Add ClientConfig proto, and JSON generation/parsing.

Add ClientConfig proto, and JSON generation/parsing.

- Add the ClientConfig proto definition corresponding to the response
  coming from the service.
- Create a means of generating a pseudo response from the built in
  params and request options.
- Create a class to parse the JSON response.
- Add necessary helper classes for the ProxyScheme and time values
  coming from the service.

BUG=466753
Committed: https://crrev.com/97652ebe50c4be712aa976776bee9694cbce1178
Cr-Commit-Position: refs/heads/master@{#321931}

Committed: https://crrev.com/9e0b2ffcac69765ab045fd6878301a8b4ea189ee
Cr-Commit-Position: refs/heads/master@{#322021}

[jeremyim, 2015-03-17 23:32:05]

jeremyim@chromium.org changed reviewers:
+ bengr@chromium.org, sclittle@chromium.org

[jeremyim, 2015-03-17 23:32:06]

PTAL. =)

[sclittle, 2015-03-18 20:31:51]

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:15:
class DictionaryValue;
Remove unused forward declaration

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:24:
// objects lives on the IO thread.
nit: s/objects/object/

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:25:
class DataReductionProxyConfigClient {
"ConfigClient" seems a bit ambiguous, maybe "ConfigFetcher" instead if it's just
retrieving the config?

Otherwise we have both "ConfigClient" and "ClientConfig".

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:29:
DataReductionProxyConfigClient(
This class doesn't have any public methods, and the constructor does nothing.
Why add an empty class? What is it supposed to do in the future? Why not just
add this class later along with it's functionality?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/client_config_parser.h (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.h:40: //
Returns false if the JSON has an unexpected structure.
I'm confused. You turn a JSON into a proto; why do you want to do this? If
you're using a JSON over the wire, then why use a proto at all?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/common/client_config_parser_unittest.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser_unittest.cc:40:
for (size_t i = 0; i < arraysize(tests); ++i) {
Use range-based for loops; e.g. for (const auto& test : tests)

Same for the other tests in this file.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/proto/BUILD.gn (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/BUILD.gn:7:
proto_library("data_reduction_proxy_proto") {
What needs to depend on this proto? Could this "proto/" directory be put under
core/?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/proto/client_config.proto (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:26: // nanosecond
resolution in UTC Epoch time.
Why do you need nanosecond resolution? Are microseconds really too coarse for
determining when to request a new session key?

[bengr, 2015-03-19 16:31:15]

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc:27:
std::string DataReductionProxyConfigClient::ConstructStaticResponse() {
Can this method be const?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc:33:

Remove blank line.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:33:
virtual ~DataReductionProxyConfigClient();
Does this need to be virtual?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:39:
// Constructs a synthetic response based on |params_|.
Why is it 'synthetic'?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client_unittest.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client_unittest.cc:49:
EXPECT_TRUE(ParseClientConfig(config_data, &config));
Maybe also add a test that expects false on a bogus config?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:78:
return base::StringPrintf("%s|%s", session.c_str(), credentials.c_str());
Can session or credentials be empty?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:85:
std::string* credentials) {
Can session or credentials be NULL?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:88:
if (auth_values.size() == 2) {
Shouldn't this always be 2? Maybe DCHECK instead and have the function return
void?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:230:
if (now > credentials_expiration_time_) {
Remove curly braces

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.h:71:
static std::string CreateLocalSessionKey(const std::string& session,
I'm assuming '|' is not a valid character in a session or credentials
identifier. Is this true? Please comment on the assumptions, what these methods
do, specifically, and why they're needed.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h:52:
scoped_refptr<base::SingleThreadTaskRunner> task_runner);
Includes needed for string, scoped_refptr and SingleThreadTaskRunner

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h:56:
base::Time Now() const override;
#include "base/time/time.h"

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/client_config_parser.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.cc:26:
std::string GetSchemeString(net::ProxyServer::Scheme scheme) {
Since these methods are all public to the namespace, do you think a prefix would
be useful, e.g., ConfigParser?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.cc:128: for (;
it != http_proxy_servers->end(); ++it) {
Does this loop never encounter conditions to return false?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/client_config_parser.h (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.h:40: //
Returns false if the JSON has an unexpected structure.
On 2015/03/18 20:31:51, sclittle wrote:
> I'm confused. You turn a JSON into a proto; why do you want to do this? If
> you're using a JSON over the wire, then why use a proto at all?

I agree that some explanation is in order. Please add a comment somewhere.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc:393:
scoped_ptr<base::Value> proxy_config(new base::DictionaryValue());
#include "base/memory/scoped_ptr.h"

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc:407:
server->SetString("scheme", GetSchemeString(origin_.scheme()));
It might be worthwhile to define constants for "scheme", "host", and "port".

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/proto/client_config.proto (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:21: optional
ProxyConfig proxy_config = 3;
Where do you say what kind of traffic should be proxied? I.e., how would we
extend this to having a configuration for HTTPS traffic?

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:37: optional int32
nanos = 2;
Why not use uint32?

[jeremyim, 2015-03-20 16:21:12]

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc:27:
std::string DataReductionProxyConfigClient::ConstructStaticResponse() {
On 2015/03/19 16:31:14, bengr wrote:
> Can this method be const?

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.cc:33:

On 2015/03/19 16:31:14, bengr wrote:
> Remove blank line.

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:15:
class DictionaryValue;
On 2015/03/18 20:31:51, sclittle wrote:
> Remove unused forward declaration

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:24:
// objects lives on the IO thread.
On 2015/03/18 20:31:51, sclittle wrote:
> nit: s/objects/object/

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:25:
class DataReductionProxyConfigClient {
On 2015/03/18 20:31:51, sclittle wrote:
> "ConfigClient" seems a bit ambiguous, maybe "ConfigFetcher" instead if it's
just
> retrieving the config?
> 
> Otherwise we have both "ConfigClient" and "ClientConfig".

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:29:
DataReductionProxyConfigClient(
On 2015/03/18 20:31:51, sclittle wrote:
> This class doesn't have any public methods, and the constructor does nothing.
> Why add an empty class? What is it supposed to do in the future? Why not just
> add this class later along with it's functionality?

The next CL is already pretty large; the more I can move into this CL that is
fairly self-contained makes the future easier.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:33:
virtual ~DataReductionProxyConfigClient();
On 2015/03/19 16:31:14, bengr wrote:
> Does this need to be virtual?

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client.h:39:
// Constructs a synthetic response based on |params_|.
On 2015/03/19 16:31:14, bengr wrote:
> Why is it 'synthetic'?

It is synthetic in that it is locally manufactured, and not from the service.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client_unittest.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_config_client_unittest.cc:49:
EXPECT_TRUE(ParseClientConfig(config_data, &config));
On 2015/03/19 16:31:14, bengr wrote:
> Maybe also add a test that expects false on a bogus config?

ParseClientConfig already has a decent set of tests. This is just making sure
that the static response that we will be generating locally is valid.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:78:
return base::StringPrintf("%s|%s", session.c_str(), credentials.c_str());
On 2015/03/19 16:31:14, bengr wrote:
> Can session or credentials be empty?

Not if they are valid session/credentials.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:85:
std::string* credentials) {
On 2015/03/19 16:31:14, bengr wrote:
> Can session or credentials be NULL?

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:88:
if (auth_values.size() == 2) {
On 2015/03/19 16:31:14, bengr wrote:
> Shouldn't this always be 2? Maybe DCHECK instead and have the function return
> void?

Even with a DCHECK, we would have to evaluate the if statement; and returning a
bool seems better from a caller perspective to know if it succeeded or not (in
non debug code).

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.cc:230:
if (now > credentials_expiration_time_) {
On 2015/03/19 16:31:14, bengr wrote:
> Remove curly braces

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_request_options.h:71:
static std::string CreateLocalSessionKey(const std::string& session,
On 2015/03/19 16:31:14, bengr wrote:
> I'm assuming '|' is not a valid character in a session or credentials
> identifier. Is this true? Please comment on the assumptions, what these
methods
> do, specifically, and why they're needed.

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h:52:
scoped_refptr<base::SingleThreadTaskRunner> task_runner);
On 2015/03/19 16:31:14, bengr wrote:
> Includes needed for string, scoped_refptr and SingleThreadTaskRunner

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/browser/data_reduction_proxy_test_utils.h:56:
base::Time Now() const override;
On 2015/03/19 16:31:15, bengr wrote:
> #include "base/time/time.h"

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/client_config_parser.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.cc:26:
std::string GetSchemeString(net::ProxyServer::Scheme scheme) {
On 2015/03/19 16:31:15, bengr wrote:
> Since these methods are all public to the namespace, do you think a prefix
would
> be useful, e.g., ConfigParser?

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.cc:128: for (;
it != http_proxy_servers->end(); ++it) {
On 2015/03/19 16:31:15, bengr wrote:
> Does this loop never encounter conditions to return false?

Correct; I am asserting that it is valid to return an empty proxy server list if
no valid servers were sent down.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/client_config_parser.h (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser.h:40: //
Returns false if the JSON has an unexpected structure.
On 2015/03/19 16:31:15, bengr wrote:
> On 2015/03/18 20:31:51, sclittle wrote:
> > I'm confused. You turn a JSON into a proto; why do you want to do this? If
> > you're using a JSON over the wire, then why use a proto at all?
> 
> I agree that some explanation is in order. Please add a comment somewhere.

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File
components/data_reduction_proxy/core/common/client_config_parser_unittest.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/client_config_parser_unittest.cc:40:
for (size_t i = 0; i < arraysize(tests); ++i) {
On 2015/03/18 20:31:51, sclittle wrote:
> Use range-based for loops; e.g. for (const auto& test : tests)
> 
> Same for the other tests in this file.

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc
(right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc:393:
scoped_ptr<base::Value> proxy_config(new base::DictionaryValue());
On 2015/03/19 16:31:15, bengr wrote:
> #include "base/memory/scoped_ptr.h"

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/core/common/data_reduction_proxy_params.cc:407:
server->SetString("scheme", GetSchemeString(origin_.scheme()));
On 2015/03/19 16:31:15, bengr wrote:
> It might be worthwhile to define constants for "scheme", "host", and "port".

Done.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/proto/BUILD.gn (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/BUILD.gn:7:
proto_library("data_reduction_proxy_proto") {
On 2015/03/18 20:31:51, sclittle wrote:
> What needs to depend on this proto? Could this "proto/" directory be put under
> core/?

The convention for several other components seems to be to have proto files in a
separate proto directory.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
File components/data_reduction_proxy/proto/client_config.proto (right):

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:21: optional
ProxyConfig proxy_config = 3;
On 2015/03/19 16:31:15, bengr wrote:
> Where do you say what kind of traffic should be proxied? I.e., how would we
> extend this to having a configuration for HTTPS traffic?

Line 42: // Provides proxy server information for HTTP URIs.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:26: // nanosecond
resolution in UTC Epoch time.
On 2015/03/18 20:31:51, sclittle wrote:
> Why do you need nanosecond resolution? Are microseconds really too coarse for
> determining when to request a new session key?

We don't need it; however, the service definition defines its presence.

https://codereview.chromium.org/1017853003/diff/40001/components/data_reducti...
components/data_reduction_proxy/proto/client_config.proto:37: optional int32
nanos = 2;
On 2015/03/19 16:31:15, bengr wrote:
> Why not use uint32?

Another restriction due to the service definition.

[bengr, 2015-03-23 21:59:11]

lgtm, with a q.

https://codereview.chromium.org/1017853003/diff/80001/components/data_reducti...
File components/data_reduction_proxy.gypi (right):

https://codereview.chromium.org/1017853003/diff/80001/components/data_reducti...
components/data_reduction_proxy.gypi:143:
'data_reduction_proxy/core/common/client_config_parser.cc',
Why no drp prefix?

[jeremyim, 2015-03-24 03:45:11]

The CQ bit was checked by jeremyim@chromium.org

[jeremyim, 2015-03-24 03:45:11]

The patchset sent to the CQ was uploaded after l-g-t-m from bengr@chromium.org
Link to the patchset: https://codereview.chromium.org/1017853003/#ps100001
(title: "Add data_reduction_proxy prefix")

[commit-bot: I haz the power, 2015-03-24 03:45:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1017853003/100001

[commit-bot: I haz the power, 2015-03-24 04:08:49]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-03-24 04:10:20]

Patchset 6 (id:??) landed as
https://crrev.com/97652ebe50c4be712aa976776bee9694cbce1178
Cr-Commit-Position: refs/heads/master@{#321931}

[Matt Giuca, 2015-03-24 05:14:33]

A revert of this CL (patchset #6 id:100001) has been created in
https://codereview.chromium.org/1020863009/ by mgiuca@chromium.org.

The reason for reverting is: Seems that some of the tests introduced in this CL
are failing on builders:

http://build.chromium.org/p/chromium.linux/builders/Linux%20Tests%20%28dbg%29...

Value of: config_parser::TimeToISO8601(test.time)
Actual: "2009-04-22T19:24:48.000Z"
Expected: test.expected
Which is: "1601-01-01T00:00:00.000Z"
[  FAILED  ] ClientConfigParserTest.TimeToISO8601 (0 ms)

Value of: config_parser::TimestampToTime(ts)
Actual: 1964-11-24 17:31:43.000 UTC
Expected: test.expected_time
Which is: 2038-01-19 03:14:07.999 UTC
[  FAILED  ] ClientConfigParserTest.TimestampToTime (0 ms)

No idea why they are failing there, or why they passed on the trybots..

[jeremyim, 2015-03-24 17:16:42]

The CQ bit was checked by jeremyim@chromium.org

[jeremyim, 2015-03-24 17:16:42]

The patchset sent to the CQ was uploaded after l-g-t-m from bengr@chromium.org
Link to the patchset: https://codereview.chromium.org/1017853003/#ps120001
(title: "Tweak test values")

[commit-bot: I haz the power, 2015-03-24 17:17:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1017853003/120001

[commit-bot: I haz the power, 2015-03-24 17:21:49]

Committed patchset #7 (id:120001)

[commit-bot: I haz the power, 2015-03-24 17:22:27]

Patchset 7 (id:??) landed as
https://crrev.com/9e0b2ffcac69765ab045fd6878301a8b4ea189ee
Cr-Commit-Position: refs/heads/master@{#322021}