chrome/browser/ui/views/frame/browser_desktop_window_tree_host_ozone.cc - Issue 2478323003: Ozone Linux: Fix linker errors in interactive_ui_tests

Side by Side Diff: chrome/browser/ui/views/frame/browser_desktop_window_tree_host_ozone.cc

Issue 2478323003: Ozone Linux: Fix linker errors in interactive_ui_tests (Closed)

Patch Set: Add CreateBrowserDesktopWindowTreeHost stub for ozone Created 4 years, 1 month ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

OLD	NEW
(Empty)
	1 // Copyright 2016 The Chromium Authors. All rights reserved.

	2 // Use of this source code is governed by a BSD-style license that can be

	3 // found in the LICENSE file.

	4

	5 #include "base/logging.h"

	6 #include "chrome/browser/ui/views/frame/browser_desktop_window_tree_host.h"

	7

	8 ////////////////////////////////////////////////////////////////////////////////

	9 // BrowserDesktopWindowTreeHost, public:

	10

	11 // static

	12 BrowserDesktopWindowTreeHost*

	13 BrowserDesktopWindowTreeHost::CreateBrowserDesktopWindowTreeHost(
	rjkroege 2016/11/07 19:54:15 What does --mash Chrome use in place of this? It a What does --mash Chrome use in place of this? It already operates in "desktop-like" mode. But perhaps can't draw itself without ash submitting a top-level compositor frame / per display. But maybe that's ok pending a TODO(tonikitoo) here? Maybe this code should submit the top-level compositor frame when running without ash? Tom (Use chromium acct) 2016/11/07 20:07:00 The only impls I see are BDWTHWin and BDWTHX11, so Show quoted text On 2016/11/07 19:54:15, rjkroege wrote: > What does --mash Chrome use in place of this? It already operates in > "desktop-like" mode. But perhaps can't draw itself without ash submitting a > top-level compositor frame / per display. But maybe that's ok pending a > TODO(tonikitoo) here? Maybe this code should submit the top-level compositor > frame when running without ash? > The only impls I see are BDWTHWin and BDWTHX11, so I guess mash chrome just never calls CreateBrowserDWTH, and this solves the linker error? rjkroege 2016/11/07 23:01:22 Yes. WindowPort and friends is the solution for th Show quoted text On 2016/11/07 20:07:00, Tom Anderson wrote: > On 2016/11/07 19:54:15, rjkroege wrote: > > What does --mash Chrome use in place of this? It already operates in > > "desktop-like" mode. But perhaps can't draw itself without ash submitting a > > top-level compositor frame / per display. But maybe that's ok pending a > > TODO(tonikitoo) here? Maybe this code should submit the top-level compositor > > frame when running without ash? > > > > The only impls I see are BDWTHWin and BDWTHX11, so I guess mash chrome just > never calls CreateBrowserDWTH, and this solves the linker error? Yes. WindowPort and friends is the solution for this with the --mash config. Tom (Use chromium acct) 2016/11/07 23:21:03 Acknowledged. Show quoted text On 2016/11/07 23:01:22, rjkroege wrote: > On 2016/11/07 20:07:00, Tom Anderson wrote: > > On 2016/11/07 19:54:15, rjkroege wrote: > > > What does --mash Chrome use in place of this? It already operates in > > > "desktop-like" mode. But perhaps can't draw itself without ash submitting a > > > top-level compositor frame / per display. But maybe that's ok pending a > > > TODO(tonikitoo) here? Maybe this code should submit the top-level > compositor > > > frame when running without ash? > > > > > > > The only impls I see are BDWTHWin and BDWTHX11, so I guess mash chrome just > > never calls CreateBrowserDWTH, and this solves the linker error? > > Yes. WindowPort and friends is the solution for this with the --mash config. Acknowledged.
	14 views::internal::NativeWidgetDelegate* native_widget_delegate,

	15 views::DesktopNativeWidgetAura* desktop_native_widget_aura,

	16 BrowserView* browser_view,

	17 BrowserFrame* browser_frame) {

	18 NOTIMPLEMENTED();
	Tom (Use chromium acct) 2016/11/07 19:40:45 I'm not sure what the right way to stub out Browse I'm not sure what the right way to stub out BrowserDesktopWindowTreeHost* is. I can see two possible implementations here 1. Have a single BrowserDesktopWindowTreeHostOzone which takes a BrowserDesktopWindowTreeHost in the constructor, and proxies all calls to it. 2. Have BrowserDesktopWindowTreeHostOzoneX11 : public DesktopWindowTreeHostOzoneX11, BrowserDesktopWindowTreeHostOzoneWayland : public DesktopWindowTreeHostOzoneWayland, etc... +rjkroege@ What do you guys think? rjkroege 2016/11/07 23:01:22 NOTREACHED(); We should never get here. Show quoted text On 2016/11/07 19:40:45, Tom Anderson wrote: > I'm not sure what the right way to stub out BrowserDesktopWindowTreeHost* is. > > I can see two possible implementations here > 1. Have a single BrowserDesktopWindowTreeHostOzone which takes a > BrowserDesktopWindowTreeHost in the constructor, and proxies all calls to it. > 2. Have BrowserDesktopWindowTreeHostOzoneX11 : public > DesktopWindowTreeHostOzoneX11, BrowserDesktopWindowTreeHostOzoneWayland : public > DesktopWindowTreeHostOzoneWayland, etc... > > +rjkroege@ > What do you guys think? NOTREACHED(); We should never get here. Tom (Use chromium acct) 2016/11/07 23:21:03 Done. Show quoted text On 2016/11/07 23:01:22, rjkroege wrote: > On 2016/11/07 19:40:45, Tom Anderson wrote: > > I'm not sure what the right way to stub out BrowserDesktopWindowTreeHost* is. > > > > I can see two possible implementations here > > 1. Have a single BrowserDesktopWindowTreeHostOzone which takes a > > BrowserDesktopWindowTreeHost in the constructor, and proxies all calls to it. > > 2. Have BrowserDesktopWindowTreeHostOzoneX11 : public > > DesktopWindowTreeHostOzoneX11, BrowserDesktopWindowTreeHostOzoneWayland : > public > > DesktopWindowTreeHostOzoneWayland, etc... > > > > +rjkroege@ > > What do you guys think? > > NOTREACHED(); We should never get here. Done.
	19 return nullptr;

	20 }

OLD	NEW

« no previous file with comments | « chrome/browser/ui/BUILD.gn ('k') | chrome/test/base/interactive_ui_tests_main.cc » ('j') | chrome/test/base/interactive_ui_tests_main.cc » ('J')