GitLab

This CL mojofies the legacy IPC
ServiceWorkerHostMsg_PostMessageToWorker.

BUG=772713,813749

Change-Id: I7e6ad8d411c820d22fe7b91b0bd2036b48aabe69
Reviewed-on: https://chromium-review.googlesource.com/933867Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Makoto Shimazu <shimazu@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Commit-Queue: Han Leon <leon.han@intel.com>
Cr-Commit-Position: refs/heads/master@{#541987}

Bug: 820070
Change-Id: I61115868be9edf2bb6be1214e7eb4283faf1cf7b
Reviewed-on: https://chromium-review.googlesource.com/956308
Commit-Queue: Łukasz Anforowicz <lukasza@chromium.org>
Reviewed-by: Nick Carter <nick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541986}

This patch makes DataTransferAccessPolicy to be an enum class for better
type safety and improving code health. Using forward declaratuion in
other header files instead of including |DataTransferAccessPolicy.h|.
Signed-off-by: Zhuoyu Qian <zhuoyu.qian@samsung.com>
Change-Id: Id516b37b0d71ba348b9f146f5ce546cf28316cf7
Reviewed-on: https://chromium-review.googlesource.com/954777Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541985}

if menu animation started.


Bug: 799524
Change-Id: I1d87a7944a8f98f9617ccb071ffae1adef491605
Reviewed-on: https://chromium-review.googlesource.com/953823
Commit-Queue: min c <minch@chromium.org>
Reviewed-by: Qiang Xu <warx@google.com>
Reviewed-by: Dan Erat <derat@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541984}

R=dpranke@chromium.org

Change-Id: I25bf5c2b3c64649a7cfa49481a4a9e3a0d20a58f
Reviewed-on: https://chromium-review.googlesource.com/941205
Commit-Queue: Yngve Pettersen <yngve@vivaldi.com>
Reviewed-by: Tomasz Mikolajewski <mtomasz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541983}

Remove include of <vector> and <string> within Blink when it isn't
needed.

Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
Change-Id: I8d3cf66ecc95c02865153dbd4af475b400804476
Bug: 820271
Reviewed-on: https://chromium-review.googlesource.com/954764Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Noel Gordon <noel@chromium.org>
Commit-Queue: Luciano Pacheco (SYD) <lucmult@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541982}

In SPv1 mode, LayoutObject::AdjustStyleDifference will cause subtree paint invalidation
(including PaintLayer repaint) on change of paint properties or z-index, but only
if there is no direct compositing reason for the object. The reason for the LayoutObject
paint invalidations was to cause raster invalidation as a side-effect. Painted
display list output of individual LayoutObjects does not change as a result of these
updates.

In SPv175 mode, we don't need to invalidate paint of individual LayoutObjects just
becaue of these changes, since raster invalidation happens via a new PaintChunk
mechanism. However, we still need to cause PaintLayer repaint if any paint property
tree nodes are added or removed, or z-index changes, to cause re-drawing of PaintChunks
with the correct paint properties and order.

There is already invalidation of PaintLayer repaint for paint property node change,
in PaintPropertyTreeBuilder::UpdateForSelf and ::UpdateForChildren (it was added
recently, after the code being removed in this CL was committed). Therefore we
only need to invalidate repaint for z-index change. Further, this is only needed
if the LayoutObject already had a PaintLayer, because z-index has no effect on
objects without a PaintLayer, and there is already invalidation happening on
PaintLayer addition or removal.

Tested by:

LayoutTests/paint/invalidation/compositing/chunk-reorder.html
svg/transforms/change-transform-to-none-shape.html

Bug:814711

Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
Change-Id: I208b03b9e6b1a6635dab9585b8884ed0e2f8688d
Reviewed-on: https://chromium-review.googlesource.com/954352
Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541981}

Flakily fails with:
../../chrome/browser/chromeos/system/tray_accessibility_browsertest.cc:1681: Failure
Value of: IsHelpAvailableOnDetailMenu()
  Actual: false
Expected: true

NOTRY=true
TBR=xiyuan@chromium.org

Bug: 819987, 787024
Change-Id: I133007c1411f714f2b7646c0c0d949c577c63037
Reviewed-on: https://chromium-review.googlesource.com/956276
Commit-Queue: James Cook <jamescook@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541980}

Convert all messages with prefix IDS_FILE_BROWSER_METADATA.

Also fix 2 types: "for" and "screenshot" that I noticed.

Tested in a local chromeOS build.

Bug: 769395
Change-Id: Ie286f00f2ec54b3227ded5fedcd944e55ade1685
Reviewed-on: https://chromium-review.googlesource.com/952108Reviewed-by: Sasha Morrissey <sashab@chromium.org>
Commit-Queue: Luciano Pacheco (SYD) <lucmult@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541979}

These trybots will go into the general luci.chromium.try bucket. Right
now only the Linux and Mac ones have been set up, but it's harmless to
register all of them now. (The optional GPU tryservers are already in
this console and conversion has only just started for them.)

Also added a separate angle.try console with just these LUCI trybots.

Bug: 815092
Change-Id: I27d04c7a2b7ef5cf773eb96427e09e8552647640
Reviewed-on: https://chromium-review.googlesource.com/955560
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Nodir Turakulov <nodir@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541978}

mTemporarySelectedStack is a variable that temporarily stores the selected tab
stack. Its value is used to update TabModelSelector's currently selected model
once the scroll animation finishes.

This CL changes mTemporarySelectedStack from a Boolean to an int so StackLayout
can support more than two tab stacks.

Bug: 648314
Change-Id: I337feb3b82f687ef554c577544d1087c8d0e9c40
Reviewed-on: https://chromium-review.googlesource.com/954285
Commit-Queue: Ryan Landay <rlanday@chromium.org>
Reviewed-by: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541977}

Bug: 771466, 776260
Test: Confirm that floating VK is enabled after setting the flag to 'Default'.
Change-Id: I2e4a26ee05dbaba16cd78f0768e6fae3d079bcc4
Reviewed-on: https://chromium-review.googlesource.com/954804Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Commit-Queue: Yuichiro Hanada <yhanada@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541976}

1. Target.IsNewInstance() was returning to true when host name is
specified explicitly. This is incorrect because the host is expected to
be provisioned and it may be reused for multiple test runs.
2. Updated deployment code to read /blob instead of /blobstore to get
list of existing blobs.

Bug: 798851
Change-Id: I62b04a352a364250e2aa2d44319494481be011c6
Reviewed-on: https://chromium-review.googlesource.com/956349
Commit-Queue: Sergey Ulanov <sergeyu@chromium.org>
Reviewed-by: Wez <wez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541975}

Android doesn't need to pull in in-browser discovery or cast channel
code. This patch moves those definitions in test_helper.{cc,h} to
existing #if !defined(OS_ANDROID) guards.

Bug: 820238
Change-Id: Id47f7fa8421daa94f2c6a804240f6541391c3998
Reviewed-on: https://chromium-review.googlesource.com/956353Reviewed-by: Bin Zhao <zhaobin@chromium.org>
Commit-Queue: Derek Cheng <imcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541974}

WebSiteDataStore crashed during the destructing of WKHTTPCookieStore
The reason:
 WKHTTPCookieStore can't be deleted in IO Thread (undocumented behavior).
The solution:
 Make WKHTTPCookieStore instance in WKHTTPSystemCookieStore as weak, so
 it's not retained in WKHTTPSystemCookieStore and it's deleted/released
 when WebSiteDataStore is deleted.

Bug: 817460, 816879
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I15b384f41496df6d0af9c550b7b13e67a75d236b
Reviewed-on: https://chromium-review.googlesource.com/955547
Commit-Queue: Mohammad Refaat <mrefaat@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541973}

Bug: 790536, 790521
Change-Id: Id677b3b0e50ce9f6b342e46269fbaa9ab9575ef9
Reviewed-on: https://chromium-review.googlesource.com/956664Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Shenghua Zhang <shenghuazhang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541972}

Make GetIdealBoundsOfItemIcon handle overflow correctly.
Enable RootWindowControllerTest.MoveWindows_Basic in Mash.
(mash adds overflow shelf items for the widgets in that test)

Bug: 814015
Test: Automated; windows minimize to shelf item or overflow button.
Change-Id: I3529f898c8201fe7fa0e2f34ecfe008c2a029272
Reviewed-on: https://chromium-review.googlesource.com/956096
Commit-Queue: Michael Wasserman <msw@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541971}

Treat the fuzzer-input as the aggregated hit-test data, and run a bunch of
queries on that.

BUG=819291

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: Iab5b09820c2751126fd6a9f630848834b47d2dd0
Reviewed-on: https://chromium-review.googlesource.com/949407Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Reviewed-by: Ken Rockot <rockot@chromium.org>
Reviewed-by: Ria Jiang <riajiang@chromium.org>
Commit-Queue: Sadrul Chowdhury <sadrul@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541970}

Bug: 790536, 790521
Change-Id: I0479b74551bbd38841721ea549b198c6dfc2cdac
Reviewed-on: https://chromium-review.googlesource.com/956551Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Shenghua Zhang <shenghuazhang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541969}

(NOPRESUBMIT because the owners check doesn't seem to properly deal
with owners reassigning their ownership!)

NOPRESUBMIT=true

Bug: none
Change-Id: I9b8cf3944a25e478105e6a241094d718666ee7a0
Reviewed-on: https://chromium-review.googlesource.com/956676
Commit-Queue: Elliot Glaysher <erg@chromium.org>
Reviewed-by: Thomas Anderson <thomasanderson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541968}

This class is in charge of creating CdmProxy for Widevine CDM.

BUG=787650

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I0b6b1fa4621b41b727f1200725dedca647939628
Reviewed-on: https://chromium-review.googlesource.com/933369Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Xiaohan Wang <xhwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541967}

The legacy implementation of LayoutReplaced::PositionForPoint()
almost works in LayoutNG, except for the case where the replaced
object is in inline layout, where it fails to obtain the line's
logical top and bottom in the containing block.

This patch adds LayoutNG support for it.

Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_layout_tests_layout_ng;master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
Change-Id: I414a445e0617770aec6461dbc497f650f586d12b
Reviewed-on: https://chromium-review.googlesource.com/949864
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Reviewed-by: Yoshifumi Inoue <yosin@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541966}

ToT V4L2 decoder enumerates all VP9 Profiles as supported but looking at
the manuals and the code this is not true (see bug). This CL restricts
the enumeration to VP9 Profile 0.

Bug: 796297
Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I4e604830790efe3b703af501f7f401bd96375f41
Reviewed-on: https://chromium-review.googlesource.com/953682Reviewed-by: Pawel Osciak <posciak@chromium.org>
Commit-Queue: Miguel Casas <mcasas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541965}

This reverts commit 23e967ba.

Reason for revert: Chrome kept crashing when trying to log in on the login screen. See crbug.com/820302 for details.

Bug: 820302

Original change's description:
> Migrate ServicesCustomizationDocument to SimpleURLLoader
> 
> Bug: 773295
> Change-Id: Id129cea8341c14bd8da4b5374034423223f03859
> Reviewed-on: https://chromium-review.googlesource.com/884063
> Commit-Queue: Mark Pilgrim <pilgrim@chromium.org>
> Reviewed-by: James Cook <jamescook@chromium.org>
> Reviewed-by: Matt Menke <mmenke@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#541884}

TBR=jamescook@chromium.org,mmenke@chromium.org,pilgrim@chromium.org

Change-Id: Ibaf3b0e0f98edf34f79374e85101b705418e7588
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 773295
Reviewed-on: https://chromium-review.googlesource.com/956763Reviewed-by: Xiaoqian Dai <xdai@chromium.org>
Commit-Queue: Xiaoqian Dai <xdai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541964}

These were generated using Shaka Packager 2.0.0.

BUG=657957
TEST=new browser_tests pass

Change-Id: I1ea784700c0d96b549f2a40221b92dd9af807d4a
Reviewed-on: https://chromium-review.googlesource.com/935193
Commit-Queue: John Rummell <jrummell@chromium.org>
Reviewed-by: Xiaohan Wang <xhwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541963}

Bug: 733662
Change-Id: Ia5d8c360225e817529f5145bd7a0e22ff2206dae
Reviewed-on: https://chromium-review.googlesource.com/956027Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Jiaquan He <hejq@google.com>
Cr-Commit-Position: refs/heads/master@{#541962}

Bug: 790536, 790521
Change-Id: Ib2761ce671a42ab6a3b2a6cdfad94d8725a6c139
Reviewed-on: https://chromium-review.googlesource.com/956495Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Shenghua Zhang <shenghuazhang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541961}

The histograms were added in https://crrev.googlesource.com/c/782642
but the histograms.xml and enums.xml didn't match the actual histogram
and enum names. This CL fixes that.

Bug: 682933
Change-Id: Id0447f7e16032dae16d45781de07bacbd96a2ad6
Reviewed-on: https://chromium-review.googlesource.com/956343Reviewed-by: Ilya Sherman <isherman@chromium.org>
Commit-Queue: Ilya Sherman <isherman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541960}

Change-Id: I32ab5f24831f7a18e5e3401503b242692f42c48b
Reviewed-on: https://chromium-review.googlesource.com/956541Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Devin Mullins <twifkak@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541959}

This CL allows users to use the Feedback keyboard shortcut to launch the
Feedback UI and send Feedback. This also fixes an issue with the current
implementation of login feedback, where if we try to send feedback after
we have launched the app before and restarted Chrome, the UI would not
launch.

This is because on the first run, the background host is always created
just to get the types of events the app can handle. Once this is done,
we save this information in the extension registry and only load the
app once the event is actually dispatched. Since the existing code
was instead waiting for the host to load, it would never send the
event and wait forever.

Instead we're not checking if the extension has been added to the
registry. If it has, we can safely dispatch the event. If not, we instead
wait for the extension to load and then send feedback.

I've tested that this works now both for the first run and subsequent runs
of the app. The existing browser test should still be applicable since the
CL just changed the entry point for login feedback; did not add any
additional functionality.

R=rdevlin.cronin@chromium.org, xiyuan@chromium.org

Bug: 809715
Change-Id: Ib26cb879b15334c010bfed539bc83384c09d2a01
Reviewed-on: https://chromium-review.googlesource.com/953372
Commit-Queue: Rahul Chaturvedi <rkc@chromium.org>
Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541958}

This CL uses ImageView instances to show the icon and answer image
instead of drawing them in OnPaint. The OnPaint has been removed from
OmniboxResultView.

Bug: None
Change-Id: Ic9b1f98d5f81a4dd8b23a39b5174996270314086
Reviewed-on: https://chromium-review.googlesource.com/952669
Commit-Queue: Dave Schuyler <dschuyler@chromium.org>
Reviewed-by: Justin Donnelly <jdonnelly@chromium.org>
Reviewed-by: Tommy Li <tommycli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541957}

Change-Id: Id7dafc9572f34b3daa2122a11efd2b7e82bd2c91
Reviewed-on: https://chromium-review.googlesource.com/956544Reviewed-by: Ryan Hamilton <rch@chromium.org>
Commit-Queue: Nick Harper <nharper@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541956}

Bug: 706389
Change-Id: Ida63780d8010deca305697c93cff34bfa3245443
Reviewed-on: https://chromium-review.googlesource.com/956483Reviewed-by: Philip Jägenstedt <foolip@chromium.org>
Reviewed-by: Bryan McQuade <bmcquade@chromium.org>
Commit-Queue: Luna Lu <loonybear@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541955}

When BaseAudioContext is being tear down, AudioWorkletGlobalScope might
be already gone at that point. When BAC accesses AWGS, we must check its
existence.

Currently, the only access point of |global_scope| in BAC is
UpdateWorkletGlobalScopeOnRenderingThread().

Bug: 819845
Test: The attached repro case does not crash after 20 min run.
Change-Id: I485668708406639b8b77ae348c57c31782db5438
Reviewed-on: https://chromium-review.googlesource.com/954228Reviewed-by: Raymond Toy <rtoy@chromium.org>
Reviewed-by: Hongchan Choi <hongchan@chromium.org>
Commit-Queue: Hongchan Choi <hongchan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541954}

This reverts commit 613ba331.

Reason for revert: Causing build failures:
https://logs.chromium.org/v/?s=chromium%2Fbb%2Ftryserver.chromium.mac%2Fmac_chromium_rel_ng%2F668246%2F%2B%2Frecipes%2Fsteps%2Fanalyze%2F0%2Fstdout

/b/c/b/mac/src/buildtools/mac/gn gen //out/Release --check
  -> returned 1
ERROR at //chrome/test/BUILD.gn:4405:7: Can't load input file.
      "//extensions/features",
      ^----------------------

Original change's description:
> Extensions: Fix memory leak in ExtensionApiFrameIdMap.
> 
> Currently in ExtensionApiFrameIdMap::UpdateTabAndWindowId and
> ExtensionApiFrameIdMap::LookupFrameDataOnUI we don't check whether a
> RenderFrameHost is alive, before storing it's frame data. It's possible that
> such a RenderFrameHost never becomes alive, and hence we don't get any
> OnRenderFrameDeleted notification for it. Since ExtensionApiFrameIdMap is a
> singleton, this causes a memory leak. To fix, ensure that only live
> RenderFrameHosts are tracked.
> 
> Also ensure in InProcessBrowserTest::TearDown that ExtensionApiFrameIdMap is not
> leaking memory for any browser test.
> 
> BUG=817205
> 
> Change-Id: I78bf06cee7e10465f9eb5973ee8e59f525a74d93
> Reviewed-on: https://chromium-review.googlesource.com/951975
> Reviewed-by: Scott Violet <sky@chromium.org>
> Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
> Commit-Queue: Karan Bhatia <karandeepb@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#541950}

TBR=sky@chromium.org,rdevlin.cronin@chromium.org,karandeepb@chromium.org

Change-Id: Ib200225c2824c865feda8707f081485d3f53d912
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 817205
Reviewed-on: https://chromium-review.googlesource.com/956610Reviewed-by: Ryan Landay <rlanday@chromium.org>
Commit-Queue: Ryan Landay <rlanday@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541953}

Bug: 813716
Cq-Include-Trybots: master.tryserver.chromium.linux:closure_compilation
Change-Id: I44ca92b701d1fc1140bdbfb577fb9cb0a5680219
Reviewed-on: https://chromium-review.googlesource.com/953972
Commit-Queue: Scott Chen <scottchen@chromium.org>
Reviewed-by: Dave Schuyler <dschuyler@chromium.org>
Reviewed-by: Demetrios Papadopoulos <dpapad@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541952}

These need a different service account than some of the other trybots
for some of the tests they check out and run.

Bug: 815092
Change-Id: Ieb4eb45971c5b057642722eb615f586e8e212d6b
Reviewed-on: https://chromium-review.googlesource.com/955681Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Nodir Turakulov <nodir@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541951}

Currently in ExtensionApiFrameIdMap::UpdateTabAndWindowId and
ExtensionApiFrameIdMap::LookupFrameDataOnUI we don't check whether a
RenderFrameHost is alive, before storing it's frame data. It's possible that
such a RenderFrameHost never becomes alive, and hence we don't get any
OnRenderFrameDeleted notification for it. Since ExtensionApiFrameIdMap is a
singleton, this causes a memory leak. To fix, ensure that only live
RenderFrameHosts are tracked.

Also ensure in InProcessBrowserTest::TearDown that ExtensionApiFrameIdMap is not
leaking memory for any browser test.

BUG=817205

Change-Id: I78bf06cee7e10465f9eb5973ee8e59f525a74d93
Reviewed-on: https://chromium-review.googlesource.com/951975Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Commit-Queue: Karan Bhatia <karandeepb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541950}

Refactors the watcher APIs a bit to support some exensibility. Also
renames them in the process because "trap" seems like a better analogy.

Some remnants of the watcher concept are left intact to keep the CL size
small.

TBR=dcheng@chromium.org

Bug: 819046
Change-Id: Ie469f715c32080824d99b3a4f85834c030cbd181
Reviewed-on: https://chromium-review.googlesource.com/952081Reviewed-by: Ken Rockot <rockot@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Jay Civelli <jcivelli@chromium.org>
Commit-Queue: Ken Rockot <rockot@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541949}

DownloadManager::InputStream is backed by either a ByteStreamReader or a
DataStreamHandle. The former is implemented in content/. And it is used
by the legacy download code. For network service, the latter is used.
To move download code to components/, this CL makes them 2 separate
child class of a common download::InputStream.
And that makes code more simple to read as there used to be a lot of
if-else statement to check which one is used in the InputStream.

Bug: 803135
Change-Id: Icab5369583b186bf253aec84f1d59769162300c9
Reviewed-on: https://chromium-review.googlesource.com/952023
Commit-Queue: Min Qin <qinmin@chromium.org>
Reviewed-by: Joy Ming <jming@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#541948}