Add UTF16ToUTF8AndAdjustOffset() to base/.

Add UTF16ToUTF8AndAdjustOffset() to base/.

Currently, only UTF8ToUTF16 version is implemented. The converse version will
be required for an upcoming patch for supporting input methods in PPAPI
(crbug.com/59425), where WebKit strings (UTF16) with segmentation information
must be converted to PPAPI strings (UTF8).

BUG=
TEST=base_unittests

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=100079

[kinaba, 2011-09-06 04:38:57]

Brett, could you take a look? The functions are just doing the reverse of
existing UTF8ToUTF16AndAdjustOffset, and the implementation is mostly shared.
Thanks.

[brettw, 2011-09-06 18:19:17]

http://codereview.chromium.org/7828092/diff/1/base/utf_offset_string_conversi...
File base/utf_offset_string_conversions.h (right):

http://codereview.chromium.org/7828092/diff/1/base/utf_offset_string_conversi...
base/utf_offset_string_conversions.h:42: BASE_EXPORT bool
UTF16ToUTF8AndAdjustOffset(const char16* src,
I'd not bother with these first two variants and just use the StringPiece ones.

[kinaba, 2011-09-06 23:55:35]

http://codereview.chromium.org/7828092/diff/1/base/utf_offset_string_conversi...
File base/utf_offset_string_conversions.h (right):

http://codereview.chromium.org/7828092/diff/1/base/utf_offset_string_conversi...
base/utf_offset_string_conversions.h:42: BASE_EXPORT bool
UTF16ToUTF8AndAdjustOffset(const char16* src,
On 2011/09/06 18:19:18, brettw wrote:
> I'd not bother with these first two variants and just use the StringPiece
ones.

Done.

[brettw, 2011-09-07 19:28:06]

Thanks just some minor nit.

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
File base/utf_offset_string_conversions.cc (right):

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.cc:109: std::string result;
The 8->16 version above just calls the vector version of the function, which
saves some code. Maybe you should do the same thing here/

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.cc:125: 
Only one blank line needed.

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
File base/utf_offset_string_conversions.h (right):

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.h:48: 
Only one blank line here.

[kinaba, 2011-09-07 19:54:47]

Thank you for comments. Updated.

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
File base/utf_offset_string_conversions.cc (right):

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.cc:109: std::string result;
On 2011/09/07 19:28:07, brettw wrote:
> The 8->16 version above just calls the vector version of the function, which
> saves some code. Maybe you should do the same thing here/

Done.

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.cc:125: 
On 2011/09/07 19:28:07, brettw wrote:
> Only one blank line needed.

Done.

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
File base/utf_offset_string_conversions.h (right):

http://codereview.chromium.org/7828092/diff/4001/base/utf_offset_string_conve...
base/utf_offset_string_conversions.h:48: 
On 2011/09/07 19:28:07, brettw wrote:
> Only one blank line here.

Done.

[brettw, 2011-09-07 20:07:42]

LGTM