electron/patches/chromium/dcheck.patch
Electron Bot d8737734bf
chore: bump chromium to bc8f70ed4cfe2262ea833750eaddb (master) (#22649)
* chore: bump chromium in DEPS to 9321f32fb1b3af8fdfce55c7bbfacf3f75118dca

* Update patches

* Update electron_swiftshader_binaries deps

https://chromium-review.googlesource.com/c/chromium/src/+/2056931

* Use Promise with RequestPointerLock calls

https://chromium-review.googlesource.com/c/chromium/src/+/2069199

* Replace content::CursorInfo with ui::Cursor

https://chromium-review.googlesource.com/c/chromium/src/+/1999201

* Convert MaterialDesignController to a true singleton.

https://chromium-review.googlesource.com/c/chromium/src/+/2090877

* Drop WebContentsView::RenderViewCreated hook

https://chromium-review.googlesource.com/c/chromium/src/+/2093535

* chore: bump chromium in DEPS to 6478123cfa0102ed754c70eb9bbdd391d676a4dd

* Splitting context_menu_params.h into separate browser VS common parts.

https://chromium-review.googlesource.com/c/chromium/src/+/2097468

* Fix DCHECK on OnThemeChanged()

https://chromium-review.googlesource.com/c/chromium/src/+/2090713

* chore: bump chromium in DEPS to b0269bb003f699bc8ea7dcba8b0795ef963696d7

* Remove no longer needed patch

* Check PointerLock requests for new options and update accordingly

https://chromium-review.googlesource.com/c/chromium/src/+/2071788

* Address issues from review

* Fixup compile error

* Add additional library files

* chore: bump chromium in DEPS to a41285fb8aebc8f70ed4cfe2262ea833750eaddb

* Update patches

Co-authored-by: John Kleinschmidt <jkleinsc@github.com>
2020-03-14 16:54:14 -04:00

65 lines
3.4 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: deepak1556 <hop2deep@gmail.com>
Date: Thu, 20 Sep 2018 17:44:38 -0700
Subject: dcheck.patch
This disables some debug checks which currently fail when running the Electron
test suite. In general there should be sustained effort to have all debug checks
enabled. If you want to help, choose one of the diffs below and remove it. Then
build Electron (debug configuration) and make sure all tests pass on the CI
systems. Unfortunately the tests don't always cover the check failures, so it's
good to also run some non-trivial Electron app to verify.
Apart from getting rid of a whole diff, you may also be able to replace one diff
with another which enables at least some of the previously disabled checks. For
example, the checks might be disabled for a whole build target, but actually
only one or two specific checks fail. Then it's better to simply comment out the
failing checks and allow the rest of the target to have them enabled.
diff --git a/content/browser/frame_host/navigation_controller_impl.cc b/content/browser/frame_host/navigation_controller_impl.cc
index f950552abbe9e9a0002156b596438fe9aa4f01b9..23931899eb58432fd8eabf4d681d2294a88fd97d 100644
--- a/content/browser/frame_host/navigation_controller_impl.cc
+++ b/content/browser/frame_host/navigation_controller_impl.cc
@@ -1250,8 +1250,10 @@ NavigationType NavigationControllerImpl::ClassifyNavigation(
return NAVIGATION_TYPE_NEW_SUBFRAME;
}
- // We only clear the session history when navigating to a new page.
- DCHECK(!params.history_list_was_cleared);
+ // Electron does its own book keeping of navigation entries and we
+ // expect content to not track any, by clearing history list for
+ // all navigations.
+ // DCHECK(!params.history_list_was_cleared);
if (rfh->GetParent()) {
// All manual subframes would be did_create_new_entry and handled above, so
@@ -1506,7 +1508,10 @@ void NavigationControllerImpl::RendererDidNavigateToNewPage(
new_entry->GetFavicon() = GetLastCommittedEntry()->GetFavicon();
}
- DCHECK(!params.history_list_was_cleared || !replace_entry);
+ // Electron does its own book keeping of navigation entries and we
+ // expect content to not track any, by clearing history list for
+ // all navigations.
+ // DCHECK(!params.history_list_was_cleared || !replace_entry);
// The browser requested to clear the session history when it initiated the
// navigation. Now we know that the renderer has updated its state accordingly
// and it is safe to also clear the browser side history.
diff --git a/ui/base/clipboard/clipboard_win.cc b/ui/base/clipboard/clipboard_win.cc
index 23681d4192ac329a68822dbcdb06369f34e5e28a..a6f07372bdc576ad8d2cce0b04ffe2adc9ab37ad 100644
--- a/ui/base/clipboard/clipboard_win.cc
+++ b/ui/base/clipboard/clipboard_win.cc
@@ -767,10 +767,10 @@ void ClipboardWin::WriteBitmapFromHandle(HBITMAP source_hbitmap,
void ClipboardWin::WriteToClipboard(ClipboardFormatType format, HANDLE handle) {
UINT cf_format = format.ToFormatEtc().cfFormat;
- DCHECK_NE(clipboard_owner_->hwnd(), nullptr);
+ // DCHECK_NE(clipboard_owner_->hwnd(), nullptr);
if (handle && !::SetClipboardData(cf_format, handle)) {
- DCHECK_NE(GetLastError(),
- static_cast<unsigned long>(ERROR_CLIPBOARD_NOT_OPEN));
+ // DCHECK_NE(GetLastError(),
+ // static_cast<unsigned long>(ERROR_CLIPBOARD_NOT_OPEN));
FreeData(cf_format, handle);
}
}