Move site_chrome_test, site_logging, site_log_reader, site_crash_test into cros dir.

Change-Id: Ib50962111d0c6c07a135679d80930af4bc52915b

Move site_chrome_test, site_logging, site_log_reader, 
site_crash_test into cros dir.

site_chrome_test --> cros/chrome_test
site_crash_test --> cros/crash_test
site_logging --> cros/cros_logging
site_log_reader --> cros/cros_logging
site_ui_test --> cros/cros_ui_test

I had removed site_logging module and moved its code into cros_logging, since its new name (logging) has conflict with python's logging module.

site_log_reader has also been merged into cros_logging module.

As I am touching all the files in the list, I also removed some unnecessary import to autotest_lib.client.bin.test.


Change-Id: I2c8f101a6c7a0499619e227e4cb0cfb8ceec4690

BUG=7236
TEST=run bvt and every other tests been touched from both source (third_party/autotest/files) and chroot (/build/board/usr/local/autotest)

Committed: http://chrome-svn/viewvc/chromeos?view=rev&revision=e7c4cab

[ericli, 2010-12-15 19:17:27]

Ping?

On 2010/12/13 22:38:43, ericli wrote:

[kmixter1, 2010-12-16 03:12:57]

FYI

The new naming is a little confusing since it conflicts with existing python,
autotest, and variables.  I would suggest cros_ui_test for instance from cros
directory.  I know it's redundant but I find it unambiguous to read.  Adding
cmasone to see his opinion since some of these classes are his.

http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py
File client/bin/site_job.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py#newco...
client/bin/site_job.py:18: log_pauser = log_reader.LogRotationPauser()
If you collapse site_log_reader and site_logging, I would suggest logging.  Of
course that conflicts with the internal python name, so maybe adopt cros_logging

http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py
File client/cros/chrome_test.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py#n...
client/cros/chrome_test.py:6: import common
style is to put these all on one line.  Should you say "from" to make it
abundantly clear these are not system libs?

http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py
File client/cros/crash_test.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py#ne...
client/cros/crash_test.py:216: utils.poll_for_condition(
this is confusing with the utils that's part of common_lib. I would suggest
prefixing with site or cros.

http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py
File client/cros/login.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py#newcode93
client/cros/login.py:93: def process_crashed(process, logreader):
suggest renaming class and using the standard _ separation of words

http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
File client/site_tests/logging_CrashServices/logging_CrashServices.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
client/site_tests/logging_CrashServices/logging_CrashServices.py:6: from
autotest_lib.client.bin import  utils
extra space

[Chris Masone, 2010-12-16 06:37:33]

On 2010/12/16 03:13:28, kmixter1 wrote:

I prefer the unambiguous naming as well.  more readable code > more pithy code

[ericli, 2010-12-17 21:48:39]

PTAL.

http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py
File client/bin/site_job.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py#newco...
client/bin/site_job.py:18: log_pauser = log_reader.LogRotationPauser()
I agree. I have some initial concern about prefix cros_* to the name, but since
you guys OK with it, I am all for it.

I will also change ui to cros_ui in a separate CL.


On 2010/12/16 03:12:57, kmixter1 wrote:
> If you collapse site_log_reader and site_logging, I would suggest logging.  Of
> course that conflicts with the internal python name, so maybe adopt
cros_logging

http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py
File client/cros/chrome_test.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py#n...
client/cros/chrome_test.py:6: import common
common is a very special hack by autotest design, you could see it was not
actually referred to use anywhere in the code. 

import common as a standalone line is an usual practice is autotest code, just
grep "import common" client/bin and client/common_lib, you will see sufficient
examples.

On 2010/12/16 03:12:57, kmixter1 wrote:
> style is to put these all on one line.  Should you say "from" to make it
> abundantly clear these are not system libs?

http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py
File client/cros/crash_test.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py#ne...
client/cros/crash_test.py:216: utils.poll_for_condition(
cros_utils would be good. But due the massive number of occurrence of the
function, I would do that in another CL.

On 2010/12/16 03:12:57, kmixter1 wrote:
> this is confusing with the utils that's part of common_lib. I would suggest
> prefixing with site or cros.

http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py
File client/cros/login.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py#newcode93
client/cros/login.py:93: def process_crashed(process, logreader):
On 2010/12/16 03:12:57, kmixter1 wrote:
> suggest renaming class and using the standard _ separation of words

With the new naming, now it makes more sense.
Done.

http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
File client/site_tests/logging_CrashServices/logging_CrashServices.py (right):

http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
client/site_tests/logging_CrashServices/logging_CrashServices.py:6: from
autotest_lib.client.bin import  utils
On 2010/12/16 03:12:57, kmixter1 wrote:
> extra space

Done.

[Chris Masone, 2010-12-21 23:43:42]

I don't understand why we're changing site_ui->ui in this CL, and then
ui->cros_ui in a separate CL.  I mean, this is a cleanup CL.  Why not just
finish the cleanup?

On 2010/12/17 21:48:39, ericli wrote:
> PTAL.
> 
> http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py
> File client/bin/site_job.py (right):
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py#newco...
> client/bin/site_job.py:18: log_pauser = log_reader.LogRotationPauser()
> I agree. I have some initial concern about prefix cros_* to the name, but
since
> you guys OK with it, I am all for it.
> 
> I will also change ui to cros_ui in a separate CL.
> 
> 
> On 2010/12/16 03:12:57, kmixter1 wrote:
> > If you collapse site_log_reader and site_logging, I would suggest logging. 
Of
> > course that conflicts with the internal python name, so maybe adopt
> cros_logging
> 
> http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py
> File client/cros/chrome_test.py (right):
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py#n...
> client/cros/chrome_test.py:6: import common
> common is a very special hack by autotest design, you could see it was not
> actually referred to use anywhere in the code. 
> 
> import common as a standalone line is an usual practice is autotest code, just
> grep "import common" client/bin and client/common_lib, you will see sufficient
> examples.
> 
> On 2010/12/16 03:12:57, kmixter1 wrote:
> > style is to put these all on one line.  Should you say "from" to make it
> > abundantly clear these are not system libs?
> 
> http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py
> File client/cros/crash_test.py (right):
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py#ne...
> client/cros/crash_test.py:216: utils.poll_for_condition(
> cros_utils would be good. But due the massive number of occurrence of the
> function, I would do that in another CL.
> 
> On 2010/12/16 03:12:57, kmixter1 wrote:
> > this is confusing with the utils that's part of common_lib. I would suggest
> > prefixing with site or cros.
> 
> http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py
> File client/cros/login.py (right):
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py#newcode93
> client/cros/login.py:93: def process_crashed(process, logreader):
> On 2010/12/16 03:12:57, kmixter1 wrote:
> > suggest renaming class and using the standard _ separation of words
> 
> With the new naming, now it makes more sense.
> Done.
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
> File client/site_tests/logging_CrashServices/logging_CrashServices.py (right):
> 
>
http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
> client/site_tests/logging_CrashServices/logging_CrashServices.py:6: from
> autotest_lib.client.bin import  utils
> On 2010/12/16 03:12:57, kmixter1 wrote:
> > extra space
> 
> Done.

[ericli, 2010-12-22 00:02:56]

Chris,

Thanks for your time on this CL. I fully understand your concerns. But   this CL
has already been too many patches and I'd rather make a push before any further
changes.

This CL does not declare any victory on the cleanup effort, instead there is
still lots of things in front of us. I have already regretted the granularity of
this CL since it went too big. Would you bear with me on this temporary check
point?

By the way, this CL does not touch site_ui --> ui, only ui_test--> cros_ui_test,
as site_ui --> ui was pushed in previous CLs. That is why I prefer to do it in a
follow up.


On 2010/12/21 23:43:42, Chris Masone wrote:
> I don't understand why we're changing site_ui->ui in this CL, and then
> ui->cros_ui in a separate CL.  I mean, this is a cleanup CL.  Why not just
> finish the cleanup?
> 
> On 2010/12/17 21:48:39, ericli wrote:
> > PTAL.
> > 
> > http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py
> > File client/bin/site_job.py (right):
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py#newco...
> > client/bin/site_job.py:18: log_pauser = log_reader.LogRotationPauser()
> > I agree. I have some initial concern about prefix cros_* to the name, but
> since
> > you guys OK with it, I am all for it.
> > 
> > I will also change ui to cros_ui in a separate CL.
> > 
> > 
> > On 2010/12/16 03:12:57, kmixter1 wrote:
> > > If you collapse site_log_reader and site_logging, I would suggest logging.

> Of
> > > course that conflicts with the internal python name, so maybe adopt
> > cros_logging
> > 
> > http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py
> > File client/cros/chrome_test.py (right):
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py#n...
> > client/cros/chrome_test.py:6: import common
> > common is a very special hack by autotest design, you could see it was not
> > actually referred to use anywhere in the code. 
> > 
> > import common as a standalone line is an usual practice is autotest code,
just
> > grep "import common" client/bin and client/common_lib, you will see
sufficient
> > examples.
> > 
> > On 2010/12/16 03:12:57, kmixter1 wrote:
> > > style is to put these all on one line.  Should you say "from" to make it
> > > abundantly clear these are not system libs?
> > 
> > http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py
> > File client/cros/crash_test.py (right):
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py#ne...
> > client/cros/crash_test.py:216: utils.poll_for_condition(
> > cros_utils would be good. But due the massive number of occurrence of the
> > function, I would do that in another CL.
> > 
> > On 2010/12/16 03:12:57, kmixter1 wrote:
> > > this is confusing with the utils that's part of common_lib. I would
suggest
> > > prefixing with site or cros.
> > 
> > http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py
> > File client/cros/login.py (right):
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py#newcode93
> > client/cros/login.py:93: def process_crashed(process, logreader):
> > On 2010/12/16 03:12:57, kmixter1 wrote:
> > > suggest renaming class and using the standard _ separation of words
> > 
> > With the new naming, now it makes more sense.
> > Done.
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
> > File client/site_tests/logging_CrashServices/logging_CrashServices.py
(right):
> > 
> >
>
http://codereview.chromium.org/5740004/diff/8001/client/site_tests/logging_Cr...
> > client/site_tests/logging_CrashServices/logging_CrashServices.py:6: from
> > autotest_lib.client.bin import  utils
> > On 2010/12/16 03:12:57, kmixter1 wrote:
> > > extra space
> > 
> > Done.

[Chris Masone, 2011-01-04 01:53:45]

K...now that everyone's coming back at the same time, I'd rather have
Ken just finish his review of this.

On Tuesday, December 21, 2010,  <ericli@chromium.org> wrote:
> Chris,
>
> Thanks for your time on this CL. I fully understand your concerns. But   this
CL
> has already been too many patches and I'd rather make a push before any
further
> changes.
>
> This CL does not declare any victory on the cleanup effort, instead there is
> still lots of things in front of us. I have already regretted the granularity
of
> this CL since it went too big. Would you bear with me on this temporary check
> point?
>
> By the way, this CL does not touch site_ui --> ui, only ui_test-->
cros_ui_test,
> as site_ui --> ui was pushed in previous CLs. That is why I prefer to do it in
a
> follow up.
>
>
> On 2010/12/21 23:43:42, Chris Masone wrote:
>
> I don't understand why we're changing site_ui->ui in this CL, and then
> ui->cros_ui in a separate CL.  I mean, this is a cleanup CL.  Why not just
> finish the cleanup?
>
>
>
> On 2010/12/17 21:48:39, ericli wrote:
>> PTAL.
>>
>> http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py
>> File client/bin/site_job.py (right):
>>
>>
>
>
>
http://codereview.chromium.org/5740004/diff/8001/client/bin/site_job.py#newco...
>
>> client/bin/site_job.py:18: log_pauser = log_reader.LogRotationPauser()
>> I agree. I have some initial concern about prefix cros_* to the name, but
> since
>> you guys OK with it, I am all for it.
>>
>> I will also change ui to cros_ui in a separate CL.
>>
>>
>> On 2010/12/16 03:12:57, kmixter1 wrote:
>> > If you collapse site_log_reader and site_logging, I would suggest logging.
>
>
>
> Of
>> > course that conflicts with the internal python name, so maybe adopt
>> cros_logging
>>
>> http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py
>> File client/cros/chrome_test.py (right):
>>
>>
>
>
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/chrome_test.py#n...
>
>> client/cros/chrome_test.py:6: import common
>> common is a very special hack by autotest design, you could see it was not
>> actually referred to use anywhere in the code.
>>
>> import common as a standalone line is an usual practice is autotest code,
>
> just
>
>> grep "import common" client/bin and client/common_lib, you will see
>
> sufficient
>
>> examples.
>>
>> On 2010/12/16 03:12:57, kmixter1 wrote:
>> > style is to put these all on one line.  Should you say "from" to make it
>> > abundantly clear these are not system libs?
>>
>> http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py
>> File client/cros/crash_test.py (right):
>>
>>
>
>
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/crash_test.py#ne...
>
>> client/cros/crash_test.py:216: utils.poll_for_condition(
>> cros_utils would be good. But due the massive number of occurrence of the
>> function, I would do that in another CL.
>>
>> On 2010/12/16 03:12:57, kmixter1 wrote:
>> > this is confusing with the utils that's part of common_lib. I would
>
> suggest
>
>> > prefixing with site or cros.
>>
>> http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py
>> File client/cros/login.py (right):
>>
>>
>
>
>
http://codereview.chromium.org/5740004/diff/8001/client/cros/login.py#newcode93
>
>> client/cros/login.py:93: def process_crashed(process, logreader):
>> On 2010/12/16 03:12:57, kmixter1 wrote:

[kmixter1, 2011-01-05 19:44:24]

For testing, I'm assuming you're not just running bvt, but also emerging
autotest-tests?  Otherwise I guess syntax checking isn't happening on the
majority of tests.  Can you update the TEST stanza?

Can you also sync your repo so that the site_ui -> ui changes don't appear in
this CL?  I'm not sure what CL they were in.

I saw a few lines where import ui was alone.  Maybe "common" is special and
needs to be on its own line (as you suggested) but I don't know why "ui" would.

[kmixter1, 2011-01-05 19:50:39]

(chrome_test.py is an example of ui on its own line, and of showing a
site_ui -> ui change).

On Wed, Jan 5, 2011 at 11:44 AM, <kmixter@chromium.org> wrote:

> For testing, I'm assuming you're not just running bvt, but also emerging
> autotest-tests?  Otherwise I guess syntax checking isn't happening on the
> majority of tests.  Can you update the TEST stanza?
>
> Can you also sync your repo so that the site_ui -> ui changes don't appear
> in
> this CL?  I'm not sure what CL they were in.
>
> I saw a few lines where import ui was alone.  Maybe "common" is special and
> needs to be on its own line (as you suggested) but I don't know why "ui"
> would.
>
>
> http://codereview.chromium.org/5740004/
>

[ericli, 2011-01-05 19:56:09]

On Wed, Jan 5, 2011 at 11:44 AM, <kmixter@chromium.org> wrote:

> For testing, I'm assuming you're not just running bvt, but also emerging
> autotest-tests?  Otherwise I guess syntax checking isn't happening on the
> majority of tests.  Can you update the TEST stanza?
>
> Done.


> Can you also sync your repo so that the site_ui -> ui changes don't appear
> in
> this CL?  I'm not sure what CL they were in.
>
site_u --> ui change was landed on Dec. 10 by
http://git.chromium.org/gitweb/?p=autotest.git;a=commit;h=c4d8f4aab4e434fcb1a...
.


>
> I saw a few lines where import ui was alone.  Maybe "common" is special and
> needs to be on its own line (as you suggested) but I don't know why "ui"
> would.
>
> You are right. since now chrome_test and ui are under the same new location
(client/cros), you could directly import it after import common statement,
without specifying the full from ... import statement.

"import common" as a standalone statement seems a less-than-perfect autotest
idiom, but I will not touch it for now.


> http://codereview.chromium.org/5740004/
>



-- 
Eric Li
李咏竹
Google Kirkland

[kmixter1, 2011-01-05 20:40:33]

LGTM

I still think from autotest_lib.client.cros import ui is more readable than
just "ui" but I don't think there's any style guide requirements for it.

But is there any chance you can finish the ui -> site_ui change this week
before leaving?

On Wed, Jan 5, 2011 at 11:56 AM, Eric Li(李咏竹) <ericli@chromium.org> wrote:

> On Wed, Jan 5, 2011 at 11:44 AM, <kmixter@chromium.org> wrote:
>
>> For testing, I'm assuming you're not just running bvt, but also emerging
>> autotest-tests?  Otherwise I guess syntax checking isn't happening on the
>> majority of tests.  Can you update the TEST stanza?
>>
>> Done.
>
>
>> Can you also sync your repo so that the site_ui -> ui changes don't appear
>> in
>> this CL?  I'm not sure what CL they were in.
>>
> site_u --> ui change was landed on Dec. 10 by
>
http://git.chromium.org/gitweb/?p=autotest.git;a=commit;h=c4d8f4aab4e434fcb1a...
> .
>
>
>>
>> I saw a few lines where import ui was alone.  Maybe "common" is special
>> and
>> needs to be on its own line (as you suggested) but I don't know why "ui"
>> would.
>>
>> You are right. since now chrome_test and ui are under the same new
> location (client/cros), you could directly import it after import common
> statement, without specifying the full from ... import statement.
>
> "import common" as a standalone statement seems a less-than-perfect
> autotest idiom, but I will not touch it for now.
>
>
>> http://codereview.chromium.org/5740004/
>>
>
>
>
> --
> Eric Li
> 李咏竹
> Google Kirkland
>
>
>

[ericli, 2011-01-05 20:54:41]

Thanks, I will try. I think your mean ui --> cros_ui?

On Wed, Jan 5, 2011 at 12:40 PM, Ken Mixter <kmixter@chromium.org> wrote:

> LGTM
>
> I still think from autotest_lib.client.cros import ui is more readable than
> just "ui" but I don't think there's any style guide requirements for it.
>
> But is there any chance you can finish the ui -> site_ui change this week
> before leaving?
>
>
> On Wed, Jan 5, 2011 at 11:56 AM, Eric Li(李咏竹) <ericli@chromium.org> wrote:
>
>> On Wed, Jan 5, 2011 at 11:44 AM, <kmixter@chromium.org> wrote:
>>
>>> For testing, I'm assuming you're not just running bvt, but also emerging
>>> autotest-tests?  Otherwise I guess syntax checking isn't happening on the
>>> majority of tests.  Can you update the TEST stanza?
>>>
>>> Done.
>>
>>
>>> Can you also sync your repo so that the site_ui -> ui changes don't
>>> appear in
>>> this CL?  I'm not sure what CL they were in.
>>>
>> site_u --> ui change was landed on Dec. 10 by
>>
http://git.chromium.org/gitweb/?p=autotest.git;a=commit;h=c4d8f4aab4e434fcb1a...
>> .
>>
>>
>>>
>>> I saw a few lines where import ui was alone.  Maybe "common" is special
>>> and
>>> needs to be on its own line (as you suggested) but I don't know why "ui"
>>> would.
>>>
>>> You are right. since now chrome_test and ui are under the same new
>> location (client/cros), you could directly import it after import common
>> statement, without specifying the full from ... import statement.
>>
>> "import common" as a standalone statement seems a less-than-perfect
>> autotest idiom, but I will not touch it for now.
>>
>>
>>> http://codereview.chromium.org/5740004/
>>>
>>
>>
>>
>> --
>> Eric Li
>> 李咏竹
>> Google Kirkland
>>
>>
>>
>


-- 
Eric Li
李咏竹
Google Kirkland

[kmixter1, 2011-01-05 20:59:13]

Yes, exactly.

On Wed, Jan 5, 2011 at 12:54 PM, Eric Li(李咏竹) <ericli@chromium.org> wrote:

> Thanks, I will try. I think your mean ui --> cros_ui?
>
>
> On Wed, Jan 5, 2011 at 12:40 PM, Ken Mixter <kmixter@chromium.org> wrote:
>
>> LGTM
>>
>> I still think from autotest_lib.client.cros import ui is more readable
>> than just "ui" but I don't think there's any style guide requirements for
>> it.
>>
>> But is there any chance you can finish the ui -> site_ui change this week
>> before leaving?
>>
>>
>> On Wed, Jan 5, 2011 at 11:56 AM, Eric Li(李咏竹) <ericli@chromium.org>wrote:
>>
>>> On Wed, Jan 5, 2011 at 11:44 AM, <kmixter@chromium.org> wrote:
>>>
>>>> For testing, I'm assuming you're not just running bvt, but also emerging
>>>> autotest-tests?  Otherwise I guess syntax checking isn't happening on
>>>> the
>>>> majority of tests.  Can you update the TEST stanza?
>>>>
>>>> Done.
>>>
>>>
>>>> Can you also sync your repo so that the site_ui -> ui changes don't
>>>> appear in
>>>> this CL?  I'm not sure what CL they were in.
>>>>
>>> site_u --> ui change was landed on Dec. 10 by
>>>
http://git.chromium.org/gitweb/?p=autotest.git;a=commit;h=c4d8f4aab4e434fcb1a...
>>> .
>>>
>>>
>>>>
>>>> I saw a few lines where import ui was alone.  Maybe "common" is special
>>>> and
>>>> needs to be on its own line (as you suggested) but I don't know why "ui"
>>>> would.
>>>>
>>>> You are right. since now chrome_test and ui are under the same new
>>> location (client/cros), you could directly import it after import common
>>> statement, without specifying the full from ... import statement.
>>>
>>> "import common" as a standalone statement seems a less-than-perfect
>>> autotest idiom, but I will not touch it for now.
>>>
>>>
>>>> http://codereview.chromium.org/5740004/
>>>>
>>>
>>>
>>>
>>> --
>>> Eric Li
>>> 李咏竹
>>> Google Kirkland
>>>
>>>
>>>
>>
>
>
> --
> Eric Li
> 李咏竹
> Google Kirkland
>
>
>