Support UTF-8 characters (e.g. Hebrew)

[orenagiv]

Due to a bug in the openapi_spec package - if the OpenAI response include UTF-8 characters of languages such as Hebrew - the response is not properly decoded and the result is "gibberish".

This affects the openai_dart package.

I've opened a bug in openapi_spec:
tazatechnology/openapi_spec#43

I hope they'll fix it soon.

[walsha2]

@davidmigloz this has been resolved in tazatechnology/openapi_spec#44 and published to openapi_spec (v0.7.8).

[davidmigloz]

Thanks @orenagiv for reporting and @walsha2 for the quick fix!

The fix is now deployed in:

• openai_dart: 0.1.1+2
• langchain_openai 0.1.2+1