Rewrite isolate_smoke_test.py to increase the coverage.

Rewrite isolate_smoke_test.py to increase the coverage.

Also fix discrepancy in the way timestamp was calculated. It was causing issues
on OSX and Windows.

R=nsylvain@chromium.org
BUG=98834
TEST=


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=132569

[nsylvain, 2012-04-13 15:49:58]

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
File tools/isolate/isolate_smoke_test.py (right):

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:93: return 420
do we really need to mess with modes on windows? Im not sure where it's used
though

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:104: #root_dir = os.path.join(root_dir,
RELATIVE_CWD[self.case()])
remove?

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:244: LEVEL = 1
i dont get the LEVEL thing yet.

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:256: pass
you expect it to go there right?  Can't you make sure it throws an exception?
Like set a variable and assert on it.

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:378: LEVEL = 2
i have a hard time understanding the difference between this test case and the
previous one.

[M-A Ruel, 2012-04-13 15:56:16]

Thanks, ptal.

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
File tools/isolate/isolate_smoke_test.py (right):

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:93: return 420
On 2012/04/13 15:49:58, nsylvain wrote:
> do we really need to mess with modes on windows? Im not sure where it's used
> though

In fact, it's not used. Let's look at removing this in another CL?

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:104: #root_dir = os.path.join(root_dir,
RELATIVE_CWD[self.case()])
On 2012/04/13 15:49:58, nsylvain wrote:
> remove?

done

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:244: LEVEL = 1
On 2012/04/13 15:49:58, nsylvain wrote:
> i dont get the LEVEL thing yet.

Used constants instead to relate to process_inputs().

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:256: pass
On 2012/04/13 15:49:58, nsylvain wrote:
> you expect it to go there right?  Can't you make sure it throws an exception?
> Like set a variable and assert on it.

See line 254, self.fail() would mark the test as failed if it is ever executed.

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:378: LEVEL = 2
On 2012/04/13 15:49:58, nsylvain wrote:
> i have a hard time understanding the difference between this test case and the
> previous one.

Cleaned up.

[nsylvain, 2012-04-14 21:56:06]

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
File tools/isolate/isolate_smoke_test.py (right):

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:378: LEVEL = 2
Still don't get it.

class Isolate_remap(IsolateBase):
  LEVEL = isolate.STATS_ONLY
  def test_fail(self):
    self._execute()
    
    
class Isolate_run(IsolateBase):
  LEVEL = isolate.STATS_ONLY
  def test_fail(self):
    self._execute()



Both tests are identical except the name, and yet the first one you expect it to
work, and the second one you expect it to fail. 

the current "Testing On The Toilet" is about making sure the "Cause" and the
"Effect" is clear.  So it needs to be obvious from the test why it would fail.  
I dont see the difference, other than the test name.. which should not matter
right?

On 2012/04/13 15:56:16, Marc-Antoine Ruel wrote:
> On 2012/04/13 15:49:58, nsylvain wrote:
> > i have a hard time understanding the difference between this test case and
the
> > previous one.
> 
> Cleaned up.

[Marc-Antoine Ruel (Google), 2012-04-14 22:33:58]

ptak

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
File tools/isolate/isolate_smoke_test.py (right):

https://chromiumcodereview.appspot.com/10068032/diff/1/tools/isolate/isolate_...
tools/isolate/isolate_smoke_test.py:378: LEVEL = 2
On 2012/04/14 21:56:07, nsylvain wrote:
> Still don't get it.

Should be much more explicit now.

[M-A Ruel, 2012-04-16 16:04:02]

On 2012/04/14 22:33:58, Marc-Antoine Ruel (Google) wrote:
> ptak

FTR, the current list in order is

10068032
10062001
10065031
10079007
10080013

They are all lgtm'ed except for the first one.

I have one on going at ~110 lines to correct case insensitivity issues with
Windows and OSX, then an .gyp/.isolate only to unit_tests_run which you don't
need to review, then I'm almost done. :)

[nsylvain, 2012-04-17 01:45:22]

lgtm

[commit-bot: I haz the power, 2012-04-17 12:21:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/maruel@chromium.org/10068032/8001

[commit-bot: I haz the power, 2012-04-17 13:45:55]

Change committed as 132569