how can we deal with multi-lingual string? (c++)

[alcohol]

without std::wstring support, how can we deal with strings consists
of Ascii, Chinese,Japanese, Korean characters?

[Evan Jones]

On Dec 25, 2010, at 21:10 , alcohol wrote:
> without std::wstring support, how can we deal with strings consists
> of Ascii, Chinese,Japanese, Korean characters?

It is expected that you put UTF-8 encoded characters into protocol
buffer strings. Alternatively, you can use the bytes type and use any
encoding you want, but then you'll need to handle the conversions,
whereas if you use Java, the protocol buffer API handles conversions
for you (to/from Java String, which uses Unicode). Hope that helps,

Evan

--
Evan Jones
http://evanjones.ca/

[Kenton Varda]

The dirty little secret of std::wstring is that it does not actually deal with non-ASCII characters on all platforms.  On some platforms wchar_t is 8-bit just like char!  You should avoid using wstring and wchar_t for this reason; define your own types that are exactly what you need.

For protocol buffers, we take the convention of using regular 8-bit chars but always using UTF-8 encoding.

On Sat, Dec 25, 2010 at 8:10 PM, alcohol <alco...@gmail.com> wrote:

without std::wstring support,  how can we deal with strings consists
of  Ascii, Chinese,Japanese, Korean characters?

--
You received this message because you are subscribed to the Google Groups "Protocol Buffers" group.
To post to this group, send email to prot...@googlegroups.com.
To unsubscribe from this group, send email to protobuf+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/protobuf?hl=en.

[Darko Miletic]

I'd say use UTF-8 for all strings and you are good to go.