Cleanup Tool WebUI: Add functionality to the Cleanup button

Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2819633003
Cr-Commit-Position: refs/heads/master@{#467408}
Committed: https://chromium.googlesource.com/chromium/src/+/b573ab6f6f5f1b9a5ce1b56595590eae69563638

[proberge, 2017-04-13 21:00:31]

Description was changed from

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
==========

to

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation
==========

[proberge, 2017-04-13 21:09:37]

proberge@chromium.org changed reviewers:
+ alito@chromium.org

[alito, 2017-04-14 00:01:19]

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:70: this.isRunning= false;
nit: space before '='

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:72: this.isInfected = false;
Could you document why isInfected is set to false here?

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:123: * @return {boolean} True
if the "Scanning" label should be displayed.
nit: True if -> Whether

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:133: * @return {boolean} True
if the "Cleaning" label should be displayed.
nit: True if -> Whether

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.cc (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.cc:1244: DCHECK(browser);
No need for this DCHECK.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:14: #include "base//callback.h"
Remove the extra '/'

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:153: void
DisplayUserInitiatedSRTPromptDialog(base::Closure callback);
In order for the organic page to be able to run the cleaner, it will need to
call some modified version of the RunSwReporters() function (The cleaner needs
to run, the IPC will have to be set up, and the info passed from the cleaner to
Chrome needs to eventually make its way to the SRTPromptController). It's still
more complicated because we need to ensure that there is one place where all
requests for running the cleaner comes in so we don't run it multiple times
resulting in multiple prompts.

How about holding off on displaying the prompt from the organic page for now? Or
if you want to just prototype the dialog being displayed, move this function to
cleanup_action_handler.cc until we are ready to hook everything up?

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
File chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.cc (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.cc:48: CHECK_EQ(1U,
args->GetSize());
Move this to the previous line?

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
File chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.h:10: #include
"chrome/browser/ui/browser.h"
Please remove this include.

[proberge, 2017-04-18 19:24:29]

Thanks!

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:70: this.isRunning= false;
On 2017/04/14 00:01:18, alito wrote:
> nit: space before '='

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:72: this.isInfected = false;
On 2017/04/14 00:01:18, alito wrote:
> Could you document why isInfected is set to false here?

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:123: * @return {boolean} True
if the "Scanning" label should be displayed.
On 2017/04/14 00:01:18, alito wrote:
> nit: True if -> Whether

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/resource...
chrome/browser/resources/cleanup_tool/manager.js:133: * @return {boolean} True
if the "Cleaning" label should be displayed.
On 2017/04/14 00:01:18, alito wrote:
> nit: True if -> Whether

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.cc (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.cc:1244: DCHECK(browser);
On 2017/04/14 00:01:18, alito wrote:
> No need for this DCHECK.

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:14: #include "base//callback.h"
On 2017/04/14 00:01:18, alito wrote:
> Remove the extra '/'

Wow, no idea what happened there.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:153: void
DisplayUserInitiatedSRTPromptDialog(base::Closure callback);
On 2017/04/14 00:01:18, alito wrote:
> In order for the organic page to be able to run the cleaner, it will need to
> call some modified version of the RunSwReporters() function (The cleaner needs
> to run, the IPC will have to be set up, and the info passed from the cleaner
to
> Chrome needs to eventually make its way to the SRTPromptController). It's
still
> more complicated because we need to ensure that there is one place where all
> requests for running the cleaner comes in so we don't run it multiple times
> resulting in multiple prompts.
> 
> How about holding off on displaying the prompt from the organic page for now?
Or
> if you want to just prototype the dialog being displayed, move this function
to
> cleanup_action_handler.cc until we are ready to hook everything up?

Scan results will need to be available before "Run Chrome Cleanup" is displayed.
I don't think we want to run the scan again; we should have all the information
needed to show the prompt. 

The user can initiate a scan by clicking on the "Scan Now" button on this page.
See https://codereview.chromium.org/2822433003/.

I would prefer to keep logic out of ui/webui.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
File chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.cc (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.cc:48: CHECK_EQ(1U,
args->GetSize());
On 2017/04/14 00:01:18, alito wrote:
> Move this to the previous line?

Done.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
File chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/ui/webui...
chrome/browser/ui/webui/cleanup_tool/cleanup_action_handler.h:10: #include
"chrome/browser/ui/browser.h"
On 2017/04/14 00:01:19, alito wrote:
> Please remove this include.

Done.

[alito, 2017-04-19 18:01:44]

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:153: void
DisplayUserInitiatedSRTPromptDialog(base::Closure callback);
On 2017/04/18 19:24:28, proberge wrote:
> On 2017/04/14 00:01:18, alito wrote:
> > In order for the organic page to be able to run the cleaner, it will need to
> > call some modified version of the RunSwReporters() function (The cleaner
needs
> > to run, the IPC will have to be set up, and the info passed from the cleaner
> to
> > Chrome needs to eventually make its way to the SRTPromptController). It's
> still
> > more complicated because we need to ensure that there is one place where all
> > requests for running the cleaner comes in so we don't run it multiple times
> > resulting in multiple prompts.
> > 
> > How about holding off on displaying the prompt from the organic page for
now?
> Or
> > if you want to just prototype the dialog being displayed, move this function
> to
> > cleanup_action_handler.cc until we are ready to hook everything up?
> 
> Scan results will need to be available before "Run Chrome Cleanup" is
displayed.
> I don't think we want to run the scan again; we should have all the
information
> needed to show the prompt. 
> 
> The user can initiate a scan by clicking on the "Scan Now" button on this
page.
> See https://codereview.chromium.org/2822433003/.
> 
> I would prefer to keep logic out of ui/webui. 

As we discussed offline, the logic could be simplified so that whenever any
"scan" or "clean" button is clicked on the webui, the full cleaner flow would be
run. That way we wouldn't need to cache the results of scans and we would ensure
that the information displayed in the prompt is accurate wrt files that will be
removed.

[proberge, 2017-04-19 18:03:10]

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:153: void
DisplayUserInitiatedSRTPromptDialog(base::Closure callback);
On 2017/04/19 18:01:44, alito wrote:
> On 2017/04/18 19:24:28, proberge wrote:
> > On 2017/04/14 00:01:18, alito wrote:
> > > In order for the organic page to be able to run the cleaner, it will need
to
> > > call some modified version of the RunSwReporters() function (The cleaner
> needs
> > > to run, the IPC will have to be set up, and the info passed from the
cleaner
> > to
> > > Chrome needs to eventually make its way to the SRTPromptController). It's
> > still
> > > more complicated because we need to ensure that there is one place where
all
> > > requests for running the cleaner comes in so we don't run it multiple
times
> > > resulting in multiple prompts.
> > > 
> > > How about holding off on displaying the prompt from the organic page for
> now?
> > Or
> > > if you want to just prototype the dialog being displayed, move this
function
> > to
> > > cleanup_action_handler.cc until we are ready to hook everything up?
> > 
> > Scan results will need to be available before "Run Chrome Cleanup" is
> displayed.
> > I don't think we want to run the scan again; we should have all the
> information
> > needed to show the prompt. 
> > 
> > The user can initiate a scan by clicking on the "Scan Now" button on this
> page.
> > See https://codereview.chromium.org/2822433003/.
> > 
> > I would prefer to keep logic out of ui/webui. 
> 
> As we discussed offline, the logic could be simplified so that whenever any
> "scan" or "clean" button is clicked on the webui, the full cleaner flow would
be
> run. That way we wouldn't need to cache the results of scans and we would
ensure
> that the information displayed in the prompt is accurate wrt files that will
be
> removed.

Ack, closing this CL.

[proberge, 2017-04-25 17:50:12]

Description was changed from

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation
==========

to

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation
==========

[proberge, 2017-04-25 17:51:10]

After more offline discussions, PTAL.

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
File chrome/browser/safe_browsing/srt_fetcher_win.h (right):

https://codereview.chromium.org/2819633003/diff/20001/chrome/browser/safe_bro...
chrome/browser/safe_browsing/srt_fetcher_win.h:153: void
DisplayUserInitiatedSRTPromptDialog(base::Closure callback);
On 2017/04/19 18:03:10, proberge wrote:
> On 2017/04/19 18:01:44, alito wrote:
> > On 2017/04/18 19:24:28, proberge wrote:
> > > On 2017/04/14 00:01:18, alito wrote:
> > > > In order for the organic page to be able to run the cleaner, it will
need
> to
> > > > call some modified version of the RunSwReporters() function (The cleaner
> > needs
> > > > to run, the IPC will have to be set up, and the info passed from the
> cleaner
> > > to
> > > > Chrome needs to eventually make its way to the SRTPromptController).
It's
> > > still
> > > > more complicated because we need to ensure that there is one place where
> all
> > > > requests for running the cleaner comes in so we don't run it multiple
> times
> > > > resulting in multiple prompts.
> > > > 
> > > > How about holding off on displaying the prompt from the organic page for
> > now?
> > > Or
> > > > if you want to just prototype the dialog being displayed, move this
> function
> > > to
> > > > cleanup_action_handler.cc until we are ready to hook everything up?
> > > 
> > > Scan results will need to be available before "Run Chrome Cleanup" is
> > displayed.
> > > I don't think we want to run the scan again; we should have all the
> > information
> > > needed to show the prompt. 
> > > 
> > > The user can initiate a scan by clicking on the "Scan Now" button on this
> > page.
> > > See https://codereview.chromium.org/2822433003/.
> > > 
> > > I would prefer to keep logic out of ui/webui. 
> > 
> > As we discussed offline, the logic could be simplified so that whenever any
> > "scan" or "clean" button is clicked on the webui, the full cleaner flow
would
> be
> > run. That way we wouldn't need to cache the results of scans and we would
> ensure
> > that the information displayed in the prompt is accurate wrt files that will
> be
> > removed.
> 
> Ack, closing this CL. 

Re-opening since it seems that we will indeed have "Start Cleaning" logic on
this WebUI.

[alito, 2017-04-25 18:53:07]

lgtm

[proberge, 2017-04-26 14:46:08]

proberge@chromium.org changed reviewers:
+ sorin@chromium.org, tommycli@chromium.org

[proberge, 2017-04-26 14:46:09]

@alito I made some changes, could you take another look?
@sorin I reverted some changes from yesterday's CLs due to recent design
changes. Could you review for browser/component_updater? Sorry for the noise.
@tommycli Once again for WebUI changes.

[alito, 2017-04-26 15:42:45]

still lgtm

[tommycli, 2017-04-26 16:10:34]

lgtm after these changes.

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/cleanup_browser_proxy.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/cleanup_browser_proxy.js:25: * ]}
This typedef seems odd. Did you mean {{ ... }} ? I've never seen {[ ... ]}.

Does this pass closure compile?

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/manager.js:116: shouldShowScan_:
function(hasScanResults, isRunning) {
Instead of all these shouldShowFoo_ methods that cover all 4 possibilities (2x2
booleans = 4?)

I recommend just making a getStatusLabel_ method that just returns a string.

You can use this.i18n method if you have the i18n behavior. See this example:
https://cs.chromium.org/chromium/src/chrome/browser/resources/settings/contro...

[Sorin Jianu, 2017-04-26 16:52:47]

lgtm

thank you!

component updater lgtm

[proberge, 2017-04-26 17:28:38]

Thanks for the quick review everyone :)

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/cleanup_browser_proxy.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/cleanup_browser_proxy.js:25: * ]}
On 2017/04/26 16:10:33, tommycli wrote:
> This typedef seems odd. Did you mean {{ ... }} ? I've never seen {[ ... ]}.
> 
> Does this pass closure compile?

Good catch! I indeed meant {{...}}.
I usually rely on the CQ for closure compile, as I'm (sadly) developing on
Windows.

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/manager.js:116: shouldShowScan_:
function(hasScanResults, isRunning) {
On 2017/04/26 16:10:33, tommycli wrote:
> Instead of all these shouldShowFoo_ methods that cover all 4 possibilities
(2x2
> booleans = 4?)
> 
> I recommend just making a getStatusLabel_ method that just returns a string.
> 
> You can use this.i18n method if you have the i18n behavior. See this example:
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/settings/contro...

Two of the methods are used by buttons, two are used by labels. I don't think we
should use a single element with updating text.

I do agree that having 4 methods to show/hide elements is rather clunky. What do
you think of something like
hidden="[[!isState_(toolState_, 'is-scanning')]]" ?
(From 
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)

Alternatively:
hidden="[[notEq_(toolState_, cleanup.SCANNING)]]
(From 
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)

[tommycli, 2017-04-26 18:01:34]

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/manager.js:116: shouldShowScan_:
function(hasScanResults, isRunning) {
On 2017/04/26 17:28:38, proberge wrote:
> On 2017/04/26 16:10:33, tommycli wrote:
> > Instead of all these shouldShowFoo_ methods that cover all 4 possibilities
> (2x2
> > booleans = 4?)
> > 
> > I recommend just making a getStatusLabel_ method that just returns a string.
> > 
> > You can use this.i18n method if you have the i18n behavior. See this
example:
> >
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/settings/contro...
> 
> Two of the methods are used by buttons, two are used by labels. I don't think
we
> should use a single element with updating text.
> 
> I do agree that having 4 methods to show/hide elements is rather clunky. What
do
> you think of something like
> hidden="[[!isState_(toolState_, 'is-scanning')]]" ?
> (From 
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)
> 
> Alternatively:
> hidden="[[notEq_(toolState_, cleanup.SCANNING)]]
> (From 
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)

Ah, I see, I didn't realize they were different elements. In that case, what you
have currently is probably the best. Both isState_ and notEq_ seem a bit too
clever for my personal taste, but that's just a matter of taste...

[proberge, 2017-04-26 18:06:20]

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
File chrome/browser/resources/cleanup_tool/manager.js (right):

https://codereview.chromium.org/2819633003/diff/140001/chrome/browser/resourc...
chrome/browser/resources/cleanup_tool/manager.js:116: shouldShowScan_:
function(hasScanResults, isRunning) {
On 2017/04/26 18:01:34, tommycli wrote:
> On 2017/04/26 17:28:38, proberge wrote:
> > On 2017/04/26 16:10:33, tommycli wrote:
> > > Instead of all these shouldShowFoo_ methods that cover all 4 possibilities
> > (2x2
> > > booleans = 4?)
> > > 
> > > I recommend just making a getStatusLabel_ method that just returns a
string.
> > > 
> > > You can use this.i18n method if you have the i18n behavior. See this
> example:
> > >
> >
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/settings/contro...
> > 
> > Two of the methods are used by buttons, two are used by labels. I don't
think
> we
> > should use a single element with updating text.
> > 
> > I do agree that having 4 methods to show/hide elements is rather clunky.
What
> do
> > you think of something like
> > hidden="[[!isState_(toolState_, 'is-scanning')]]" ?
> > (From 
> >
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)
> > 
> > Alternatively:
> > hidden="[[notEq_(toolState_, cleanup.SCANNING)]]
> > (From 
> >
>
https://cs.chromium.org/chromium/src/chrome/browser/resources/chromeos/login/...)
> 
> Ah, I see, I didn't realize they were different elements. In that case, what
you
> have currently is probably the best. Both isState_ and notEq_ seem a bit too
> clever for my personal taste, but that's just a matter of taste...

Glad we agree :)
Will stay with the 4 methods for now and keep isState_ in the back of my mind if
we eventually need to add a third boolean.

[proberge, 2017-04-26 18:06:33]

The CQ bit was checked by proberge@chromium.org

[proberge, 2017-04-26 18:06:33]

The patchset sent to the CQ was uploaded after l-g-t-m from sorin@chromium.org,
alito@chromium.org, tommycli@chromium.org
Link to the patchset: https://codereview.chromium.org/2819633003/#ps160001
(title: "Fix typdef {[ -> {{")

[commit-bot: I haz the power, 2017-04-26 18:07:16]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-04-26 19:21:06]

CQ is committing da patch.

Bot data: {"patchset_id": 160001, "attempt_start_ts": 1493229993383320,
"parent_rev": "aa6248548d241ea5af0bc8a6aaa2f28cf59cadf6", "commit_rev":
"b573ab6f6f5f1b9a5ce1b56595590eae69563638"}

[commit-bot: I haz the power, 2017-04-26 19:21:14]

Description was changed from

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation
==========

to

==========
Cleanup Tool WebUI: Add functionality to the Cleanup button

This will allow users to manually launch the new Cleanup Tool prompt to
begin cleanup.

BUG=690020
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2819633003
Cr-Commit-Position: refs/heads/master@{#467408}
Committed:
https://chromium.googlesource.com/chromium/src/+/b573ab6f6f5f1b9a5ce1b5659559...
==========

[commit-bot: I haz the power, 2017-04-26 19:21:15]

Committed patchset #9 (id:160001) as
https://chromium.googlesource.com/chromium/src/+/b573ab6f6f5f1b9a5ce1b5659559...