Pull SecurityStateModel out into a component

Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071
Committed: https://crrev.com/d9e54fb05c6c0a2d780d5db406ce2cd47da61ab3
Cr-Commit-Position: refs/heads/master@{#368636}

[estark, 2015-12-21 04:01:05]

Patchset #5 (id:80001) has been deleted

[estark, 2015-12-21 22:46:50]

estark@chromium.org changed reviewers:
+ felt@chromium.org

[estark, 2015-12-21 22:46:50]

felt: we are nearing the end of SecurityStateModel c14n CLs... this is a fairly
mechanical CL that just moves SSM and SSMClient into its own component now that
all the relevant dependencies have been removed. (It looks large but almost all
of the changes are just renaming places that use the SecurityLevel enum.) I left
in a duplicated enum in security_state_model_android.h; in a follow-up CL I'll
remove that and update all the references to the SecurityLevel enum in
Java-land.

No rush on this one because I'll want blundell@ to take a look at it and he's
OOO for a while.

[estark, 2015-12-21 22:48:36]

Oh, I forgot to mention, there's a linux_chromium_rel_ng components_unittest
failure that's quite mysterious:
`net::ImportCertFromFile(net::GetTestCertsDirectory(), "sha1_2016.pem")` seems
to return null for some reason when the unit tests are pulled into
components/security_state instead of living in chrome/browser/ssl. Let me know
if anything jumps out at you as a likely culprit for that.

[felt, 2015-12-22 17:19:49]

On 2015/12/21 22:48:36, estark wrote:
> Oh, I forgot to mention, there's a linux_chromium_rel_ng components_unittest
> failure that's quite mysterious:
> `net::ImportCertFromFile(net::GetTestCertsDirectory(), "sha1_2016.pem")` seems
> to return null for some reason when the unit tests are pulled into
> components/security_state instead of living in chrome/browser/ssl. Let me know
> if anything jumps out at you as a likely culprit for that.

my guess would be a problem with the BUILD.gn not correctly bringing in the net
dependency, but you already have net:test_support, so it's not clear to me
what's missing.

[felt, 2015-12-22 17:24:10]

On 2015/12/22 17:19:49, felt wrote:
> On 2015/12/21 22:48:36, estark wrote:
> > Oh, I forgot to mention, there's a linux_chromium_rel_ng components_unittest
> > failure that's quite mysterious:
> > `net::ImportCertFromFile(net::GetTestCertsDirectory(), "sha1_2016.pem")`
seems
> > to return null for some reason when the unit tests are pulled into
> > components/security_state instead of living in chrome/browser/ssl. Let me
know
> > if anything jumps out at you as a likely culprit for that.
> 
> my guess would be a problem with the BUILD.gn not correctly bringing in the
net
> dependency, but you already have net:test_support, so it's not clear to me
> what's missing.

is GetTestCertsDirectory returning the right value?

[felt, 2015-12-22 17:29:08]

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
File chrome/browser/ssl/chrome_security_state_model_client.h (right):

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
chrome/browser/ssl/chrome_security_state_model_client.h:25: const
security_state::SecurityStateModel::SecurityInfo& GetSecurityInfo()
nit: this is repeated often enough that it might merit a `using
security_state::SecurityStateModel`

same comment throughout the files

[tfarina, 2015-12-22 17:30:30]

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
File chrome/browser/ssl/chrome_security_state_model_client.h (right):

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
chrome/browser/ssl/chrome_security_state_model_client.h:25: const
security_state::SecurityStateModel::SecurityInfo& GetSecurityInfo()
On 2015/12/22 17:29:08, felt wrote:
> nit: this is repeated often enough that it might merit a `using
> security_state::SecurityStateModel`
> 
certainly, you won't be doing this in a header file.

> same comment throughout the files
maybe on the source files though.

[felt, 2015-12-22 17:34:30]

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
File chrome/browser/ssl/chrome_security_state_model_client.h (right):

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
chrome/browser/ssl/chrome_security_state_model_client.h:25: const
security_state::SecurityStateModel::SecurityInfo& GetSecurityInfo()
On 2015/12/22 17:30:30, tfarina wrote:
> On 2015/12/22 17:29:08, felt wrote:
> > nit: this is repeated often enough that it might merit a `using
> > security_state::SecurityStateModel`
> > 
> certainly, you won't be doing this in a header file.
> 
> > same comment throughout the files
> maybe on the source files though.

sorry, i meant to make this comment on the next (.cc) file :)

[tfarina, 2015-12-22 17:44:29]

estark, could you describe in the CL description why are you making this change?
It is not clear, without looking at the bug (which I haven't), why
SecurityStateModel class needs to be made into a component.

[estark, 2015-12-22 17:47:04]

Description was changed from

==========
Pull SecurityStateModel out into a component

BUG=515071
==========

to

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic to calculate a page's overall security
status for UI on all platforms.

BUG=515071
==========

[estark, 2015-12-22 17:47:28]

Description was changed from

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic to calculate a page's overall security
status for UI on all platforms.

BUG=515071
==========

to

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071
==========

[estark, 2015-12-22 17:47:48]

On 2015/12/22 17:44:29, tfarina wrote:
> estark, could you describe in the CL description why are you making this
change?
> It is not clear, without looking at the bug (which I haven't), why
> SecurityStateModel class needs to be made into a component.

Done.

[estark, 2015-12-22 18:31:12]

Thanks. I'm setting up a gn build locally so I can investigate the test failure
more efficiently (probably shoulda done that earlier).

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
File chrome/browser/ssl/chrome_security_state_model_client.h (right):

https://codereview.chromium.org/1539043002/diff/120001/chrome/browser/ssl/chr...
chrome/browser/ssl/chrome_security_state_model_client.h:25: const
security_state::SecurityStateModel::SecurityInfo& GetSecurityInfo()
On 2015/12/22 17:34:30, felt wrote:
> On 2015/12/22 17:30:30, tfarina wrote:
> > On 2015/12/22 17:29:08, felt wrote:
> > > nit: this is repeated often enough that it might merit a `using
> > > security_state::SecurityStateModel`
> > > 
> > certainly, you won't be doing this in a header file.
> > 
> > > same comment throughout the files
> > maybe on the source files though.
> 
> sorry, i meant to make this comment on the next (.cc) file :)

Done in .cc files where it's used more than a handful of times.

[estark, 2016-01-04 18:12:21]

estark@chromium.org changed reviewers:
+ blundell@chromium.org

[estark, 2016-01-04 18:12:21]

+blundell: this was supposed to be a pretty straightforward CL where I just pull
the SecurityStateModel code out into a component, but I'm getting some
mysterious linux_chromium_rel_ng test failures and was wondering if you might be
able to help. I can't reproduce the failures locally (urgh!) but, on the bot,
unit tests are failing because the certificate file doesn't seem to exist. That
is, net::ImportCertFromFile() returns null, even though the path looks correct.

[blundell, 2016-01-05 15:47:29]

On 2016/01/04 18:12:21, estark wrote:
> +blundell: this was supposed to be a pretty straightforward CL where I just
pull
> the SecurityStateModel code out into a component, but I'm getting some
> mysterious linux_chromium_rel_ng test failures and was wondering if you might
be
> able to help. I can't reproduce the failures locally (urgh!) but, on the bot,
> unit tests are failing because the certificate file doesn't seem to exist.
That
> is, net::ImportCertFromFile() returns null, even though the path looks
correct.

Hi Emily,

I asked the master of swarming (maruel@) and got a quick resolution to the
problem :).

TL;DR: You need to add //net/data to the |data| list in the components_unittests
target. 

These tests run on the bots using swarming. One side-effect of that is that they
copy data into a tmp dir to allow multiple parallel runs to use that data in
parallel as necessary. They don't know what data to copy a priori, so they only
copy the data that is specified. The specification is done differently depending
on whether GYP or GN is being used. In GYP, it's <foo>.isolate that specifies
the data dependencies (e.g., components_unittests.isolate in this case). In GN,
it's specified as part of the |data| list in the target in question (e.g.,
components_unittests in this case). The system then queries the data
dependencies of the test target being run in order to know what data needs to be
copied (GN is by design much more supportive of introspection than GYP in
general).

Once GYP is killed, the .isolate files will go away. Until then, this
duplication is another unhappy consequence of having the two meta-build systems
(note that ../net/data *is* present in components_unittests.isolate, which is
why the GYP-based bots aren't failing).

Somewhat surprisingly, your test appears to be the first one in
components_unittests that is actually using data from //net/data.

[blundell, 2016-01-05 16:02:16]

Looks good, just one question about the java.

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
File chrome/browser/ssl/security_state_model_android.h (right):

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
chrome/browser/ssl/security_state_model_android.h:10: // TODO(estark): This is a
temporary duplicated enum that is only used
Hmm, this seems unfortunate. What's the blocker on switching over the Java code
at this time?

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
File chrome/browser/ui/toolbar/chrome_toolbar_model.h (right):

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
chrome/browser/ui/toolbar/chrome_toolbar_model.h:12: // TODO(blundell): Now that
SecurityStateModel::SecurityLevel is componentized,
woohoo!!!!

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
File components/security_state/security_state_model_unittest.cc (right):

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
components/security_state/security_state_model_unittest.cc:30: LOG(ERROR) <<
"Test certs directory: " << net::GetTestCertsDirectory().value();
this should go before landing, of course.

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
File components/security_state/switches.h (right):

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
components/security_state/switches.h:13: extern const char
kMarkNonSecureAsNonSecure[];
nit: blank space after (and in cc file).

[estark, 2016-01-05 22:12:43]

Patchset #10 (id:200001) has been deleted

[estark, 2016-01-05 22:59:30]

Woo, thanks blundell for the fix and explanation. Worked like a charm.

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
File chrome/browser/ssl/security_state_model_android.h (right):

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
chrome/browser/ssl/security_state_model_android.h:10: // TODO(estark): This is a
temporary duplicated enum that is only used
On 2016/01/05 16:02:16, blundell wrote:
> Hmm, this seems unfortunate. What's the blocker on switching over the Java
code
> at this time?

No blocker, I was just trying to keep the CL to a somewhat manageable size.
Switching over the Java code would just involve touching a whole bunch more
files. I'm working on doing it in a separate CL right now so we can see how big
it actually is and then see if it makes sense to just fold it into this one.

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
File chrome/browser/ui/toolbar/chrome_toolbar_model.h (right):

https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
chrome/browser/ui/toolbar/chrome_toolbar_model.h:12: // TODO(blundell): Now that
SecurityStateModel::SecurityLevel is componentized,
On 2016/01/05 16:02:16, blundell wrote:
> woohoo!!!!

:)

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
File components/security_state/security_state_model_unittest.cc (right):

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
components/security_state/security_state_model_unittest.cc:30: LOG(ERROR) <<
"Test certs directory: " << net::GetTestCertsDirectory().value();
On 2016/01/05 16:02:16, blundell wrote:
> this should go before landing, of course.

Done.

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
File components/security_state/switches.h (right):

https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
components/security_state/switches.h:13: extern const char
kMarkNonSecureAsNonSecure[];
On 2016/01/05 16:02:16, blundell wrote:
> nit: blank space after (and in cc file).

Done.

[blundell, 2016-01-06 10:38:59]

On 2016/01/05 22:59:30, estark wrote:
> Woo, thanks blundell for the fix and explanation. Worked like a charm.
> 
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
> File chrome/browser/ssl/security_state_model_android.h (right):
> 
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
> chrome/browser/ssl/security_state_model_android.h:10: // TODO(estark): This is
a
> temporary duplicated enum that is only used
> On 2016/01/05 16:02:16, blundell wrote:
> > Hmm, this seems unfortunate. What's the blocker on switching over the Java
> code
> > at this time?
> 
> No blocker, I was just trying to keep the CL to a somewhat manageable size.
> Switching over the Java code would just involve touching a whole bunch more
> files. I'm working on doing it in a separate CL right now so we can see how
big
> it actually is and then see if it makes sense to just fold it into this one.
> 

My preference would be to have it in the same CL if possible even if it makes
the CL larger; it will be a different person reviewing those changes in any
case.

>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
> File chrome/browser/ui/toolbar/chrome_toolbar_model.h (right):
> 
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
> chrome/browser/ui/toolbar/chrome_toolbar_model.h:12: // TODO(blundell): Now
that
> SecurityStateModel::SecurityLevel is componentized,
> On 2016/01/05 16:02:16, blundell wrote:
> > woohoo!!!!
> 
> :)
> 
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> File components/security_state/security_state_model_unittest.cc (right):
> 
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> components/security_state/security_state_model_unittest.cc:30: LOG(ERROR) <<
> "Test certs directory: " << net::GetTestCertsDirectory().value();
> On 2016/01/05 16:02:16, blundell wrote:
> > this should go before landing, of course.
> 
> Done.
> 
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> File components/security_state/switches.h (right):
> 
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> components/security_state/switches.h:13: extern const char
> kMarkNonSecureAsNonSecure[];
> On 2016/01/05 16:02:16, blundell wrote:
> > nit: blank space after (and in cc file).
> 
> Done.

[estark, 2016-01-06 21:55:40]

On 2016/01/06 10:38:59, blundell wrote:
> On 2016/01/05 22:59:30, estark wrote:
> > Woo, thanks blundell for the fix and explanation. Worked like a charm.
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
> > File chrome/browser/ssl/security_state_model_android.h (right):
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ssl/sec...
> > chrome/browser/ssl/security_state_model_android.h:10: // TODO(estark): This
is
> a
> > temporary duplicated enum that is only used
> > On 2016/01/05 16:02:16, blundell wrote:
> > > Hmm, this seems unfortunate. What's the blocker on switching over the Java
> > code
> > > at this time?
> > 
> > No blocker, I was just trying to keep the CL to a somewhat manageable size.
> > Switching over the Java code would just involve touching a whole bunch more
> > files. I'm working on doing it in a separate CL right now so we can see how
> big
> > it actually is and then see if it makes sense to just fold it into this one.

Ok, no problem, added it to this CL.

> > 
> 
> My preference would be to have it in the same CL if possible even if it makes
> the CL larger; it will be a different person reviewing those changes in any
> case.
> 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
> > File chrome/browser/ui/toolbar/chrome_toolbar_model.h (right):
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/chrome/browser/ui/tool...
> > chrome/browser/ui/toolbar/chrome_toolbar_model.h:12: // TODO(blundell): Now
> that
> > SecurityStateModel::SecurityLevel is componentized,
> > On 2016/01/05 16:02:16, blundell wrote:
> > > woohoo!!!!
> > 
> > :)
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> > File components/security_state/security_state_model_unittest.cc (right):
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> > components/security_state/security_state_model_unittest.cc:30: LOG(ERROR) <<
> > "Test certs directory: " << net::GetTestCertsDirectory().value();
> > On 2016/01/05 16:02:16, blundell wrote:
> > > this should go before landing, of course.
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> > File components/security_state/switches.h (right):
> > 
> >
>
https://codereview.chromium.org/1539043002/diff/160001/components/security_st...
> > components/security_state/switches.h:13: extern const char
> > kMarkNonSecureAsNonSecure[];
> > On 2016/01/05 16:02:16, blundell wrote:
> > > nit: blank space after (and in cc file).
> > 
> > Done.

[blundell, 2016-01-07 08:39:29]

blundell@chromium.org changed reviewers:
+ yfriedman@chromium.org

[blundell, 2016-01-07 08:39:30]

w00t!

LGTM

yfriedman@: Can you do an OWNERS review of //chrome/android for the move of the
generated Java enum corresponding to the C++ enum being componentized in
security_state_model.h? You'll probably also want to look at the various gypfile
changes in //chrome and //components, as that's where the actual meaningful
change is.

https://codereview.chromium.org/1539043002/diff/280001/components/security_st...
File components/security_state/security_state_model.h (right):

https://codereview.chromium.org/1539043002/diff/280001/components/security_st...
components/security_state/security_state_model.h:27: // A Java counterpart will
be generated for this enum.
nit: the comments added at lines 26-29 seem unnecessary, unless I'm missing
something about how the process works.

[felt, 2016-01-07 14:24:28]

lgtm

[estark, 2016-01-07 19:18:25]

https://codereview.chromium.org/1539043002/diff/280001/components/security_st...
File components/security_state/security_state_model.h (right):

https://codereview.chromium.org/1539043002/diff/280001/components/security_st...
components/security_state/security_state_model.h:27: // A Java counterpart will
be generated for this enum.
On 2016/01/07 08:39:30, blundell wrote:
> nit: the comments added at lines 26-29 seem unnecessary, unless I'm missing
> something about how the process works.

Oops, just a copy/paste error. Removed.

[Yaron, 2016-01-07 21:59:42]

lgtm

[estark, 2016-01-08 15:56:00]

estark@chromium.org changed reviewers:
+ eroman@chromium.org, sky@chromium.org

[estark, 2016-01-08 15:56:01]

sky, can you please review changes to chrome/browser/BUILD.gn,
chrome/browser/DEPS, chrome/browser/ui, chrome/test/base?

And eroman, I think I need a net owner's approval because the DEPS file of the
new component includes net.

[sky, 2016-01-08 17:21:26]

security_state should is a long namespace name. LGTM

[eroman, 2016-01-11 18:12:12]

lgtm

https://codereview.chromium.org/1539043002/diff/300001/components/security_st...
File components/security_state/security_state_model_unittest.cc (right):

https://codereview.chromium.org/1539043002/diff/300001/components/security_st...
components/security_state/security_state_model_unittest.cc:78: bool
IsOriginSecure(const GURL& url) override { return true; }
Without knowing more about this test, why is this change correct? Is the
assumption that all the tests are calling this with secure origins, or that this
doesn't matter for the test?

[estark, 2016-01-11 18:15:52]

Thanks for the reviews, everyone!

https://codereview.chromium.org/1539043002/diff/300001/components/security_st...
File components/security_state/security_state_model_unittest.cc (right):

https://codereview.chromium.org/1539043002/diff/300001/components/security_st...
components/security_state/security_state_model_unittest.cc:78: bool
IsOriginSecure(const GURL& url) override { return true; }
On 2016/01/11 18:12:12, eroman wrote:
> Without knowing more about this test, why is this change correct? Is the
> assumption that all the tests are calling this with secure origins, or that
this
> doesn't matter for the test?

The former (all tests are using secure origins). I'll add a comment saying so.

[estark, 2016-01-11 18:18:10]

The CQ bit was checked by estark@chromium.org

[estark, 2016-01-11 18:18:10]

The patchset sent to the CQ was uploaded after l-g-t-m from felt@chromium.org,
blundell@chromium.org, yfriedman@chromium.org, eroman@chromium.org,
sky@chromium.org
Link to the patchset: https://codereview.chromium.org/1539043002/#ps320001
(title: "add comment to address eroman's question")

[commit-bot: I haz the power, 2016-01-11 18:18:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1539043002/320001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1539043002/320001

[commit-bot: I haz the power, 2016-01-11 18:21:27]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-01-11 18:21:28]

Try jobs failed on following builders:
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  ios_rel_device_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ni...)
  mac_chromium_compile_dbg_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)

[estark, 2016-01-11 18:29:52]

The CQ bit was checked by estark@chromium.org

[estark, 2016-01-11 18:29:52]

The patchset sent to the CQ was uploaded after l-g-t-m from
yfriedman@chromium.org, eroman@chromium.org, felt@chromium.org,
sky@chromium.org, blundell@chromium.org
Link to the patchset: https://codereview.chromium.org/1539043002/#ps340001
(title: "rebase")

[commit-bot: I haz the power, 2016-01-11 18:30:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1539043002/340001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1539043002/340001

[commit-bot: I haz the power, 2016-01-11 19:37:22]

Description was changed from

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071
==========

to

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071
==========

[commit-bot: I haz the power, 2016-01-11 19:37:24]

Committed patchset #16 (id:340001)

[commit-bot: I haz the power, 2016-01-11 19:38:28]

Description was changed from

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071
==========

to

==========
Pull SecurityStateModel out into a component

As a component, the SecurityStateModel logic can be reused on iOS, so
that we use consistent logic and definitions to describe a page's
overall security status for UI on all platforms.

BUG=515071

Committed: https://crrev.com/d9e54fb05c6c0a2d780d5db406ce2cd47da61ab3
Cr-Commit-Position: refs/heads/master@{#368636}
==========

[commit-bot: I haz the power, 2016-01-11 19:38:30]

Patchset 16 (id:??) landed as
https://crrev.com/d9e54fb05c6c0a2d780d5db406ce2cd47da61ab3
Cr-Commit-Position: refs/heads/master@{#368636}

[blundell, 2016-01-12 08:42:40]

On 2016/01/11 19:38:30, commit-bot: I haz the power wrote:
> Patchset 16 (id:??) landed as
> https://crrev.com/d9e54fb05c6c0a2d780d5db406ce2cd47da61ab3
> Cr-Commit-Position: refs/heads/master@{#368636}

\o/ \o/ \o/