Upgrade Chromium from 73.0.3683.75 to Chromium 74.0.3729.22 #2013
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
16 tasks
mkarolin
commented
Mar 19, 2019
| {{declare_initial(property.name.to_upper_camel_case() if not header)|indent(spaces, true)}} { | ||
| {{caller(property)|indent(spaces, true)-}} | ||
| -{{'}'|indent(spaces, true)}} | ||
| +{{'\n}'|indent(spaces, true)}} |
There was a problem hiding this comment.
The initial patch was:
commit d1410d069d22188896cdc13a526b3732cc21c90b Author: Brian R. Bondy Date: Fri Feb 15 00:03:08 2019 -0500 Add newline on line ending for } for generated file This was causing an error on some computers when building (Anthony and me)
In C74 the code has been changed to:
{% if property.style_builder_inline == header and property.style_builder_generate_initial %}
{{declare_initial(property.name.to_upper_camel_case() if not header)|indent(spaces, true)}} {
{{(caller(property) ~ '}')|indent(spaces, true)}}
19 tasks
Chromium's Service Manager no longer supports JSON manifests. All JSON manifests and overlays need to be converted to C++. See: https://bugs.chromium.org/p/chromium/issues/detail?id=895616 This change affects the following services: - bat_ads - bat_ledger - tor_launcher
base::JSONReader::ReadAndReturnError.
base::JSONReader changed signatue of Read to return Optional<Value>
instead of std::unique_ptr<Value>. ReadAndReturnError has been
deprecated and replaced with ReadAndReturnValueWithError that
returns a struct with error instead of having multiple out parameters.
Chromium chages:
commit 025edc25d8c8649d5a8dfd062e952855b3dfa13e
Author: Lei Zhang <thestig@chromium.org>
Date: Sat Feb 16 05:02:25 2019 +0000
Change base::JSONReader::Read()/ReadAndReturnError() to return Optional.
BUG=925165
commit 015397daa81409c43afbcd11ac41978cd7175cd0
Author: Lei Zhang <thestig@chromium.org>
Date: Wed Feb 20 04:26:59 2019 +0000
Change base::JSONReader::ReadAndReturnError() to not use out params.
Instead of returning Optional<Value> and putting the error status in 4
out parameters, return struct ValueWithError which contains the value
and the error status.
ContextLifecycleStateObserver.
Chromium change:
commit 6bbacc4529f0ed67a55e49128b927ea1c3e5d33a
Author: Dave Tapuska <dtapuska@chromium.org>
Date: Mon Feb 11 20:00:17 2019 +0000
Rename PausableObject to ContextLifecycleStateObserver
Get rid of PauseState enum and use the FrameLifecycleState
BUG=907125
Tranport socket is now a meber of the base class and doesn't need to be
passed around. Destination is now a host/port pair.
Chromium changes:
commit 4b412da433b4b3e4cb70b00210395bbc73fbdfaa
Author: Matt Menke <mmenke@chromium.org>
Date: Fri Jan 25 19:31:12 2019 +0000
Socket Pools Refactor 8: Merge SOCKS socket pool into transport pool.
This CL merges the SOCKSClientSocketPool with the
TransportClientSocketPool it used to sit above. In doing this, the
SOCKSClientSocketPool class is removed, in favor of a
TransportClientSocketPool. SOCKSConnectJobs now directly create
TransportConnectJobs, instead of going through a socket pool.
This doesn't affect any other socket pools - there's still an SSL socket
pool on top of the socket pool for a SOCKS proxy, for instance.
This is part of an effort to flatten the socket pools.
https://docs.google.com/document/d/1g0EA4iDqaDhNXA_mq-YK3SlSX-xRkoKvZetAQqdRrxM/edit
Bug: 472729
commit 6b85fafc752ff33b9e92318f8577032897c9e40b
Author: Eric Orth <ericorth@chromium.org>
Date: Wed Feb 6 18:53:35 2019 +0000
Modernize DNS in SOCKS[5]ClientSocket
Convert all |destination| storage to use HostPortPair instead of
HostResolver::ResolveInfo, and convert Resolve() request to
CreateRequest().
Removed HangingHostResolverWithCancel from test as it didn't support
CreateRequest(), and a working equivalent already exists in
mock_host_resolver.h.
Bug: 922699
URLLoaderFactory method is replaced with GetNetworkFetcherFactory.
Chromium changes:
commit 75e6bf2135e7c42a24bed2343f63f5a07e17fb13
Author: Sorin Jianu <sorin@chromium.org>
Date: Tue Feb 12 16:07:12 2019 +0000
Introduce update_client::NetworkFetcher as a way to abstract the network.
This is a mechanical change and work in progress.
The goal here is to decouple update client from its Chrome network
dependencies such as //net and //services/network.
This change avoids inejction of a direct dependency of the servicified
network stack in favor of injecting a NetworkFetcherFactory, which
in turns can create instances of NetworkFetcher, where they are needed.
The next change will further isolate the implementation dependencies on
the Chrome network, and allow for selecting different implementations
of the NetworkFetcher.
Bug: 929167
commit 4e0b41d09eeb9cf79def67b567fa2cf396d7f339
Author: Sorin Jianu <sorin@chromium.org>
Date: Wed Feb 20 17:21:58 2019 +0000
Decouple update_client::NetworkFetcher interface from Chromium network.
This change is mechanical.
It enables NetworkFetcherFactory, which is injected by the embedder, and
which creates instances of NetworkFetcher.
Bug: 929167
Debug/DCHECK compilation now uses [-Werror,-Wextra-semi] which causes
semicolons after function definitions or namespaces to be treated as
errors. For example, void foo() {}; would produce an error due to the
trailing semicolon.
Chromium change:
commit a02a8f4b72fe5abad3c123a9c52c6ef78940de45
Author: Nico Weber <thakis@chromium.org>
Date: Sat Feb 23 04:13:21 2019 +0000
Reland "Enable -Wextra-semi in non-cros non-chromecast non-fuzzer linux builds that have DCHECKs enabled."
This is a reland of 5b911be859f216fe70bad5c41cc91a8684d773f6
Original change's description:
> Enable -Wextra-semi in non-cros non-chromecast non-fuzzer linux builds that have DCHECKs enabled.
>
> Only enables the warning for chromium_code.n
>
> Also fix the last few -Wextra-semi instances in sanitizer and linux/ozone builds.
>
> Getting to this state required fixing > 3000 unique warnings in 15+ different
> repositories. If this breaks some internal-only build somewhere, either consider
> using no_chromium_code for your internal code, or fix the warnings -- it's
> pretty easy. (Build with treat_warnings_as_errors=false, then open the error log
> in vim, run `:%g/warning:/t$` to copy the warnings to the end of the buffer,
> remove all but just those lines, run `:sort u` to get uniques, then save as
> `tmp.txt` and run `:cf %` and then use something like `:map m :w<cr>:cn<cr>`
> to go through them quickly.)
>
> TBR=rsesek
>
> Bug: 926235
> Change-Id: Ica629737523ff6bbd756edc44d182f35ff6cf8ac
> Reviewed-on: https://chromium-review.googlesource.com/c/1483396
> Reviewed-by: Scott Violet <sky@chromium.org>
> Auto-Submit: Nico Weber <thakis@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#634826}
TBR=rsesek
Bug: 926235, 935033
The functionality was called from override of
ChromeBrowserMainParts::PreShutdown, which was removed along with the
call to it from BrowserMainLoop::PreShutdown. This fix subclasses
BrowserMainLoop and calls BraveBrowserMainParts::PreShutdown from it.
It also patches BrowserMainParts to add virtual PreShutdown method
and BrowserMainRunnerImpl to instantiate BraveBrowserMainLoop instead
of BrowserMainLoop.
Chromium change:
commit 0f15e7a22545884fd259b873366a511686f4e5d9
Author: Alexei Filippov <alph@chromium.org>
Date: Thu Feb 14 21:34:28 2019 +0000
[heap profiler] Migrate UMA heap profiler off LegacyCallStackProfileBuilder
... and remove the latter as it was the last client.
BUG=916303,913570
BraveBrowserCommandController::ExecuteCommandWithDisposition.
Chromium change:
commit 6f6fc1e333f968d4caddb52f11b6ff9f3ef1c0f9
Author: Edwin Joe <ejoe@google.com>
Date: Wed Feb 27 20:00:37 2019 +0000
Added a new UMA to track tab switching latency time
Created a new TabSwitchEventLatencyRecorder class that will track
the latency between input event timestamp and the time when tab
switching begin processing, and upload the metric to a UMA.
Changed the TabStripModel::ActivateTabAt method interface to take
enough input event context from the Tab to the TabStripModel,
and changed all TabStripModel::ActivateTabAt callsites accordingly.
Bug: 921120
class constructor.
Also, browser_ pointer was moved to the parent class of
ProfileChooserView and now needs to be accessed via a getter.
Chromium change:
commit 6c85b61a18bad83016864a9a7b54c85d465aea90
Author: Ramin Halavati <rhalavati@chromium.org>
Date: Thu Feb 28 09:28:03 2019 +0000
Separate ProfileChooserView into two classes.
ProfileChooserView prepares several menus, including different flavors
of profile menu, guest menu, and incognito menu.
The class is now divided into two classes, ProfileMenuViewBase which is
in charge of the UI implementation and styling, and ProfileChooserView
which decides about the elements in the menu based on the selected
menu mode.
Bug: 934689
Chromium change:
commit bf2ef6b1f33e8c1653c50a2614e90be61c5bcc2e
Author: Lucas Furukawa Gadani <lfg@chromium.org>
Date: Wed Feb 20 18:38:39 2019 +0000
Remove const interfaces from content::ResourceRequestInfo.
Bug: 908139
secure/insecure.
Chromium change:
commit 44525ced669fe2c54d914d3df86c44453afd3402
Author: Maks Orlovich <morlovich@chromium.org>
Date: Mon Feb 25 14:17:58 2019 +0000
Reland "Take scheme in CookieStore::SetCanonicalCookieAsync, not just whether it's secure."
This reverts commit 90c86f0ce73498a900dbb601f1de95dc519de18a.
Justification for revert of revert: Fixed build/merged with a trunk change
Bug:850044
A bool for user gesture or not has been replaces with a struct.
Opted to use GestureType::kOther to maintain the previous value of 'true' that
we were passing in.
Chromium change:
commit 6f6fc1e333f968d4caddb52f11b6ff9f3ef1c0f9
Author: Edwin Joe <ejoe@google.com>
Date: Wed Feb 27 20:00:37 2019 +0000
Added a new UMA to track tab switching latency time
Created a new TabSwitchEventLatencyRecorder class that will track
the latency between input event timestamp and the time when tab
switching begin processing, and upload the metric to a UMA.
Changed the TabStripModel::ActivateTabAt method interface to take
enough input event context from the Tab to the TabStripModel,
and changed all TabStripModel::ActivateTabAt callsites accordingly.
Bug: 921120
Chromium change:
commit 9da196a5beb8fd4739939937953f9e8ba6f50770
Author: Lei Zhang <thestig@chromium.org>
Date: Fri Mar 1 17:01:27 2019 +0000
Make ShowConstrainedWebDialog() parameters unique_ptrs.
Do the same for ConstrainedDialogWebView. Fix lint errors along the way.
For base::Bind
base::DoNothing
base::ThreadTaskRunnerHandle
Chromium change:
commit 708977c9da301d08676f218a2163f382e3607276
Author: Robert Liao <robliao@chromium.org>
Date: Fri Mar 1 02:13:50 2019 +0000
Rename view_properties.* to view_class_properties.*
This is a mechanical execution of tools/git/mass-rename.py.
BUG=937018
ChromeComponentExtensionResourceManager::AddComponentResourceEntries now takes
GzippedGritResourceMap, so all the maps we pass in
BraveComponentExtensionResourceManager need to be gzipped.
Updated .grd files map/header directives to gzipped_resource...
Updated BasicUI to consume GzippedGritResourceMap.
Chromium change:
commit 20b9be0f2dcbece3929c1a7fa8da6b4a3d925f73
Author: Sam McNally <sammc@chromium.org>
Date: Wed Jan 30 02:29:00 2019 +0000
Add support for gzipped component app/extension resources.
Applied to the ink wasm module, this saves 3.9MB.
If applied to all Files app, Gallery etc. html/css/js resources this
could save a further 3.2MB.
Bug: 923204
…num.
Chromium change:
commit 193c09db4eb3b886512b3c43dfa3432f0fcef719
Author: Yao Xiao <yaoxia@chromium.org>
Date: Wed Feb 13 17:30:19 2019 +0000
Change allow_download in ResourceRequestInfoImpl to an enum
Motivation: Currently the |allow_download| flag in
|ResourceRequestInfoImpl| can be set to false in the circumstances of
Viewsource, Interstitial, etc. In all these cases, it's safe to
skip intercepting the resource before plugins handler is set up
(see MimeSniffingResourceHandler::MaybeStartInterception()). However,
this may not always be the case if we want to exploit this flag to
prevent download in other circumstances, e.g. in adframe. In this case,
bypassing the setup of plugin handler will also prevent a PDF to
display in adframe, although the restriction was to prevent just
download in adframe.
Therefore, I'm changing the boolean |allow_download| inside
|ResourceRequestInfoImpl| to 3 state enum, which allows us to differentiate
between a safe immediate prevention (Viewsource, Interstitial, etc.) that we
have always been doing, and a prevention that should happen only after the
plugin handler is checked and possibly set up (AdFrame).
The following CL gives an idea of how the new enum will be used:
https://chromium-review.googlesource.com/c/chromium/src/+/1367912
Bug: 929911
Updated consequent calls to GetList to GetAsList (soon to go away).
GetList now returns a vector<base::Value> and more code will need to be
refactored once GetAsList goes away complitely.
Chromium change:
commit 9dd8968271bf31642e97511c0b3cca00534a1ee0
Author: Lei Zhang <thestig@chromium.org>
Date: Tue Feb 19 20:14:53 2019 +0000
Change base::JSONReader::ReadToValue() to return Optional.
Update most unit tests to reflect this change, or change the
JSONReader().ReadToValue() pattern to JSONReader::Read().
Convert existing callers that did not get updated to use the deprecated
version of the API.
BUG=925165
… errors.
Debug/DCHECK compilation now uses [-Werror,-Wextra-semi] which causes
semicolons after function definitions or namespaces to be treated as
errors. For example, void foo() {}; would produce an error due to the
trailing semicolon.
Chromium change:
commit a02a8f4b72fe5abad3c123a9c52c6ef78940de45
Author: Nico Weber <thakis@chromium.org>
Date: Sat Feb 23 04:13:21 2019 +0000
Reland "Enable -Wextra-semi in non-cros non-chromecast non-fuzzer linux builds that have DCHECKs enabled."
This is a reland of 5b911be859f216fe70bad5c41cc91a8684d773f6
Bug: 926235, 935033
Chromium change:
commit 572755bab76f7d4aa697fffebbdcb03ac92a44cd
Author: Alexandre Frechette <frechette@chromium.org>
Date: Wed Feb 13 22:30:20 2019 +0000
Moving the core language prefs (kAcceptLanguages and kPreferredLanguages) to components/language.
Bug: 902354, 872096
This feature was disabled by brave/brave-browser#1051 to mitigate Spectre. Spectre was later addressed by Chromium via site isolation and Chromium re-enabled SharedArrayBuffer, but Brave didn't. Now in C74, WebAssemblyThreads feature is being enabled by default. Since WebAssemblyThreads is built with and auto-enables SharedArrayBuffer, there is no reason for us to continue keeping SharedArrayBuffer disabled. Aslo, removed a browser test for checking that SharedArrayBuffer is disabled. Chromium change making WebAssemblyThreads enabled by default: commit f9cc6b912e601070728b449892097cec0d59cbfa Author: Ben Smith <binji@chromium.org> Date: Wed Feb 27 02:09:43 2019 +0000 [wasm] Enable wasm threads by default See intent to ship blink-dev thread: https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/tD6np-OG2PU Bug: chromium:754910
When enabled this feature does not instantiate ChromeNetworkDelegate and by its virtue BraveNetworkDelegateBase which is where we add hooks into network stack. Related: brave/brave-browser#2351 Chromium change to not instantiate ChromeNetworkDelegate: commit 4b7d02f22b70c27f32fa5508f785ba310d08aad3 Author: John Abd-El-Malek < jam@chromium.org > Date: Thu May 03 17:46:35 2018 Don't instantiate ChromeNetworkDelegate when network service is enabled. Bug: 598073 Chromium change in C74 to enable network service by default: commit 708db9bc46bec1d0037f44f8fdaff7b7d72724e0 Author: John Abd-El-Malek <jam@chromium.org> Date: Mon Feb 25 18:10:35 2019 +0000 Enable network service on Win/Mac/Linux by default as it launched on stable. Removing running integration tests on these bots (waterfall + CQ). Keep running the layout tests with network service disabled on on debug Linux Tests bots as we're still supporting ChromeOS/Android/Cast which haven't switched over yet. Also remove the disabled-service-worker-servicification virtual test. Bug: 933880,926114
mkarolin
changed the title
mkarolin
commented
Mar 20, 2019
| @@ -0,0 +1,12 @@ | |||
| diff --git a/chrome/browser/resources/history/history_toolbar.html b/chrome/browser/resources/history/history_toolbar.html | |||
There was a problem hiding this comment.
Moved from patches/chrome-browser-resources-md_history-history_toolbar.html.patch (deleted below).
bbondy
reviewed
Mar 20, 2019
bbondy
reviewed
Mar 20, 2019
| @@ -10,7 +10,7 @@ | |||
| namespace net { | |||
|
|
|||
| int SOCKS5ClientSocket::DoAuth(int rv) { | |||
| rv = Authenticate(rv, *transport_, net_log_, io_callback_); | |||
There was a problem hiding this comment.
cc @yrliou pls review this commit: "Updated SOCKS5 client socket overrides."
bbondy
reviewed
Mar 20, 2019
| @@ -22,10 +22,12 @@ | |||
| #include "components/prefs/pref_registry_simple.h" | |||
There was a problem hiding this comment.
cc @jumde please review this commit: "Change to classes inheriting from update_client::Configurator."
bbondy
reviewed
Mar 20, 2019
| @@ -145,6 +146,7 @@ bool BraveMainDelegate::BasicStartupComplete(int* exit_code) { | |||
| autofill::features::kAutofillServerCommunication.name, | |||
| features::kAudioServiceOutOfProcess.name, | |||
| features::kDefaultEnableOopRasterization.name, | |||
| network::features::kNetworkService.name, | |||
There was a problem hiding this comment.
I posted an issue here to revert this and support the network service:
brave/brave-browser#3819
|
I edited Comment 0 to add a section "Special notes to QA:" with some things I wanted them to pay attention to based on the code changes. |
bbondy
approved these changes
Mar 20, 2019
|
Just noting that this will need an uplift to 0.63.x as well after it is merged to master. We should wait for Nightly builds for a couple days before merging in the Dev uplift though. |
pilgrim-brave
approved these changes
Mar 21, 2019
mkarolin
added a commit
that referenced
this pull request
Mar 21, 2019
Upgrade Chromium from 73.0.3683.75 to Chromium 74.0.3729.22
16 tasks
19 tasks
bsclifton
added a commit
that referenced
this pull request
Mar 27, 2019
(originally implemented with #2013) Fixes brave/brave-browser#3898
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#3502
Related PR: brave/brave-browser#3790
It's easiest to review this per commit.
Each commit explains where the change came from in Chromium's history.
Special notes to QA:
Submitter Checklist:
npm test brave_unit_tests && npm test brave_browser_tests) onnpm run lint)git rebase master(if needed).git rebase -ito squash commits (if needed).Test Plan:
See brave/brave-browser#3502.
Reviewer Checklist: