Backfill unit tests for SkRecord

Backfill unit tests for SkRecord

BUG=skia:2378
Committed: http://code.google.com/p/skia/source/detail?r=14455

[f(malita), 2014-04-30 12:58:18]

lgtm

https://codereview.chromium.org/251133008/diff/1/tests/RecordOptsTest.cpp
File tests/RecordOptsTest.cpp (right):

https://codereview.chromium.org/251133008/diff/1/tests/RecordOptsTest.cpp#new...
tests/RecordOptsTest.cpp:35: SkASSERT(ptr != NULL);
Should we use REPORTER_ASSERT instead?

[mtklein, 2014-04-30 13:03:15]

lgtm

https://codereview.chromium.org/251133008/diff/1/tests/RecordOptsTest.cpp
File tests/RecordOptsTest.cpp (right):

https://codereview.chromium.org/251133008/diff/1/tests/RecordOptsTest.cpp#new...
tests/RecordOptsTest.cpp:35: SkASSERT(ptr != NULL);
On 2014/04/30 12:58:18, Florin Malita wrote:
> Should we use REPORTER_ASSERT instead?
Ah, yeah, good question.  I started with that, but it was hard to debug. 
REPORTER_ASSERT just gives you "Failed: ptr != NULL", without any sort of
provenance.  You end up putting in an SkASSERT(ptr != NULL) anyway so that you
can break gdb and see who called assert_type.

We could add REPORTER_ASSERT in addition, but since it doesn't bail out early,
if the assert fails we'll still just crash (here in Debug, dereferencing NULL
later in Release).

[commit-bot: I haz the power, 2014-04-30 13:03:31]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/mtklein@chromium.org/251133008/1

[commit-bot: I haz the power, 2014-04-30 13:18:15]

Change committed as 14455