Require length in sk_wchar_to_string.

Require length in sk_wchar_to_string.

This information is already available at all call sites and allows
the call to WideCharToMultiByte to not overwrite the '\0' in the
writable_str() which isn't really writable.

BUG=skia:1989
Committed: https://skia.googlesource.com/skia/+/5e7b4f967b37cc303a20b87617446605f26f1867

[bungeman-skia, 2014-08-25 15:37:28]

Perhaps we should also document that in general SkString::writable_string() only
gives write access to SkString::size() bytes.

[reed1, 2014-08-25 17:04:18]

lgtm w/ documentation request

https://codereview.chromium.org/500113002/diff/1/src/utils/win/SkDWrite.h
File src/utils/win/SkDWrite.h (right):

https://codereview.chromium.org/500113002/diff/1/src/utils/win/SkDWrite.h#new...
src/utils/win/SkDWrite.h:32: HRESULT sk_wchar_to_skstring(WCHAR* name, int
nameLen, SkString* skname);
Can we document that nameLen is the number-of-WCHARs, and not a byte count.

[bungeman-skia, 2014-08-25 17:08:52]

https://codereview.chromium.org/500113002/diff/1/src/utils/win/SkDWrite.h
File src/utils/win/SkDWrite.h (right):

https://codereview.chromium.org/500113002/diff/1/src/utils/win/SkDWrite.h#new...
src/utils/win/SkDWrite.h:32: HRESULT sk_wchar_to_skstring(WCHAR* name, int
nameLen, SkString* skname);
On 2014/08/25 17:04:18, reed1 wrote:
> Can we document that nameLen is the number-of-WCHARs, and not a byte count.

Done.

[bungeman-skia, 2014-08-25 17:08:56]

The CQ bit was checked by bungeman@google.com

[commit-bot: I haz the power, 2014-08-25 17:09:26]

CQ is trying da patch. Follow status at
 https://skia-tree-status.appspot.com/cq/bungeman@google.com/500113002/20001

[commit-bot: I haz the power, 2014-08-25 17:16:05]

Committed patchset #2 (20001) as 5e7b4f967b37cc303a20b87617446605f26f1867