Fix: debug.logging not work in retry provider

[Lin-jun-xiang]

When using RetryProvider, we want to identify the specific Provider being used (according to #1094 requirements).

According to the source code, I noticed that g4f.logging is likely intended for this purpose, facilitating better debugging for developers (please correct me if I'm wrong).

However, the logging attribute in retry_provider.py does not change along with g4f.logging = True. (always be False)

gpt4free/g4f/Provider/retry_provider.py

Lines 43 to 44 in cb3677a

	if logging:
	print(f"Using {provider.__name__} provider")

Therefore, in this case, I have defined logging as a class variable in RetryProvider.

Example

import g4f

g4f.logging = True

print(
    g4f.ChatCompletion.create(
            model='gpt-3.5-turbo',
            messages=[{"role": "user", "content": "hello"}],
    )
)

Results

Using RetryProvider provider
Using ChatgptAi provider
Using AItianhuSpace provider
AItianhuSpace: RuntimeError: g4f.provider.AItianhuSpace requires cookies
Using GptGo provider
Hello! How can I assist you today?

Noted: I'm not sure if this modification is the optimal solution. If there is a better way, I welcome assistance.

[xtekky]

Yes, I noticed that, you found a great solution !!

[nitsik]

Hello. I always see "Using RetryProvider provider" when I use "g4f.logging = True". I not see "Using ChatgptAi provider" and others. Why? What its means? Is RetryProvider it single rpovider? Why I not see other providers? Thank you.

[Lin-jun-xiang]

@nitsik

Use g4f.debug.logging=True instead of g4f.logging=True.

[nitsik]

@nitsik

Use g4f.debug.logging=True instead of g4f.logging=True.

When I use "g4f.debug.logging=True" I not see, what kind of provider I use.

[Lin-jun-xiang]

@nitsik

Make sure you have used new version g4f: pip install -U g4f

If you still not see the provider, could you show your code for me ? Thanks to u!

The following is my test:

import g4f

g4f.debug.logging = True

print(
    g4f.ChatCompletion.create(
            model='gpt-3.5-turbo',
            messages=[{"role": "user", "content": "hello"}],
            # ignored=["Ylokh", "GptGo", "AItianhu", "Aibn", "Myshell", "FreeGpt"],
        )
)

Using RetryProvider provider
Using ChatBase provider
Hello! How can I assist you today?

[nitsik]

I am use docker. How can I update? How to view version g4f?

@nitsik

Make sure you have used new version g4f: pip install -U g4f

If you still not see the provider, could you show your code for me ? Thanks to u!

The following is my test:

import g4f

g4f.debug.logging = True

print(
    g4f.ChatCompletion.create(
            model='gpt-3.5-turbo',
            messages=[{"role": "user", "content": "hello"}],
            # ignored=["Ylokh", "GptGo", "AItianhu", "Aibn", "Myshell", "FreeGpt"],
        )
)

Using RetryProvider provider
Using ChatBase provider
Hello! How can I assist you today?