| Name | unitycatalog-langchain JSON |
| Version |
0.1.1
JSON |
| download |
| home_page | None |
| Summary | Support for Unity Catalog functions as LangChain tools |
| upload_time | 2025-02-13 00:25:34 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.9 |
| license | Apache-2.0 |
| keywords |
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# 🦜🔗 Using Unity Catalog AI with Langchain
Integrate the Unity Catalog AI package with `Langchain` to allow seamless usage of UC functions as tools in agentic applications.
## Installation
### Client Library
To install the Unity Catalog function client SDK and the `LangChain` (and `LangGraph`) integration, simply install from PyPI:
```sh
pip install unitycatalog-langchain
```
If you are working with **Databricks Unity Catalog**, you can install the optional package:
```sh
pip install unitycatalog-langchain[databricks]
```
## Getting started
### Creating a Unity Catalog Client
To interact with your Unity Catalog server, initialize the `UnitycatalogFunctionClient` as shown below:
```python
import asyncio
from unitycatalog.ai.core.client import UnitycatalogFunctionClient
from unitycatalog.client import ApiClient, Configuration
# Configure the Unity Catalog API client
config = Configuration(
host="http://localhost:8080/api/2.1/unity-catalog" # Replace with your UC server URL
)
# Initialize the asynchronous ApiClient
api_client = ApiClient(configuration=config)
# Instantiate the UnitycatalogFunctionClient
uc_client = UnitycatalogFunctionClient(api_client=api_client)
# Example catalog and schema names
CATALOG = "my_catalog"
SCHEMA = "my_schema"
```
### Creating a Unity Catalog Function
You can create a UC function either by providing a Python callable or by submitting a `FunctionInfo` object. Below is an example (recommended) of using the `create_python_function` API that accepts a Python callable (function) as input.
To create a UC function from a Python function, define your function with appropriate type hints and a Google-style docstring:
```python
def add_numbers(a: float, b: float) -> float:
"""
Adds two numbers and returns the result.
Args:
a (float): First number.
b (float): Second number.
Returns:
float: The sum of the two numbers.
"""
return a + b
# Create the function within the Unity Catalog catalog and schema specified
function_info = uc_client.create_python_function(
func=add_numbers,
catalog=CATALOG,
schema=SCHEMA,
replace=False, # Set to True to overwrite if the function already exists
)
print(function_info)
```
### Databricks-managed Unity Catalog
To use Databricks-managed Unity Catalog with this package, follow the [instructions](https://docs.databricks.com/en/dev-tools/cli/authentication.html#authentication-for-the-databricks-cli) to authenticate to your workspace and ensure that your access token has workspace-level privilege for managing UC functions.
#### Client setup
Initialize a client for managing UC functions in a Databricks workspace, and set it as the global client.
```python
from unitycatalog.ai.core.base import set_uc_function_client
from unitycatalog.ai.core.databricks import DatabricksFunctionClient
client = DatabricksFunctionClient()
# sets the default uc function client
set_uc_function_client(client)
```
#### Create a function in UC
Create a python UDF in Unity Catalog with the client
```python
# replace with your own catalog and schema
CATALOG = "catalog"
SCHEMA = "schema"
func_name = f"{CATALOG}.{SCHEMA}.python_exec"
# define the function body in SQL
sql_body = f"""CREATE OR REPLACE FUNCTION {func_name}(code STRING COMMENT 'Python code to execute. Remember to print the final result to stdout.')
RETURNS STRING
LANGUAGE PYTHON
COMMENT 'Executes Python code and returns its stdout.'
AS $$
import sys
from io import StringIO
stdout = StringIO()
sys.stdout = stdout
exec(code)
return stdout.getvalue()
$$
"""
client.create_function(sql_function_body=sql_body)
```
Now the function is created and stored in the corresponding catalog and schema.
## Using the Function as a GenAI Tool
### Create a UCFunctionToolkit instance
[Langchain tools](https://python.langchain.com/v0.2/docs/concepts/#tools) are utilities designed to be called by a model, and UCFunctionToolkit provides the ability to use UC functions as tools that are recognized natively by LangChain.
```python
from unitycatalog.ai.langchain.toolkit import UCFunctionToolkit
# create a UCFunctionToolkit that includes the above UC function
toolkit = UCFunctionToolkit(function_names=[f"{CATALOG}.{SCHEMA}.python_exec"])
# fetch the tools stored in the toolkit
tools = toolkit.tools
python_exec_tool = tools[0]
# execute the tool directly
python_exec_tool.invoke({"code": "print(1)"})
```
### Use the tools in a Langchain Agent
Now we create an agent and use the tools.
```python
from langchain_community.chat_models.databricks import ChatDatabricks
from langchain.agents import AgentExecutor, create_tool_calling_agent
from langchain_core.prompts import ChatPromptTemplate
# Use Databricks foundation models
llm = ChatDatabricks(endpoint="databricks-meta-llama-3-1-70b-instruct")
prompt = ChatPromptTemplate.from_messages(
[
(
"system",
"You are a helpful assistant. Make sure to use tool for information.",
),
("placeholder", "{chat_history}"),
("human", "{input}"),
("placeholder", "{agent_scratchpad}"),
]
)
agent = create_tool_calling_agent(llm, tools, prompt)
# Create the agent executor
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
agent_executor.invoke({"input": "36939 * 8922.4"})
```
### Configurations for Databricks managed UC functions execution
We provide configurations for databricks client to control the function execution behaviors, check [function execution arguments section](../../README.md#function-execution-arguments-configuration).
Raw data
{
"_id": null,
"home_page": null,
"name": "unitycatalog-langchain",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": null,
"author": null,
"author_email": "Serena Ruan <serena.ruan@databricks.com>",
"download_url": "https://files.pythonhosted.org/packages/c1/40/87e3cfd5caf35d2cac01723662728316b6204393cb87fa96690c7d6f7768/unitycatalog_langchain-0.1.1.tar.gz",
"platform": null,
"description": "# \ud83e\udd9c\ud83d\udd17 Using Unity Catalog AI with Langchain\n\nIntegrate the Unity Catalog AI package with `Langchain` to allow seamless usage of UC functions as tools in agentic applications.\n\n## Installation\n\n### Client Library\n\nTo install the Unity Catalog function client SDK and the `LangChain` (and `LangGraph`) integration, simply install from PyPI:\n\n```sh\npip install unitycatalog-langchain\n```\n\nIf you are working with **Databricks Unity Catalog**, you can install the optional package:\n\n```sh\npip install unitycatalog-langchain[databricks]\n```\n\n## Getting started\n\n### Creating a Unity Catalog Client\n\nTo interact with your Unity Catalog server, initialize the `UnitycatalogFunctionClient` as shown below:\n\n```python\nimport asyncio\nfrom unitycatalog.ai.core.client import UnitycatalogFunctionClient\nfrom unitycatalog.client import ApiClient, Configuration\n\n# Configure the Unity Catalog API client\nconfig = Configuration(\n host=\"http://localhost:8080/api/2.1/unity-catalog\" # Replace with your UC server URL\n)\n\n# Initialize the asynchronous ApiClient\napi_client = ApiClient(configuration=config)\n\n# Instantiate the UnitycatalogFunctionClient\nuc_client = UnitycatalogFunctionClient(api_client=api_client)\n\n# Example catalog and schema names\nCATALOG = \"my_catalog\"\nSCHEMA = \"my_schema\"\n```\n\n### Creating a Unity Catalog Function\n\nYou can create a UC function either by providing a Python callable or by submitting a `FunctionInfo` object. Below is an example (recommended) of using the `create_python_function` API that accepts a Python callable (function) as input.\n\nTo create a UC function from a Python function, define your function with appropriate type hints and a Google-style docstring:\n\n```python\ndef add_numbers(a: float, b: float) -> float:\n \"\"\"\n Adds two numbers and returns the result.\n\n Args:\n a (float): First number.\n b (float): Second number.\n\n Returns:\n float: The sum of the two numbers.\n \"\"\"\n return a + b\n\n# Create the function within the Unity Catalog catalog and schema specified\nfunction_info = uc_client.create_python_function(\n func=add_numbers,\n catalog=CATALOG,\n schema=SCHEMA,\n replace=False, # Set to True to overwrite if the function already exists\n)\n\nprint(function_info)\n```\n\n### Databricks-managed Unity Catalog\n\nTo use Databricks-managed Unity Catalog with this package, follow the [instructions](https://docs.databricks.com/en/dev-tools/cli/authentication.html#authentication-for-the-databricks-cli) to authenticate to your workspace and ensure that your access token has workspace-level privilege for managing UC functions.\n\n#### Client setup\n\nInitialize a client for managing UC functions in a Databricks workspace, and set it as the global client.\n\n```python\nfrom unitycatalog.ai.core.base import set_uc_function_client\nfrom unitycatalog.ai.core.databricks import DatabricksFunctionClient\n\nclient = DatabricksFunctionClient()\n\n# sets the default uc function client\nset_uc_function_client(client)\n```\n\n#### Create a function in UC\n\nCreate a python UDF in Unity Catalog with the client\n\n```python\n# replace with your own catalog and schema\nCATALOG = \"catalog\"\nSCHEMA = \"schema\"\n\nfunc_name = f\"{CATALOG}.{SCHEMA}.python_exec\"\n# define the function body in SQL\nsql_body = f\"\"\"CREATE OR REPLACE FUNCTION {func_name}(code STRING COMMENT 'Python code to execute. Remember to print the final result to stdout.')\nRETURNS STRING\nLANGUAGE PYTHON\nCOMMENT 'Executes Python code and returns its stdout.'\nAS $$\n import sys\n from io import StringIO\n stdout = StringIO()\n sys.stdout = stdout\n exec(code)\n return stdout.getvalue()\n$$\n\"\"\"\n\nclient.create_function(sql_function_body=sql_body)\n```\n\nNow the function is created and stored in the corresponding catalog and schema.\n\n## Using the Function as a GenAI Tool\n\n### Create a UCFunctionToolkit instance\n\n[Langchain tools](https://python.langchain.com/v0.2/docs/concepts/#tools) are utilities designed to be called by a model, and UCFunctionToolkit provides the ability to use UC functions as tools that are recognized natively by LangChain.\n\n```python\nfrom unitycatalog.ai.langchain.toolkit import UCFunctionToolkit\n\n# create a UCFunctionToolkit that includes the above UC function\ntoolkit = UCFunctionToolkit(function_names=[f\"{CATALOG}.{SCHEMA}.python_exec\"])\n\n# fetch the tools stored in the toolkit\ntools = toolkit.tools\npython_exec_tool = tools[0]\n\n# execute the tool directly\npython_exec_tool.invoke({\"code\": \"print(1)\"})\n```\n\n### Use the tools in a Langchain Agent\n\nNow we create an agent and use the tools.\n\n```python\nfrom langchain_community.chat_models.databricks import ChatDatabricks\nfrom langchain.agents import AgentExecutor, create_tool_calling_agent\nfrom langchain_core.prompts import ChatPromptTemplate\n\n# Use Databricks foundation models\nllm = ChatDatabricks(endpoint=\"databricks-meta-llama-3-1-70b-instruct\")\nprompt = ChatPromptTemplate.from_messages(\n [\n (\n \"system\",\n \"You are a helpful assistant. Make sure to use tool for information.\",\n ),\n (\"placeholder\", \"{chat_history}\"),\n (\"human\", \"{input}\"),\n (\"placeholder\", \"{agent_scratchpad}\"),\n ]\n)\nagent = create_tool_calling_agent(llm, tools, prompt)\n\n# Create the agent executor\nagent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)\nagent_executor.invoke({\"input\": \"36939 * 8922.4\"})\n```\n\n### Configurations for Databricks managed UC functions execution\n\nWe provide configurations for databricks client to control the function execution behaviors, check [function execution arguments section](../../README.md#function-execution-arguments-configuration).\n",
"bugtrack_url": null,
"license": "Apache-2.0",
"summary": "Support for Unity Catalog functions as LangChain tools",
"version": "0.1.1",
"project_urls": null,
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "a2a035e490d49b9684abced53889a436f1205be9ab36ece273060f603f09689f",
"md5": "82eb360cf037bb40e56855243abe7e05",
"sha256": "fd01c56b9b71f99a9fa8ad89c2dacab9cd38a3ce0a6caf078faf879876b3ce9c"
},
"downloads": -1,
"filename": "unitycatalog_langchain-0.1.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "82eb360cf037bb40e56855243abe7e05",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 5093,
"upload_time": "2025-02-13T00:25:32",
"upload_time_iso_8601": "2025-02-13T00:25:32.532948Z",
"url": "https://files.pythonhosted.org/packages/a2/a0/35e490d49b9684abced53889a436f1205be9ab36ece273060f603f09689f/unitycatalog_langchain-0.1.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c14087e3cfd5caf35d2cac01723662728316b6204393cb87fa96690c7d6f7768",
"md5": "6778f17e18c58f6df17b1c06935a8c3e",
"sha256": "fe004ee07cfbb90f78be927941f400e319f44df9f591b5248f9d0eec81f0624b"
},
"downloads": -1,
"filename": "unitycatalog_langchain-0.1.1.tar.gz",
"has_sig": false,
"md5_digest": "6778f17e18c58f6df17b1c06935a8c3e",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 4751,
"upload_time": "2025-02-13T00:25:34",
"upload_time_iso_8601": "2025-02-13T00:25:34.470153Z",
"url": "https://files.pythonhosted.org/packages/c1/40/87e3cfd5caf35d2cac01723662728316b6204393cb87fa96690c7d6f7768/unitycatalog_langchain-0.1.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-13 00:25:34",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "unitycatalog-langchain"
}
