# Dify API Python
[English](README.md) | [中文](README_CN.md)
Dify API Python is a Python implementation for interacting with Dify's API. It provides an easy-to-use interface for making API calls, with support for both streaming and non-streaming responses. The library also includes a method to merge stream responses, making it convenient to use in scenarios where streaming is not well supported.
## Installation
Install the package using pip:
```bash
pip install dify-api-python
```
## Usage
### Importing the Client
```python
from dify_api_python import DifyClient
```
### Initializing the Client
You can initialize the `DifyClient` in several ways:
1. Using parameters:
```python
client = DifyClient(api_key='your_api_key', base_url='https://api.dify.ai/v1')
```
2. Using a default configuration file:
```python
client = DifyClient()
```
3. Using a custom configuration file:
```python
client = DifyClient(config_path='/path/to/your/config.ini')
```
The configuration file should be in INI format and include the following:
```ini
[DEFAULT]
API_KEY = your_api_key
BASE_URL = https://api.dify.ai/v1
```
Note: If you provide `api_key` or `base_url` as parameters, they will override any values in the configuration file.
### Making API Calls
#### Chat Completion
For a chat completion request:
```python
response = client.chat_completion(
query="Hello, how are you?",
user="user123",
inputs={},
stream=False
)
print(response)
```
For a streaming chat completion:
```python
for chunk in client.chat_completion(
query="Tell me a story",
user="user123",
inputs={},
stream=True
):
print(chunk)
```
#### Text Completion
For a text completion request:
```python
response = client.text_completion(
prompt="Once upon a time",
user_id="user123",
inputs={},
stream=False
)
print(response)
```
For a streaming text completion:
```python
for chunk in client.text_completion(
prompt="Write a poem about",
user_id="user123",
inputs={},
stream=True
):
print(chunk)
```
#### Combined Chat Completion
For a combined chat completion (merges streaming response):
```python
response = client.chat_completion_combined(
query="Explain quantum computing",
user="user123",
inputs={}
)
print(response)
```
### Example Usage
Here's a complete example of how to use the Dify API Python client:
```python
from dify_api_python import DifyClient
# Initialize DifyClient
client = DifyClient(api_key='your_api_key', base_url='https://api.dify.ai/v1')
def chat_with_dify(query, user_id="user123", inputs={}):
"""
Chat with Dify API
:param query: User's question or input
:param user_id: User ID, default is "user123"
:param inputs: Additional input parameters, default is an empty dictionary
:return: API response
"""
try:
response = client.chat_completion_combined(
query=query,
user=user_id,
inputs=inputs
)
return response
except Exception as e:
print(f"An error occurred: {str(e)}")
return None
# Example usage
if __name__ == "__main__":
user_query = "Please explain the basic principles of quantum computing"
result = chat_with_dify(user_query)
if result:
print("Dify's answer:")
print(result)
else:
print("Failed to get an answer")
```
## Key Features
- Flexible client initialization (via parameters or config file)
- Support for both streaming and non-streaming API calls
- Method to merge streaming responses
- Easy-to-use interface for chat and text completions
## Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
## License
This project is licensed under the Apache License 2.0. See the [LICENSE](LICENSE) file for details.
## Support
If you encounter any issues or have questions, please open an issue on the GitHub repository.
Raw data
{
"_id": null,
"home_page": "https://github.com/zlz3907/dify-api-python",
"name": "dify-api-python",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": null,
"author": "zlz3907",
"author_email": "zlz3907@gmail.com",
"download_url": null,
"platform": null,
"description": "# Dify API Python\n\n[English](README.md) | [\u4e2d\u6587](README_CN.md)\n\nDify API Python is a Python implementation for interacting with Dify's API. It provides an easy-to-use interface for making API calls, with support for both streaming and non-streaming responses. The library also includes a method to merge stream responses, making it convenient to use in scenarios where streaming is not well supported.\n\n## Installation\n\nInstall the package using pip:\n\n```bash\npip install dify-api-python\n```\n\n## Usage\n\n### Importing the Client\n\n```python\nfrom dify_api_python import DifyClient\n```\n\n### Initializing the Client\n\nYou can initialize the `DifyClient` in several ways:\n\n1. Using parameters:\n ```python\n client = DifyClient(api_key='your_api_key', base_url='https://api.dify.ai/v1')\n ```\n\n2. Using a default configuration file:\n ```python\n client = DifyClient()\n ```\n\n3. Using a custom configuration file:\n ```python\n client = DifyClient(config_path='/path/to/your/config.ini')\n ```\n\nThe configuration file should be in INI format and include the following:\n\n```ini\n[DEFAULT]\nAPI_KEY = your_api_key\nBASE_URL = https://api.dify.ai/v1\n```\n\nNote: If you provide `api_key` or `base_url` as parameters, they will override any values in the configuration file.\n\n### Making API Calls\n\n#### Chat Completion\n\nFor a chat completion request:\n\n```python\nresponse = client.chat_completion(\n query=\"Hello, how are you?\",\n user=\"user123\",\n inputs={},\n stream=False\n)\nprint(response)\n```\n\nFor a streaming chat completion:\n\n```python\nfor chunk in client.chat_completion(\n query=\"Tell me a story\",\n user=\"user123\",\n inputs={},\n stream=True\n):\n print(chunk)\n```\n\n#### Text Completion\n\nFor a text completion request:\n\n```python\nresponse = client.text_completion(\n prompt=\"Once upon a time\",\n user_id=\"user123\",\n inputs={},\n stream=False\n)\nprint(response)\n```\n\nFor a streaming text completion:\n\n```python\nfor chunk in client.text_completion(\n prompt=\"Write a poem about\",\n user_id=\"user123\",\n inputs={},\n stream=True\n):\n print(chunk)\n```\n\n#### Combined Chat Completion\n\nFor a combined chat completion (merges streaming response):\n\n```python\nresponse = client.chat_completion_combined(\n query=\"Explain quantum computing\",\n user=\"user123\",\n inputs={}\n)\nprint(response)\n```\n\n### Example Usage\n\nHere's a complete example of how to use the Dify API Python client:\n\n```python\nfrom dify_api_python import DifyClient\n\n# Initialize DifyClient\nclient = DifyClient(api_key='your_api_key', base_url='https://api.dify.ai/v1')\n\ndef chat_with_dify(query, user_id=\"user123\", inputs={}):\n \"\"\"\n Chat with Dify API\n \n :param query: User's question or input\n :param user_id: User ID, default is \"user123\"\n :param inputs: Additional input parameters, default is an empty dictionary\n :return: API response\n \"\"\"\n try:\n response = client.chat_completion_combined(\n query=query,\n user=user_id,\n inputs=inputs\n )\n return response\n except Exception as e:\n print(f\"An error occurred: {str(e)}\")\n return None\n\n# Example usage\nif __name__ == \"__main__\":\n user_query = \"Please explain the basic principles of quantum computing\"\n result = chat_with_dify(user_query)\n \n if result:\n print(\"Dify's answer:\")\n print(result)\n else:\n print(\"Failed to get an answer\")\n```\n\n## Key Features\n\n- Flexible client initialization (via parameters or config file)\n- Support for both streaming and non-streaming API calls\n- Method to merge streaming responses\n- Easy-to-use interface for chat and text completions\n\n## Contributing\n\nContributions are welcome! Please feel free to submit a Pull Request.\n\n## License\n\nThis project is licensed under the Apache License 2.0. See the [LICENSE](LICENSE) file for details.\n\n## Support\n\nIf you encounter any issues or have questions, please open an issue on the GitHub repository.\n\n",
"bugtrack_url": null,
"license": "Apache License 2.0",
"summary": "A Python client for the Dify API",
"version": "0.1.4",
"project_urls": {
"Homepage": "https://github.com/zlz3907/dify-api-python"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "0120b3fa1def2a4466f28488e82f70af30c3a106abd43b8caa2bb41eff55e170",
"md5": "0b0634754ab60f0b33a82d88466b9751",
"sha256": "a638122119ff3b77fd61ec0b5a675f0c336c1a009af886e02d6c3270a303adbe"
},
"downloads": -1,
"filename": "dify_api_python-0.1.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0b0634754ab60f0b33a82d88466b9751",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 9488,
"upload_time": "2024-10-14T10:02:16",
"upload_time_iso_8601": "2024-10-14T10:02:16.083787Z",
"url": "https://files.pythonhosted.org/packages/01/20/b3fa1def2a4466f28488e82f70af30c3a106abd43b8caa2bb41eff55e170/dify_api_python-0.1.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-10-14 10:02:16",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "zlz3907",
"github_project": "dify-api-python",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "dify-api-python"
}
JSON
