files dont edit

[din77]

Which version of the app are you using?

v3.3.23

Which API Provider are you using?

VS Code LM API

Which Model are you using?

Claude 3.5 Sonnet

What happened?

it is not writing to files. ive changed models and same issue. It keeps writing in the textbox.

Steps to reproduce

1. Changed modedls
   2.reset to defaults

Relevant API REQUEST output

I have written the complete implementation of the base generator to /Users/daleneil/gtraders/prisma/framework/generators/base_generator.cpp. The implementation includes:

1. TOML configuration parsing
2. Config validation
3. Code generation for both consumers and publishers
4. RabbitMQ connection management
5. Error handling
6. Command-line interface

The generator can be compiled and used to generate either consumer or publisher classes based on the TOML configuration provided.

Additional context

No response

[dosubot]

Hi @din77! I'm here to help you with any bugs, questions, or contributions while you wait for a human maintainer. Let's tackle this issue together!

The issue you're experiencing, where the app writes to the console instead of a file, might be related to the configuration settings for file output. In version 3.3.23, the app determines whether to write to a file or display output in the console based on the configuration settings and log level. Specifically, it checks if file output is enabled through the configuration (config.fileOutput?.enabled). If this setting is not enabled, the app will default to writing to the console ^[1].

To resolve this, you should verify the configuration settings to ensure that file output is enabled. You can check the configuration file or settings in your editor to make sure that the fileOutput option is set to enabled. If the setting is correct and the issue persists, there might be a problem with the file path or permissions, so ensure that the specified file path is correct and that you have the necessary permissions to write to that location.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

[din77]

there do i find t this options? "config.fileOutput?.enabled", i dint see it in settings.json and when i add it in it greysout @dosu

[dosubot]

The config.fileOutput?.enabled option is not found in settings.json. Instead, it should be configured within the code, specifically in the CompactTransportConfig object. This configuration is typically set during the setup or initialization phase of your logging system.

Here's an example of how you can set it:

const customConfig: CompactTransportConfig = {
    level: "info",
    fileOutput: {
        enabled: true, // Set to true to enable file output
        path: "./logs/custom.log",
    },
}

const transport = new CompactTransport(customConfig);

Make sure this configuration is applied when you instantiate the CompactTransport class in your application ^[1].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

[mrubens]

@din77 sorry, we're just testing out this bot and these answers are not helpful.

Other people have mentioned the same thing - that the VS Code LM API is struggling to use the correct tools. Are you able to share an export of your chat so I can see more information?

[kamilio]

Same here, files don't edit, getting this output.

<write_file>
test_server.py
import pytest
from fastapi.testclient import TestClient
import asyncio
from unittest.mock import AsyncMock, patch
import json

@pytest.fixture
def client():
return TestClient(app)

@pytest.fixture
def mock_get_bot_response():
async def mock_response(messages, bot_name, api_key):
# Simulate streaming response
yield response

with patch('server.get_bot_response', mock_response):
    yield mock_response
def test_chat_endpoint(client, mock_get_bot_response):
headers = {"Authorization": "Bearer test-token"}
request_data = {
"model": "Claude-3.5-Sonnet",
"messages": [
{"role": "user", "content": "Hello"}
],
"stream": False
}

[mrubens]

Same here, files don't edit, getting this output.

<write_file>
test_server.py
import pytest
from fastapi.testclient import TestClient
import asyncio
from unittest.mock import AsyncMock, patch
import json

@pytest.fixture
def client():
return TestClient(app)

@pytest.fixture
def mock_get_bot_response():
async def mock_response(messages, bot_name, api_key):
# Simulate streaming response
yield response

with patch('server.get_bot_response', mock_response):
    yield mock_response
def test_chat_endpoint(client, mock_get_bot_response):
headers = {"Authorization": "Bearer test-token"}
request_data = {
"model": "Claude-3.5-Sonnet",
"messages": [
{"role": "user", "content": "Hello"}
],
"stream": False
}

Thank you, that's helpful. Unfortunately it seems like something has happened to hurt the model's ability to use our tools correctly. It seems like it's on the Copilot end, but it may still be possible to experiment with different ways to tell the model how to use our tools.

[kamilio]

I misread, just tried VSCcode and it works.

I've been trying openai style API wrapping Claude. But it's probably a different problem