Fix deps2git cache bootstrapping and make it more verbose

Fix deps2git cache bootstrapping and make it more verbose

bootstrap defaults to False in populate(), which disables the Google Storage bootstrap codepath.  We actually really want this to be on.

Also make deps2git print more, to make these errors more obvious.

BUG=
TEST=ran locally

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=292377

[Ryan Tseng, 2014-10-08 20:43:56]

hinoka@google.com changed reviewers:
+ szager@google.com

[Ryan Tseng, 2014-10-08 20:51:23]

hinoka@google.com changed reviewers:
+ hinoka@google.com

[szager1, 2014-10-08 20:58:09]

szager@chromium.org changed reviewers:
+ szager@chromium.org

[szager1, 2014-10-08 20:58:09]

https://codereview.chromium.org/641693004/diff/20001/git_tools.py
File git_tools.py (right):

https://codereview.chromium.org/641693004/diff/20001/git_tools.py#newcode158
git_tools.py:158: print_fn = print_fn or (lambda *args: None)
Not sure about this; it will bury errors.

[Ryan Tseng, 2014-10-08 21:00:35]

https://codereview.chromium.org/641693004/diff/20001/git_tools.py
File git_tools.py (right):

https://codereview.chromium.org/641693004/diff/20001/git_tools.py#newcode158
git_tools.py:158: print_fn = print_fn or (lambda *args: None)
On 2014/10/08 20:58:09, szager1 wrote:
> Not sure about this; it will bury errors.

Should I make print_fn required instead?

ie. assert hasattr(print_fn, '__call__')

[Ryan Tseng, 2014-10-08 21:01:24]

One more note: local testing reveals that the git cache printout is working, but
its not bootstrapping from google storage.  Investigating.

[Ryan Tseng, 2014-10-08 23:35:26]

Nevermind, it worked.  My credentials just wern't set up correctly.

[szager1, 2014-10-08 23:38:46]

https://codereview.chromium.org/641693004/diff/20001/git_tools.py
File git_tools.py (right):

https://codereview.chromium.org/641693004/diff/20001/git_tools.py#newcode158
git_tools.py:158: print_fn = print_fn or (lambda *args: None)
On 2014/10/08 21:00:35, Ryan Tseng wrote:
> On 2014/10/08 20:58:09, szager1 wrote:
> > Not sure about this; it will bury errors.
> 
> Should I make print_fn required instead?
> 
> ie. assert hasattr(print_fn, '__call__')

Just pass print_fn through; if it's None, the git_cache code will use an
appropriate default.

[Ryan Tseng, 2014-10-08 23:41:59]

https://codereview.chromium.org/641693004/diff/20001/git_tools.py
File git_tools.py (right):

https://codereview.chromium.org/641693004/diff/20001/git_tools.py#newcode158
git_tools.py:158: print_fn = print_fn or (lambda *args: None)
On 2014/10/08 23:38:46, szager1 wrote:
> On 2014/10/08 21:00:35, Ryan Tseng wrote:
> > On 2014/10/08 20:58:09, szager1 wrote:
> > > Not sure about this; it will bury errors.
> > 
> > Should I make print_fn required instead?
> > 
> > ie. assert hasattr(print_fn, '__call__')
> 
> Just pass print_fn through; if it's None, the git_cache code will use an
> appropriate default.

Done.

[szager1, 2014-10-08 23:55:40]

lgtm

[Ryan Tseng, 2014-10-08 23:59:49]

The CQ bit was checked by hinoka@google.com

[commit-bot: I haz the power, 2014-10-09 00:00:40]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/641693004/40001

[commit-bot: I haz the power, 2014-10-09 00:00:59]

Committed patchset #3 (id:40001) as 292377