Represent BINARY properties using std::string instead of BinaryValue,

Represent BINARY properties using std::string instead of BinaryValue,
for easier use by C++.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=147591

[Jeffrey Yasskin, 2012-07-13 00:13:33]

Brett, please review the small change to let BinaryValue take a StringPiece. The
extra code if you don't like that addition is minimal.

[brettw, 2012-07-13 16:46:46]

https://chromiumcodereview.appspot.com/10700194/diff/1/base/values.h
File base/values.h (right):

https://chromiumcodereview.appspot.com/10700194/diff/1/base/values.h#newcode33
base/values.h:33: #include "base/string_piece.h"
I don't actually know the answer but I wonder if we made the right tradeoff
here. base/value is used in a lot of places and string_piece.h isn't a simple
header file. Given that this is generated code and the the end result will be
the same, do you think it would be better to just have the generator emit the
(data(), size()) calls?

[Jeffrey Yasskin, 2012-07-15 04:07:23]

Thanks for looking at the base/values.h change, Brett. Now that it's reverted,
you don't need to review the change anymore.

https://chromiumcodereview.appspot.com/10700194/diff/1/base/values.h
File base/values.h (right):

https://chromiumcodereview.appspot.com/10700194/diff/1/base/values.h#newcode33
base/values.h:33: #include "base/string_piece.h"
On 2012/07/13 16:46:46, brettw wrote:
> I don't actually know the answer but I wonder if we made the right tradeoff
> here. base/value is used in a lot of places and string_piece.h isn't a simple
> header file. Given that this is generated code and the the end result will be
> the same, do you think it would be better to just have the generator emit the
> (data(), size()) calls?

I tend to think that StringPiece should be used everywhere we pass const
character data around, but it's not an important part of this particular change.
Reverted.

[Jeffrey Yasskin, 2012-07-19 22:09:17]

I'm not sure who's best to review json_schema_compiler changes, but "ping".

[Matt Perry, 2012-07-19 22:33:48]

LGTM. +cc kalman since I think he knows about json_schema_compiler

[not at google - send to devlin, 2012-07-20 00:12:45]

lgtm

[commit-bot: I haz the power, 2012-07-20 00:16:08]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jyasskin@chromium.org/10700194/4001

[commit-bot: I haz the power, 2012-07-20 01:27:36]

Change committed as 147591