Easy to use python client for the Clio Manage API

This API client is designed to interact with the Clio API, providing robust handling for HTTP request methods, query parameters, payload validation, and dynamic endpoint management.

Installation

pip install clio-manage-api-client

Initialization

import clio_manage_python_client as Clio

client = Clio.Manage(access_token="your_access_token") # 'store_responses=True' for sqlite response handler

Or

from clio_manage_python_client import Manage as Client

client = Client(access_token="your_access_token")

Key Features

1. Dynamic Endpoint Management:

   • Dynamically initializes and manages endpoints using metadata from the models.endpoints module.
   • Supports operations like index, show, post, patch, delete as well as sub-methods for relational endpoints (Explained Below)
   • Endpoints that return content can use the download method. Output path can be provided in the call or set within the client

2. Integrated Query Parameter and Payload Handling:

   • Automatically validates, converts, and processes query parameters and payload data.
   • Ensures type safety and required field validation for all inputs.
   • Query parameters are processed first and removed from kwargs. Remaining kwargs are processed for payload data.

3. Pagination Support:

   • Handles paginated responses seamlessly when using the all method.
   • Automatically appends results from all pages.

4. Reserved Keywords and Parameter Mappings:

   • Handles reserved keywords and endpoint-specific parameters via a mappings dictionary.
   • Reserved keywords such as from and X-API-VERSION are mapped to their correct representations (e.g., from_ to from).
   • Parameters like matter_id and id are interchangeable, ensuring ease of use for endpoints that deviate from the general schema.

5. Dynamic Endpoint Resolution:

   • Determines whether to call the index or show endpoint based on the presence of id in kwargs.
   • For example:
     • client.get.matters(id=123) calls the show endpoint.
     • client.get.matters(limit=10) calls the index endpoint.

6. Rate Limiting:

   • Enforces rate limits using the RateMonitor class to ensure compliance with API restrictions.

7. Error Reporting:

   • Provides detailed validation error messages when inputs do not meet the expected types or requirements.

---

Examples

• Models are now generated when the client is first ran directly from the latest API documentation
• The Openapi spec file that is used to generate the dataclasses is now stored in the models/ subdirectory
• Get the latest changes made to the API by using the update.py script. Existing model classes will be backed up

Single Request Example

response = client.get.matters() # Default limit is 200
print(json.dumps(response, indent=2))

Relational Endpoint Example

related_contacts = client.get.matters.related_contacts(id=12345, fields="id,name")
print(json.dumps(related_contacts, indent=2))

Fetch All Paginated Results

response_all = client.all.matters(fields="id,name")
print(f"Total items retrieved: {len(response_all['data'])}")
print(json.dumps(response_all, indent=2))

Request With Generated Fields

Using all

# Returns data from all available non nested fields
response = client.get.matters(fields="all", limit=10)
print(json.dumps(response, indent=2))

# With nested fields
response = client.get.matters(fields="all,client{all}", limit=10)
print(json.dumps(response, indent=2))

# Returns the id and etag of every nested resource within the endpoint
response = client.get.matters(fields="all_ids", limit=10)
print(json.dumps(response, indent=2))

Datetime and Timedeltas

# Absolute dates
one_year_ago = date.today() - timedelta(days=365)
response = client.all.matters(limit=200, open_date__= f'>={one_year_ago}', order="open_date(asc)", fields="all,client{name},practice_area{name,category},responsible_attorney{name}")
print(json.dumps(response, indent=2))

# Helper functions: end_of_the_month()
entries = client.all.calendar_entries(fields="start_at,end_at,all_day,location,description,summary,attendees{name}", from_=datetime.now(), to=end_of_the_month())
print(json.dumps(response, indent=2))

Exporting

To Excel Spreadsheet:

Requires pandas < 2.0 and openpyxl

save_to_xlsx(client.all.contacts(fields="all,custom_field_values{field_name,value}"), "contacts.xlsx")

---

Logical Argument and Type Conversion

Example Endpoint:

• https://app.clio.com/api/v4/matters/{matter_id}/client.json
• https://app.clio.com/api/v4/matters/{id}.json

Both matter_id and id can be used interchangeably. The client automatically maps matter_id to id for consistency.

Example Usage

# Calls the 'show' endpoint for a specific matter
response = client.get.matters(id=123)

# Calls the 'client' relation endpoint for a specific matter
response = client.get.matters.client(matter_id=123)

# Calls the 'client' relation endpoint for a specific matter
response = client.get.matters.client(id=123)

Arrays(Needs more testing):

• Can be provided using by including a double underscore
• Any text after the underscores get converted into a query parameter keyword
  • ids__ = [123,456,789] gets converted to ids[]=123&ids[]=456&ids[]=789
  • jurisdiction__id = 123 gets converted to jurisdiction[id]=123

---

Reserved Keywords and Mapping

The client automatically handles reserved keywords and non-standard endpoint parameters through a predefined mappings dictionary. This ensures compatibility and ease of use for all endpoints.

Example Mappings

mappings = {
    "X_API_VERSION": "X-API-VERSION",
    "from_": "from", 
    "ids__": "ids",
    "jurisdiction__id": "jurisdiction[id]",
    "service_type__id": "service_type[id]",
    "trigger__id": "trigger[id]",
}

Internal Design

Endpoint Resolution

• The show endpoint is prioritized if id is present in the kwargs.
• The index endpoint is called if id is not provided.

_request_handler

Handles requests, including:

• Pagination support.
• Query parameter and payload validation.
• Dynamic path formatting with provided arguments.

---

Known Issues

• On some endpoints, DELETE throws an HTTPEXCEPTION but the command completes successfully in Clio. This can be circumvented by using a try/except block inside the running loop

License

This project is licensed under the MIT License.