WIP: Upgrade Chromium from 74.0.3729.131 to Chromium 75.0.3770.18 #2374
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
28 tasks
mkarolin
force-pushed
the
75.0.3770.18
branch
2 times, most recently
from
67d22e6 to
3ac50e4
Compare
25 tasks
mkarolin
force-pushed
the
75.0.3770.18
branch
3 times, most recently
from
65c0d37 to
b516336
Compare
Chromium change: commit ebb902010f96a6c192d8dbb17a1ceac8ed1d783a Author: Christopher Lam <[email protected]> Date: Thu Apr 4 03:42:36 2019 +0000 Move OneShotEvent into base/. This CL moves the OneShotEvent primitive into base/, as it now satisfies the rule of 3 (chromecast/, extensions/, chrome/).
Due to Chromium change: commit 0f5bdf53856d1891336823bb76a901ee0593b2d1 Author: Mounir Lamouri <[email protected]> Date: Thu Apr 4 16:32:35 2019 +0000 Move MediaPlayerId to its own file. This avoids including web_contents_observer.h for consumers of the struct that aren't WebContentsObserver. Bug: 947806
Patch NativeExtensionBindingsSystem to have virtual destrcutor. This class no longer inhertis from a base class, so we need to explicitly make the destructor virtual so that we can subclass. Also patch dispatcher.cc as redefinition in chromium_src is replacing occurrances of NativeExtensionBindingsSystem that shouldn't be replaced. Chromium change: commit 095dc1341b05f736b93466941fa455d075efe8d5 Author: Devlin Cronin <[email protected]> Date: Fri Mar 29 19:43:30 2019 +0000 [Extensions Bindings] Consolidate ExtensionBindingsSystem Since native bindings are now the only bindings system for extensions, remove the base class of ExtensionBindingsSystem, and conslidate all functionality in NativeExtensionBindingsSystem: - Remove extension_bindings_system.* - De-virtualize methods and transfer comments to NativeExtensionBindingsSystem - Move functionality from extension_bindings_system.cc to native_extension_bindings_system.cc - Update references and includes to point to NativeExtensionBindingsSystem Bug: 938998
Chromium change: commit be09e054ddeb4d9c402ae6a071143a286a615c9e Author: Mario Sanchez Prada <[email protected]> Date: Thu Apr 4 16:11:22 2019 +0000 Remove net/base/completion_callback.h, which is no longer used After the cleanup work done to migrate away from net::CompletionCallback() and use net::Completion{Once|Repeating}Callback instead, this file is not needed anymore, so let's remove it to prevent accidental new usages of those generic callbacks, and force using the new ones. Bug: 807724
As indicated in the message below, these are now only intended for tests. A review of our code using these may be needed. Chromium change: commit c113169ee60372b8fe17f0f449860032bf3b859d Author: danakj <[email protected]> Date: Thu Mar 28 15:19:05 2019 +0000 Move Shell messages to web_test and rename to BlinkTest. These messages are only used by web_test/blink_test_{controller,runner}. So move them into common/web_test, to indicate they are for tests not content shell. Also give them their own target so that test_runner can be made to use them, since they are only for tests. I believe this would let us move BlinkTestRunner into test_runner, and remove WebTestDelegate interface out of content/shell/renderer. [email protected] TBR=inferno Bug: 866140
Chromium change: commit 024afb89267a5eb6200cc22e1e752f70ef22b674 Author: Wei Li <[email protected]> Date: Fri Mar 15 20:26:35 2019 +0000 Change GetTooltipText() to return the value directly Instead of using an output parameter of pointer type, this CL changes GetTooltipText() function to directly return the tooltip's value. There is one case where the return value is used by ChromeOS ShelfButton to indicate whether the tooltip should not be drawn by View. We added one API for such case. BUG=155933
Chromium change: commit fc21834b89ddf98ab536e2d927101c978545c292 Author: Tim Volodine <[email protected]> Date: Fri Mar 8 15:34:38 2019 +0000 [AW NS] Add support for custom intercept-only URLLoaderFactory when handling external protocols. In Android WebView we need to be able to intercept so-called external schemes, i.e. schemes that are not known to the content/ layer and for which HandleExternalProtocol() is invoked. Examples of such schemes include: generally unsupported schemes and "cid:" schemes. This patch: - Adds support for providing a custom URLLoader when HandleExternalProtocol is invoked. - Adds the concept of intercept-only AwProxyingURLLoaderFactory, i.e. a factory that only loads responses from the interceptor and does not proceed otherwise (i.e. no actual network traffic). This type of factory/loader is used when handling schemes not supported by content/ loader. - Factors out InterceptedResponse::SendErrorAndCompleteImmediately method. - Removes OnReceivedErrorOnUIThread callback from aw_content_browser_client. - Make sure to notify InterceptedRequest loader when the shouldInterceptRequest response input stream failed to load. In that case produce an appropriate error and stop immediately. Fixed tests: -AwContentsClientShouldInterceptRequestTest.testCalledForUnsupportedSchemes -AwContentsClientShouldInterceptRequestTest.testContentIdIframe BUG=893566,841556
Chromium change: commit d998539fbd8b8beacd48432335e683f73d0603ed Author: Jared Saul <[email protected]> Date: Tue Mar 19 15:11:17 2019 +0000 [Autofill] Split autofill_payments_features from autofill_features Also adds jsaul@ as OWNERS for autofill_payments_features.* Bug: 940042
Chromium changes: commit 3b5078ddea27c6ad15275c66a78596283343600a Author: Gayane Petrosyan <[email protected]> Date: Tue Mar 26 23:35:51 2019 +0000 Remove COLOR_CONTROL_BACKGROUND from theme properties. COLOR_CONTROL_BACKGROUND always defaults to white and doesn't have calculated value based on other theme properties. Therefore it shouldn't be part of the theme properties. This change removes COLOR_CONTROL_BACKGROUND from theme properties and its only call in browser_view as it has no effect. commit c0d9e6fb635067afd79d76f226a22944e373d409 Author: Kyle Milka <[email protected]> Date: Thu Mar 21 22:35:04 2019 +0000 Remove theme properties for the detached bookmark bar The detached bookmark bar has been removed, also remove the theme properties associated with it. Bug: 941196
LocationBarModel::input_in_progress has been removed. Chromium change: commit 3f2285d27ee93d5a69ba49ab92cdf527f324c877 Author: Tommy C. Li <[email protected]> Date: Fri Apr 5 19:13:58 2019 +0000 Omnibox: Code health - Remove LocationBarModel::input_in_progress() LocationBarModel was originally designed to provide data on the current navigation entry's state, as described by the class comment. Eventually, it grew to track input_in_progress() state as well, duplicating this state canonically tracked by OmniboxEditModel. This CL removes this method. All of the callsites (other than a few) have already been migrated to calling the View to determine if user input is in progress. This is part of the overall UI architecture effort. Bug: 912722
Now takes full CookieOptions instead of a bool. Chromium change: commit fdbc8beebe3e89c3e7c1a9ca2d6b039ba784fca4 Author: Maks Orlovich <[email protected]> Date: Mon Mar 18 18:34:52 2019 +0000 Pass in full Options to SetCanonicalCookieAsync This will be used to check whether same site cookies are being set in the proper context. This change alters all the callsites to include whether it should be trusted to set such cookies or not, so it's not a simple/mechanical transformation. (The old bool for httponly gets passed inside the options object as well, like it would be for the get). Bug: 837412
Chromium change: commit d5409e147575eec363aee4f3ecb4dd38e0728055 Author: Joshua Pawlicki <[email protected]> Date: Sat Apr 6 00:23:11 2019 +0000 Refactor unzipper and patcher to allow for non-mojo implementations. Supply a non-mojo unzipper/patcher in chrome/updater. Change-Id: Ic884058a1b7b1de5b6cd631153550dfccfec9873 Bug: 947606
Xml protocol hander has been removed, JSON is now used instead. Updated endspoints and allowed pattern in brave_static_redirect_network_delegate_helper.cc Chromium changeis: commit 8ac39328330738ded0e2fc1fe94e56efd7ed88d1 Author: Joshua Pawlicki <[email protected]> Date: Thu Apr 4 21:22:46 2019 +0000 Remove XML from update_client. Bug: 881076 commit 705087b2a392fdb79c0ffc298ded98397179e15d Author: Joshua Pawlicki <[email protected]> Date: Tue Mar 12 15:40:57 2019 +0000 Launch JSON Omaha protocol. Bug: 881076 https://bugs.chromium.org/p/chromium/issues/detail?id=881076
Chromium change: commit 03c6b7855dffb5c16f431ce383024da1db3945a7 Author: Lei Zhang <[email protected]> Date: Thu Mar 21 05:22:24 2019 +0000 Remove default argument from NotificationDisplayService::Display(). Default arguments are banned on virtual functions.
ExtensionActionViewController now requires a valid instance of ToolbarActionBar. To get around it patched ExtensionActionViewController to not DCHECK for it and patched the class definition to allow for overriding a number of methods that use the passed in instance of the ToolbarActionBar. Overrode those in BraveActionViewController to not call out into ToolbarActionBar. Chromium change: commit 4d71c399c2cfe34f45114e13ba83e293122ecf2c Author: Peter Boström <[email protected]> Date: Wed Apr 10 00:21:04 2019 +0000 Add extension pop-out to Extension menu This changes ExtensionActionViewController to be aware of a main ToolbarActionsBar, which it asks directly in ::GetPreferredPopupViewController() instead of delegating into the owner's ::GetMainControllerForAction(), which is now gone. Separating the main bar from the owning bar allows ExtensionMenuView to make use of it even though it's not a ToolbarActionsBar. Bug: chromium:943702
* Incognito profile menu is now in a separate class, so need to add Tor menu to it separately. * ProfileChooseView is now instantiated in ProfileMenuViewBase so instantiate our subclasses there. * Menu is no longer built from constructor so we can use overridden method to fix the profile button instead of casting. * We only need to add the button to the menu_items now, don't need to worry about the layout. * Switch to 'using' base class constructors since we don't do anything in the subclassed ones. * ResetView was renamed to Reset. Chromium changes: commit 10f5093c58bc6785735a4d67498e24a6af81c5da Author: Ramin Halavati <[email protected]> Date: Thu Mar 28 09:58:30 2019 +0000 Separate content generation and formatting of ProfileChooserView. ProfileChooserView generates several menus and draws them. Drawing is done using different code paths causing difficulty in maintenance of similar output. This CL separates content generation code and menu formatting to make a unified code path for drawing all menus and prepare the class for separation of Incognito menu (which is used on all desktop platform) from Profile menu which is not used on Chrome OS. Bug: 934689 commit ef75f6cddf70f78459048c6c0eb48919c2864b45 Author: Ramin Halavati <[email protected]> Date: Tue Apr 2 19:24:51 2019 +0000 Separate Incognito menu from Profile menu. ProfileChooserView created all menus that are generated when user clicks on avatar menu. Except incognito menu, the other menus are not used in ChromeOS and have dependencies that are not supported on ChromeOS. To add the incognito menu to ChromeOS, this menu is separated into a different class. Bug: 934689
Chromium change: commit 52fa3aed17a23a6d1dc7459879c9fb2abc689bc8 Author: Gabriel Charette <[email protected]> Date: Mon Apr 15 21:44:37 2019 +0000 [base] Rename TaskScheduler to ThreadPool Reason: with the advent of other scheduling primitives in //base (i.e. SequenceManager), TaskScheduler was no longer the only component responsible for scheduling tasks. We will from now on refer to the whole of //base/task as the "task scheduling infrastructure". There are other types named "TaskScheduler" outside of base:: so s/TaskScheduler/ThreadPool/ across the codebase wasn't possible. Instead, this CL did: 1) base/task/task_scheduler => base/task/thread_pool (catches all files with includes) 1.1) Careful manual search to add files without includes (e.g. missing IWYU, docs, etc.) 2) TaskScheduler => ThreadPool in all files affected by (1) 3) task_scheduler => thread_pool in all files affected by (1) 4) "task scheduler" => "thread pool" in all files affected by (1) 4) Move task_scheduler_util like headers in //content //components and //ios Also: * Renamed UMA metrics from TaskScheduler.* to ThreadPool.* and dropped "Pool" from worker pool name suffixes. * Renamed TaskScheduler*Worker thread names to ThreadPool*Worker * In base/android: NativeTaskScheduler => NativeScheduler as it was referring to the whole of base/task. TaskSchedulerTest.java => NativePostTaskTest.java (former DNE) * Intentionally ignoring IWYU violations in this already too large CL. In follow-up: * Rename other types as well: SchedulerWorker => WorkerThread SchedulerWorkerPool* => WorkerThreadGroup* Bug: 951388
Chromium change: commit aee5c031a69ebb58dffc6618244c0cb8efbcbb39 Author: Daniel Cheng <[email protected]> Date: Thu Apr 18 23:49:05 2019 +0000 Make content::ResourceType an enum class. This allows us to be more explicit when content::ResourceType needs to be smuggled in an untyped int. It has the effect of introducing more static_cast<int>, but this makes the conversion sites explicit. This also allows several other incidental fixes: - the Mojo version of the enum no longer defines a sentinel value (which violates best practices in the security guidelines) - the C++ version of the enum can now define a compiler-enforced kMaxValue enumerator. - Since kMaxValue is no longer a distinct value, dummy handling for this value can be removed. This revealed a quirk in chrome/browser/predictors/loading_data_collector.cc
Chromium change: commit 136759f05184acd08fcb9b15cd9d8c1e9b8648e1 Author: John Lee <[email protected]> Date: Mon Apr 15 20:19:11 2019 +0000 Settings WebUI: Remove 'md' prefix from Settings code Bug: 887150
ChromeContentBrowserClient. Chromium change: commit 85079c267464cfb90c4f5dff217d7d515ded6139 Author: Xi Han <[email protected]> Date: Thu Apr 18 21:43:05 2019 +0000 Create Profile's PrefService before the Profile is created on Android. [1/4] This is a precursor CL for https://crrev.com/c/1532361. We introduce StartupData which owns all of the pre-create objects, including the ChromeFeatureListCreator. Bug: 937469
Added a replacement for kAccessibilityLabelsLearnMoreURL and kGoogleAccountChooserURL. Chromium changes: commit b7e1f19bb266074c9e454e5bf34e9a4d96fd7768 Author: Katie D <[email protected]> Date: Wed Apr 17 18:15:47 2019 +0000 Adds link to help center article for accessibility labels. The article is being written for M75. Bug: 953095 commit a588c5d073e2f9b96e039fa79ad20e6d5a623f9d Author: Thomas Tangl <[email protected]> Date: Wed Apr 10 16:16:08 2019 +0000 Don't show Google Account button in user menu when signed out of web Additional change: The account email is specified in the URL for the Google Account page, so the page always shows the account that is seen in the user menu. Bug: 950534
This prevents us from triggering a DCHECK for now, but crbug.com/754178 means to remove resource identifiers from content settings prefs API altogether. Chromium change: commit 116896fff47428ed53cb8daad883cc5fbda9c065 Author: Andy Paicu <[email protected]> Date: Tue Apr 2 09:29:48 2019 +0000 Add DCHECKS to ensure content settings API isn't used with resource id This was a reland of https://chromium-review.googlesource.com/790075/ which was reverted by https://chromium-review.googlesource.com/817993/ This removes one of the NOTREACHED assertions as it was causing issues. The reason the NOTREACHED assertion was reached is because it is still possible to have flash settings saved from before the ephemereal settings provider was introduced, but they are just not used. I've added a different DCHECK in GetRuleIterator to ensure the value is not used. As I'm pretty unfamiliar with the code any and all of the above might be wrong. Original CR message: Add DCHECKS to ensure content settings API isn't used with resource ids Using resource ids with the content settings API is supported but is apparently not being used. This change adds some DCHECKS to check that this is true as preparation for removing resource ids from the API. Bug: 754178
This script has been a noop for a while and now i18n_template.html has been removed completely, so the js file is no longer included in resources. Instead include i18n_template_no_process.js directly. Chromium change: commit 4b411e69edcdfda5fa1deb9aa3f54b1f1bd0558d Author: Igor Kobylin <[email protected]> Date: Fri Jan 25 21:45:01 2019 +0000 [i18n] Remove i18n_template.html All pages importing it have already been migrated to $i18n{}, so these dependencies don't seem needed anymore. This changes following pages: * chrome://media-router * chrome://md-user-manager [email protected] Bug: 677338
Chromium change added a NOTREACHED guard for accessing SystemRequestContext beyond its initial creation to help identify what still needs to transition to network service. The intent is to have this failure triggered when network service is enabled, however there is no check for network service feature state. Patching it to only trigger NOTREACHED check if network service is enabled. Chromium change: commit a97e0c861c150cecd77683417ddf49845eb47e6f Author: John Abd-El-Malek <[email protected]> Date: Tue Apr 2 23:36:58 2019 +0000 Stop creating the failing system URLRequestContext when network service is enabled. This isn't used by any code after r646920. Bug: 825242
Chromium change: commit 20c2c6baa6d9847db46eca1e0dcfaa3c915ebb92 Author: Robert Sesek <[email protected]> Date: Wed Apr 3 20:30:36 2019 +0000 Rename entitlements.plist to app-entitlements.plist There will be more than one entitlements file (e.g., for helper app bundles) in the future. This will allow differentiation, just like for the Info.plists. Bug: 850193
Chromium change: commit 0724dc56429fcd713d357151c744adfef5e43db1 Author: Charlene Yan <[email protected]> Date: Fri Apr 12 17:57:41 2019 +0000 Migrate flags from chrome/browser/browser_features to chrome/browser/ui/ui_features The flags moved are only used under chrome/browser/ui
On windows, std::ifstream constructor cannot take std::wstring param. The likely reason why this to no longer work is this Chromium change: commit ff480e1803a097f5485fbee9833edb04e96676e0 Author: Tom Anderson <[email protected]> Date: Mon Apr 29 17:31:15 2019 +0000 [Merge to M75] Reenable libc++ on Windows > BUG=801780 > R=thakis > > Change-Id: Ife6f11c4e02534c879731274d427b0f563b8fa08 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1556590 > Auto-Submit: Thomas Anderson <[email protected]> > Commit-Queue: Nico Weber <[email protected]> > Reviewed-by: Nico Weber <[email protected]> > Cr-Commit-Position: refs/heads/master@{#652740} BUG=801780 Also, affects vendor/bat-native-usermodel/test/usermodel_unittest.cc and is fixed by brave-intl/bat-native-usermodel#9 and updating DEPS with the fix's commit.
Chromium renamed md_extensions to extensions, we need to keep parity with their tree for clarity. Also, this makes sure we unpack resources into right places during build.
BraveSchemeLoadBrowserTest.NotAllowedToLoadTestByWindowOpenWithNoOpener was failing because it expects the JS it executes to pop up a new window. In c74 it was doing just that, but in c75 it opens a new tab in the same window. Adjusted the test to observe tab_strip_model instead of the BrowserList.
A settings-box we were overriding is now a cr-link-row. Chromium change: commit 1ff51064c917d3de7e454cc9792f980792326351 Author: dpapad <[email protected]> Date: Wed Mar 13 08:12:42 2019 +0000 Settings: Use cr-link-row in people and privace_page.html where possible. This not only removes redundant code, but also fixes the paddings when large fonts are used. Bug: 686905
|
Running into a weird situation here when trying to run browser tests on macOS
|
|
Closing in favor of #2401 |
gdregalo
added a commit
that referenced
this pull request
Jan 6, 2020
gdregalo
added a commit
that referenced
this pull request
Jan 6, 2020
Unverified Channels Marked as Verified.Issue #2374
gdregalo
added a commit
that referenced
this pull request
Feb 6, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix brave/brave-browser#4333
Related PR: brave/brave-browser#4334
It's easiest to review this per commit.
Each commit explains where the change came from in Chromium's history.
Special notes to QA:
See brave/brave-browser#4333 for areas of QA coverage.
Submitter Checklist:
npm test brave_unit_tests && npm test brave_browser_tests && npm run test-security) onnpm run lint)git rebase master(if needed).git rebase -ito squash commits (if needed).Reviewer Checklist:
After-merge Checklist:
changes has landed on.