Fix ProductTest.ProductInstallBasic and InstallerStateTest.InitializeTwice flakes.

Fix ProductTest.ProductInstallBasic and InstallerStateTest.InitializeTwice flakes.

Use registry virtualization so that stale data on the test machines
don't interfere with the tests.

BUG=375739
Committed: https://chromium.googlesource.com/chromium/src/+/6d6ac988d6a7a311876f3b0d0257086ac58c8181

Committed: https://crrev.com/4b1d16a9cbc61817d584c16fe2b3cc79dc0febc2
Cr-Commit-Position: refs/heads/master@{#292822}

[grt (UTC plus 2), 2014-08-29 18:41:34]

grt@chromium.org changed reviewers:
+ robertshield@google.com

[grt (UTC plus 2), 2014-08-29 18:42:15]

grt@chromium.org changed reviewers:
+ robertshield@chromium.org
- robertshield@google.com

[grt (UTC plus 2), 2014-08-29 18:42:45]

Hi Robert,

Please take a look at these two test improvements which should resolve some of
the test flakes. Thanks.

[robertshield, 2014-08-29 18:49:20]

https://codereview.chromium.org/521703002/diff/20001/chrome/installer/util/in...
File chrome/installer/util/installer_state_unittest.cc (right):

https://codereview.chromium.org/521703002/diff/20001/chrome/installer/util/in...
chrome/installer/util/installer_state_unittest.cc:611:
PathService::Get(base::DIR_PROGRAM_FILESX86, &temp);
Can you instead use ScopedPathOverride to explicitly override the PathService
return values with the current ones for the test?

Not sure I like relying on PathService caching which seems more like an
implementation detail than a contract.

[grt (UTC plus 2), 2014-08-29 18:58:31]

PTAL, thanks.

https://codereview.chromium.org/521703002/diff/20001/chrome/installer/util/in...
File chrome/installer/util/installer_state_unittest.cc (right):

https://codereview.chromium.org/521703002/diff/20001/chrome/installer/util/in...
chrome/installer/util/installer_state_unittest.cc:611:
PathService::Get(base::DIR_PROGRAM_FILESX86, &temp);
On 2014/08/29 18:49:20, robertshield wrote:
> Can you instead use ScopedPathOverride to explicitly override the PathService
> return values with the current ones for the test?
> 
> Not sure I like relying on PathService caching which seems more like an
> implementation detail than a contract.

Done.

[robertshield, 2014-08-29 19:08:41]

https://codereview.chromium.org/521703002/diff/40001/chrome/installer/util/in...
File chrome/installer/util/installer_state_unittest.cc (right):

https://codereview.chromium.org/521703002/diff/40001/chrome/installer/util/in...
chrome/installer/util/installer_state_unittest.cc:611:
base::FilePath(FILE_PATH_LITERAL("C:\\Program Files (x86)")));
Should this path not contain the (x86) bit?

Also, to be stricter could the override value be pulled from PathService here
before the reg override stuff kicks in?

[grt (UTC plus 2), 2014-08-29 19:13:07]

PTAL again!

https://codereview.chromium.org/521703002/diff/40001/chrome/installer/util/in...
File chrome/installer/util/installer_state_unittest.cc (right):

https://codereview.chromium.org/521703002/diff/40001/chrome/installer/util/in...
chrome/installer/util/installer_state_unittest.cc:611:
base::FilePath(FILE_PATH_LITERAL("C:\\Program Files (x86)")));
On 2014/08/29 19:08:40, robertshield wrote:
> Should this path not contain the (x86) bit?
> 
> Also, to be stricter could the override value be pulled from PathService here
> before the reg override stuff kicks in?

Ultimately, it doesn't matter what the strings are since they're never used to
access the disk, but I can see a pure correctness argument.

[robertshield, 2014-08-29 20:17:37]

lgtm

[grt (UTC plus 2), 2014-08-29 20:17:57]

The CQ bit was checked by grt@chromium.org

[commit-bot: I haz the power, 2014-08-30 00:21:07]

Committed patchset #4 (id:80001) as 6d6ac988d6a7a311876f3b0d0257086ac58c8181

[ckocagil, 2014-08-31 04:27:50]

A revert of this CL (patchset #4 id:80001) has been created in
https://codereview.chromium.org/520863005/ by ckocagil@chromium.org.

The reason for reverting is: Breaks InstallerStateTest.InitializeTwice on XP:
http://build.chromium.org/p/chromium.win/builders/XP%20Tests%20%281%29/builds....

[grt (UTC plus 2), 2014-08-31 23:39:34]

The CQ bit was checked by grt@chromium.org

[commit-bot: I haz the power, 2014-08-31 23:40:25]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/grt@chromium.org/521703002/100001

[commit-bot: I haz the power, 2014-09-01 00:35:34]

Committed patchset #5 (id:100001) as 190dbe2800433984ccdf9cbcdc7b0d9edaa72905

[commit-bot: I haz the power, 2014-09-10 03:12:15]

Patchset 5 (id:??) landed as
https://crrev.com/a12d6132de967aef5c3203bc21cb926a08bfc299
Cr-Commit-Position: refs/heads/master@{#292731}

[commit-bot: I haz the power, 2014-09-10 03:15:13]

Patchset 5 (id:??) landed as
https://crrev.com/4b1d16a9cbc61817d584c16fe2b3cc79dc0febc2
Cr-Commit-Position: refs/heads/master@{#292822}