.mozilla.org/integration/autoland/rev/eceeef4fc062
Part 3 - Convert the Linux virtual desktop handling to accept string-type workspace IDs. r=stransky,nika
https://hg.mozilla.org/integration/autoland/rev/1c8115a9a684
Part 4 - Implement virtual desktop switching for Windows 10 and above. r=mhowell,emk

Comment 35

[Oana Pop-Rus]

Backed out 4 changesets (bug 890125) for Windows MinGW build bustages tier2 failures.

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=autoland&resultStatus=pending%2Crunning%2Csuccess%2Csuperseded%2Ctestfailed%2Cbusted%2Cexception%2Crunnable&searchStr=windows%2Cmingw%2Call%2Cmingw-clang%2Cbuilds%2Cfor%2Cwindows%2C64-bits%2Cbuild-win64-mingwclang%2Fdebug%2Cwmc64%28bd%29&fromchange=43f3b4e931163c71d18a51fc49db46d984995694&tochange=b1a9655f50e3d780c72463f89a21c1aa0a9c80c2&selectedJob=296172494

Failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=296172494&repo=autoland&lineNumber=94867

Backout: https://hg.mozilla.org/integration/autoland/rev/449fd8a4035ec5f19127f2d042c6ee6d7b6bf740

Comment 36

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/6e71761c1018
Part 1 - Change Window widget base classes to return and use strings for workspace IDs, rather than integers. r=nika
https://hg.mozilla.org/integration/autoland/rev/17b15e227d3b
Part 2 - Convert the OSX Spaces handling to accept string-type workspace IDs. r=mstange,nika
https://hg.mozilla.org/integration/autoland/rev/6e2c1facb7e0
Part 3 - Convert the Linux virtual desktop handling to accept string-type workspace IDs. r=stransky,nika
https://hg.mozilla.org/integration/autoland/rev/68475f12d4f1
Part 4 - Implement virtual desktop switching for Windows 10 and above. r=mhowell,emk

Comment 37

[Andrei Ciure[:aciure]]

https://hg.mozilla.org/mozilla-central/rev/6e71761c1018
https://hg.mozilla.org/mozilla-central/rev/17b15e227d3b
https://hg.mozilla.org/mozilla-central/rev/6e2c1facb7e0
https://hg.mozilla.org/mozilla-central/rev/68475f12d4f1

Comment 38

[Bogdan Tara[:bogdan_tara | bogdant]]

Backed out 4 changesets (bug 890125) for central bustages complaining about nsWindow.cpp

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=mozilla-central&revision=004536f666bfc2acf7002da703f003216f8aa5a5&searchStr=build&selectedJob=296503420

Backout link: https://hg.mozilla.org/integration/autoland/rev/d8da923e24edc189f72b44945065818aef2090cb

Failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=296503420&repo=mozilla-central&lineNumber=43317

[task 2020-04-06T22:10:47.219Z] 22:10:47     INFO -  make[4]: Leaving directory '/builds/worker/workspace/obj-build/layout/forms'
[task 2020-04-06T22:10:51.363Z] 22:10:51     INFO -  make[4]: Entering directory '/builds/worker/workspace/obj-build/widget/gtk'
[task 2020-04-06T22:10:51.368Z] 22:10:51     INFO -  /builds/worker/fetches/sccache/sccache /builds/worker/fetches/clang/bin/clang++ -std=gnu++17 -m32 -o nsWindow.o -c  -I/builds/worker/workspace/obj-build/dist/stl_wrappers -I/builds/worker/workspace/obj-build/dist/system_wrappers -include /builds/worker/checkouts/gecko/config/gcc_hidden.h -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -fstack-protector-strong -DDEBUG=1 -DOS_POSIX=1 -DOS_LINUX=1 -DCAIRO_GFX '-DMOZ_APP_NAME="firefox"' -DMOZ_HAS_MOZGLUE -DMOZILLA_INTERNAL_API -DIMPL_LIBXUL -DSTATIC_EXPORTABLE_JS_API -I/builds/worker/checkouts/gecko/widget/gtk -I/builds/worker/workspace/obj-build/widget/gtk -I/builds/worker/workspace/obj-build/ipc/ipdl/_ipdlheaders -I/builds/worker/checkouts/gecko/ipc/chromium/src -I/builds/worker/checkouts/gecko/ipc/glue -I/builds/worker/checkouts/gecko/layout/base -I/builds/worker/checkouts/gecko/layout/generic -I/builds/worker/checkouts/gecko/layout/xul -I/builds/worker/checkouts/gecko/other-licenses/atk-1.0 -I/builds/worker/checkouts/gecko/widget -I/builds/worker/checkouts/gecko/widget/headless -I/builds/worker/checkouts/gecko/widget/x11 -I/builds/worker/workspace/obj-build/dist/include -I/builds/worker/workspace/obj-build/dist/include/nspr -I/builds/worker/workspace/obj-build/dist/include/nss -fPIC -DMOZILLA_CLIENT -include /builds/worker/workspace/obj-build/mozilla-config.h -Qunused-arguments -Qunused-arguments -Wall -Wbitfield-enum-conversion -Wempty-body -Wignored-qualifiers -Woverloaded-virtual -Wpointer-arith -Wshadow-field-in-constructor-modified -Wsign-compare -Wtype-limits -Wunreachable-code -Wunreachable-code-return -Wwrite-strings -Wno-invalid-offsetof -Wclass-varargs -Wempty-init-stmt -Wfloat-overflow-conversion -Wfloat-zero-conversion -Wloop-analysis -Wc++2a-compat -Wcomma -Wimplicit-fallthrough -Wunused-function -Wunused-variable -Werror=non-literal-null-conversion -Wstring-conversion -Wtautological-overlap-compare -Wtautological-unsigned-enum-zero-compare -Wtautological-unsigned-zero-compare -Wno-error=tautological-type-limit-compare -Wno-inline-new-delete -Wno-error=deprecated-declarations -Wno-error=array-bounds -Wno-error=backend-plugin -Wno-error=return-std-move -Wno-error=atomic-alignment -Wformat -Wformat-security -Wno-gnu-zero-variadic-macro-arguments -Wno-unknown-warning-option -D_GLIBCXX_USE_CXX11_ABI=0 -fno-sized-deallocation -fno-aligned-new -fcrash-diagnostics-dir=/builds/worker/artifacts -march=pentium-m -msse -msse2 -mfpmath=sse -fno-exceptions -fno-strict-aliasing -fno-rtti -ffunction-sections -fdata-sections -fno-exceptions -fno-math-errno -pthread -pipe -g -Xclang -load -Xclang /builds/worker/workspace/obj-build/build/clang-plugin/libclang-plugin.so -Xclang -add-plugin -Xclang moz-check -Os -fno-omit-frame-pointer -funwind-tables -Werror -I/builds/worker/checkouts/gecko/widget/gtk/compat-gtk3 -pthread -I/usr/include/gtk-3.0 -I/usr/include/atk-1.0 -I/usr/include/at-spi2-atk/2.0 -I/usr/include/pango-1.0 -I/usr/include/gio-unix-2.0/ -I/usr/include/cairo -I/usr/include/gdk-pixbuf-2.0 -I/usr/include/glib-2.0 -I/usr/lib/i386-linux-gnu/glib-2.0/include -I/usr/include/harfbuzz -I/usr/include/freetype2 -I/usr/include/pixman-1 -I/usr/include/libpng12 -I/usr/include/gtk-3.0/unix-print -pthread -I/usr/include/gtk-3.0 -I/usr/include/atk-1.0 -I/usr/include/at-spi2-atk/2.0 -I/usr/include/pango-1.0 -I/usr/include/gio-unix-2.0/ -I/usr/include/cairo -I/usr/include/gdk-pixbuf-2.0 -I/usr/include/glib-2.0 -I/usr/lib/i386-linux-gnu/glib-2.0/include -I/usr/include/harfbuzz -I/usr/include/freetype2 -I/usr/include/pixman-1 -I/usr/include/libpng12 -I/usr/include/libdrm -I/usr/include/dbus-1.0 -I/usr/lib/i386-linux-gnu/dbus-1.0/include -I/usr/include/glib-2.0 -I/usr/lib/i386-linux-gnu/glib-2.0/include -Wno-error=shadow -fexperimental-new-pass-manager  -MD -MP -MF .deps/nsWindow.o.pp   /builds/worker/checkouts/gecko/widget/gtk/nsWindow.cpp
[task 2020-04-06T22:10:51.369Z] 22:10:51    ERROR -  /builds/worker/checkouts/gecko/widget/gtk/nsWindow.cpp:1630:15: error: call to member function 'AppendInt' is ambiguous
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -    workspaceID.AppendInt(wm_desktop[0]);
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -    ~~~~~~~~~~~~^~~~~~~~~
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:652:8: note: candidate function
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -    void AppendInt(int32_t aInteger) { AppendIntDec(aInteger); }
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:662:8: note: candidate function
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -    void AppendInt(uint32_t aInteger) { AppendIntDec(aInteger); }
[task 2020-04-06T22:10:51.369Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.370Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:672:8: note: candidate function
[task 2020-04-06T22:10:51.370Z] 22:10:51     INFO -    void AppendInt(int64_t aInteger) { AppendIntDec(aInteger); }
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:682:8: note: candidate function
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -    void AppendInt(uint64_t aInteger) { AppendIntDec(aInteger); }
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:653:8: note: candidate function not viable: requires 2 arguments, but 1 was provided
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -    void AppendInt(int32_t aInteger, int aRadix) {
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:663:8: note: candidate function not viable: requires 2 arguments, but 1 was provided
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -    void AppendInt(uint32_t aInteger, int aRadix) {
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.371Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:673:8: note: candidate function not viable: requires 2 arguments, but 1 was provided
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -    void AppendInt(int64_t aInteger, int aRadix) {
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -  /builds/worker/workspace/obj-build/dist/include/nsTSubstring.h:683:8: note: candidate function not viable: requires 2 arguments, but 1 was provided
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -    void AppendInt(uint64_t aInteger, int aRadix) {
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -         ^
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -  1 error generated.
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -  /builds/worker/checkouts/gecko/config/rules.mk:750: recipe for target 'nsWindow.o' failed
[task 2020-04-06T22:10:51.372Z] 22:10:51    ERROR -  make[4]: *** [nsWindow.o] Error 1
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -  make[4]: Leaving directory '/builds/worker/workspace/obj-build/widget/gtk'
[task 2020-04-06T22:10:51.372Z] 22:10:51     INFO -  /builds/worker/checkouts/gecko/config/recurse.mk:74: recipe for target 'widget/gtk/target-objects' failed
[task 2020-04-06T22:10:51.372Z] 22:10:51    ERROR -  make[3]: *** [widget/gtk/target-objects] Error 2

Comment 39

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

The bustage affected only Linux 32-bit.
Backout merged: https://hg.mozilla.org/mozilla-central/rev/d8da923e24ed

Comment 40

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/bd936bfc3a0f
Part 1 - Change Window widget base classes to return and use strings for workspace IDs, rather than integers. r=nika
https://hg.mozilla.org/integration/autoland/rev/36e331c78398
Part 2 - Convert the OSX Spaces handling to accept string-type workspace IDs. r=mstange,nika
https://hg.mozilla.org/integration/autoland/rev/70b5f44adb00
Part 3 - Convert the Linux virtual desktop handling to accept string-type workspace IDs. r=stransky,nika
https://hg.mozilla.org/integration/autoland/rev/8d31ce849dfb
Part 4 - Implement virtual desktop switching for Windows 10 and above. r=mhowell,emk

Comment 41

[Narcis Beleuzu [:NarcisB]]

https://hg.mozilla.org/mozilla-central/rev/bd936bfc3a0f
https://hg.mozilla.org/mozilla-central/rev/36e331c78398
https://hg.mozilla.org/mozilla-central/rev/70b5f44adb00
https://hg.mozilla.org/mozilla-central/rev/8d31ce849dfb

Comment 43

[Roy Orbitson]

Is this supposed to have landed yet? I'm on Dev Edition 79.0b2 (64-bit), use Windows' built-in virtual desktops, and I still have to shunt my windows to the correct desktop every time I open Firefox.

Comment 44

[Malte Forkel]

Does not work for me either with regular Windows virtual desktops: All Firefox windows opn up on the first virtual desktop after restart (Windows 10 1909 64-bit, Firefox 78.0.1)

Comment 45

[Roy Orbitson]

A pref, maybe?

Comment 46

[Martijn]

It says fixed in Firefox 77, but it's not quite fixed yet (actually, not at all).

• Browser windows on a VD other than 1 will always start at VD 1.
• A maximized secondary browser window on a secondary monitor that is minimized, when the browser restarts, will start "restored" (meaning neither maximized nor minimized).

Comment 47

[LoneDev]

I can't believe this is still an issue, I'm tired of having to move all Firefox windows to each virtual desktop each time I restart my PC

Comment 48

[ericrboidi]

Yeah this is 100% not fixed.

Comment 49

[Mathew Hodson]

This bug is closed. Further issues should be tracked in bug 1643936 instead.