Added example UTF-8 conversion functions

[ocrow]

Added push_utf8_string() and get_utf8_string() functions that implement the suggested encoding/decoding of surrogate pairs when passing UTF8 strings in to or out of duktape.

I'm not sure if this is the right place to put these functions. Would they be better as C source somewhere in the duktape distribution? The bit-twiddling required was involved enough that it seems worth sharing this code.

[svaarala]

I'll comment on the content itself separately, but regarding placement: having this in the Wiki is fine and the page you chose seems best for the purpose.

If the code were in Duktape repo, there'd be two logical places:

• examples/ if the code is intended to be just an example and not maintained and/or production quality
• extras/ if the code is maintained and production quality

There's no crystal clear policy now for what kind of code should go into Duktape repo/distributable. It makes sense when the code is somehow Duktape version relevant and/or very close to the core functionality (e.g. the module loader implementations) and/or encountered very commonly in user code (e.g. allocators). String encoding/decoding stuff fits the latter criteria I think and there's already a codepage conversion utility in the examples/ directory.

Finally, you could put the code in a separate repo and the Wiki could link there.

Code which is inline in the Wiki articles should naturally explain or illustrate the text surrounding it. For code which is useful but doesn't necessarily fit that purpose (e.g. it is long or very technical and not illustrative w.r.t. the surrounding text) linking may be a better idea.

[ocrow]

Putting it in examples/ seems perhaps most appropriate. I'll put together some test code to demonstrate the functionality and resubmit.