Update syslog parser for new ChromeOS log format

Update syslog parser for new ChromeOS log format

In R39, ChromeOS rsyslogd no longer includes the hostname
field. Update SyslogParser accordingly.

While there:
- Update example shill log message to include process ID.
  (ChromeOS syslog recently started recording process IDs
  whenever possible.)
- Add a test log message from the kernel. This message does
  not include a process ID.
- Fix parsing of messages that do not have a process ID.
  Previously, for such messages, the process name would include
  a trailing ':'. With this change, they no longer do. (The
  process name will now be 'kernel', instead of 'kernel:'.)
- Simplify string EXPECTations. (No need for cstr() and STREQ.)

BUG=403558
TEST=unit test (ExtensionSyslogParserTest.ParseLog)

Committed: https://crrev.com/1a2f0e0889de76e03e14d454fef33711b4be3a3b
Cr-Commit-Position: refs/heads/master@{#292058}

[Zachary Kuznia, 2014-08-19 20:13:28]

lgtm

[mukesh agrawal, 2014-08-19 21:26:06]

The CQ bit was checked by quiche@chromium.org

[commit-bot: I haz the power, 2014-08-19 21:26:51]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/quiche@chromium.org/486873005/1

[commit-bot: I haz the power, 2014-08-19 22:13:16]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  android_dbg_tests_recipe on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/android_dbg_tes...)
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[commit-bot: I haz the power, 2014-08-19 22:16:53]

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

[commit-bot: I haz the power, 2014-08-19 22:16:54]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[mukesh agrawal, 2014-08-20 00:01:57]

jyasskin, can I trouble you for an OWNERS review?

[Jeffrey Yasskin, 2014-08-26 18:44:27]

"ChromeOS syslogs will soon drop the hostname field." implies that the hostname
field is still present, in which case, the new parsing logic will fail.

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
File chrome/browser/extensions/api/log_private/syslog_parser.cc (left):

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
chrome/browser/extensions/api/log_private/syslog_parser.cc:50: // Skips
"localhost" field.
Does this parser need to keep working before the change to drop the hostname? If
so, you need more logic here.

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
File chrome/browser/extensions/api/log_private/syslog_parser_unittest.cc
(right):

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
chrome/browser/extensions/api/log_private/syslog_parser_unittest.cc:43:
EXPECT_STREQ("unknown", output[0]->level.c_str());
FWIW, you can use:
  EXPECT_EQ("unknown", output[0]->level);
since output[9]->level is a std::string for which == works.

[mukesh agrawal, 2014-08-26 22:41:15]

thanks, and ptal

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
File chrome/browser/extensions/api/log_private/syslog_parser.cc (left):

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
chrome/browser/extensions/api/log_private/syslog_parser.cc:50: // Skips
"localhost" field.
On 2014/08/26 18:44:27, Jeffrey Yasskin wrote:
> Does this parser need to keep working before the change to drop the hostname?
If
> so, you need more logic here.

Updated commit message to reflect that ChromeOS syslog format has already
changed.

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
File chrome/browser/extensions/api/log_private/syslog_parser_unittest.cc
(right):

https://codereview.chromium.org/486873005/diff/1/chrome/browser/extensions/ap...
chrome/browser/extensions/api/log_private/syslog_parser_unittest.cc:43:
EXPECT_STREQ("unknown", output[0]->level.c_str());
On 2014/08/26 18:44:27, Jeffrey Yasskin wrote:
> FWIW, you can use:
>   EXPECT_EQ("unknown", output[0]->level);
> since output[9]->level is a std::string for which == works.

Done.

[Jeffrey Yasskin, 2014-08-26 22:46:51]

lgtm

[mukesh agrawal, 2014-08-27 00:08:06]

The CQ bit was checked by quiche@chromium.org

[commit-bot: I haz the power, 2014-08-27 00:11:46]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/quiche@chromium.org/486873005/20001

[commit-bot: I haz the power, 2014-08-27 01:08:26]

Committed patchset #2 (20001) as de75a4ca607ec0a35cb28654afb0add3fe51ff58

[commit-bot: I haz the power, 2014-09-10 02:48:22]

Patchset 2 (id:??) landed as
https://crrev.com/1a2f0e0889de76e03e14d454fef33711b4be3a3b
Cr-Commit-Position: refs/heads/master@{#292058}