rebaseline.py self-test: validate that unknown fields are maintained

rebaseline.py self-test: validate that unknown fields are maintained
BUG=skia:1569
R=bsalomon@google.com

Committed: https://code.google.com/p/skia/source/detail?r=11008

[epoger, 2013-08-28 17:42:55]

This might be a good opportunity to discuss self-test do's and don'ts, as per
https://code.google.com/p/skia/issues/detail?id=1568 ('gm self-tests considered
harmful').

I certainly would like to know as soon as possible, if rebaseline.py was changed
such that unknown fields were thrown away during the rebaseline process...

[epoger, 2013-08-29 16:33:01]

Ping...

On 2013/08/28 17:42:55, epoger wrote:
> This might be a good opportunity to discuss self-test do's and don'ts, as per
> https://code.google.com/p/skia/issues/detail?id=1568 ('gm self-tests
considered
> harmful').

Actually, https://codereview.chromium.org/23725003 ('gm self-tests: stop
comparing stdout/stderr/command_line, just look at JSON summary') is a better
opportunity to discuss that.  This one, let's just commit. :-)

[bsalomon, 2013-08-29 17:49:59]

On 2013/08/29 16:33:01, epoger wrote:
> Ping...
> 
> On 2013/08/28 17:42:55, epoger wrote:
> > This might be a good opportunity to discuss self-test do's and don'ts, as
per
> > https://code.google.com/p/skia/issues/detail?id=1568 ('gm self-tests
> considered
> > harmful').
> 
> Actually, https://codereview.chromium.org/23725003 ('gm self-tests: stop
> comparing stdout/stderr/command_line, just look at JSON summary') is a better
> opportunity to discuss that.  This one, let's just commit. :-)

lgtm

[epoger, 2013-08-29 18:04:24]

Committed patchset #1 manually as r11008 (presubmit successful).