Relativity API Ediscovery Automation for SOC 2

import requests
import time
import json
import os

# --- Configuration ---
RELATIVITY_API_URL = os.environ.get('RELATIVITY_API_URL', 'https://your-relativity-instance.com/api/v2') # Replace with your Relativity API base URL
API_KEY = os.environ.get('RELATIVITY_API_KEY', 'YOUR_API_KEY') # Use environment variable for security
CASE_ID = 'YOUR_CASE_ID' # Replace with your Case ID
SEARCH_ID = 'YOUR_SEARCH_ID' # Replace with your Search ID (for documents to monitor)
LOG_FILE = 'relativity_api_log.json'

# --- Exponential Backoff Settings ---
INITIAL_SLEEP_TIME = 5 # seconds
MAX_SLEEP_TIME = 300 # seconds (5 minutes)
BACKOFF_FACTOR = 2 # Factor for exponential increase
MAX_RETRIES = 10

def log_event(event_type, message, status='success', data=None):
    log_entry = {
        'timestamp': time.strftime('%Y-%m-%dT%H:%M:%SZ'),
        'event_type': event_type,
        'message': message,
        'status': status,
        'data': data
    }
    try:
        with open(LOG_FILE, 'a') as f:
            json.dump(log_entry, f)
            f.write('\n')
    except Exception as e:
        print(f"Error writing to log file: {e}")

def get_document_status(document_artifact_id):
    endpoint = f"{RELATIVITY_API_URL}/Cases/{CASE_ID}/Documents/{document_artifact_id}"
    headers = {
        'Content-Type': 'application/json',
        'X-API-Key': API_KEY
    }
    try:
        response = requests.get(endpoint, headers=headers)
        response.raise_for_status() # Raise HTTPError for bad responses (4xx or 5xx)
        return response.json(), 'success'
    except requests.exceptions.HTTPError as e:
        log_event('API_REQUEST', f"HTTP error getting document status for {document_artifact_id}: {e}", 'error', {'status_code': e.response.status_code, 'response_text': e.response.text})
        return None, f"HTTP Error: {e.response.status_code}"
    except requests.exceptions.RequestException as e:
        log_event('API_REQUEST', f"Request error getting document status for {document_artifact_id}: {e}", 'error')
        return None, f"Request Error: {e}"

def poll_documents_for_status():
    # In a real scenario, you'd fetch a list of documents to monitor, perhaps from a database or another API call.
    # For demonstration, we'll simulate polling for a specific document ID or a search result.
    # A more robust approach involves using Relativity's Search functionality via API to get document IDs.
    
    # Example: Simulate fetching document IDs from a search
    # NOTE: You'd need an endpoint to get search results, this is a placeholder.
    # Example: endpoint = f"{RELATIVITY_API_URL}/Cases/{CASE_ID}/Searches/{SEARCH_ID}/Documents"
    # For simplicity, we'll hardcode a document artifact ID for demonstration.
    
    document_ids_to_check = ['12345', '67890'] # Replace with actual document artifact IDs or fetch dynamically
    
    for doc_id in document_ids_to_check:
        sleep_time = INITIAL_SLEEP_TIME
        retries = 0
        while retries < MAX_RETRIES:
            data, status = get_document_status(doc_id)
            if status == 'success' and data:
                log_event('DOCUMENT_STATUS', f"Successfully retrieved status for document {doc_id}", 'success', data)
                # Process document data here - e.g., check 'ArtifactId', 'ControlNumber', 'InstanceStatus' etc.
                # For example, you might check if 'InstanceStatus' indicates processing is complete.
                if data.get('InstanceStatus') == 'Completed':
                    print(f"Document {doc_id} processing complete.")
                    # Trigger next step in workflow (e.g., export, notification)
                    break # Move to the next document
                else:
                    print(f"Document {doc_id} status: {data.get('InstanceStatus')}. Continuing poll.")
                    # Potentially wait longer before checking this specific document again if needed,
                    # but generally you'd move to the next document in the list.
                    break # Assuming we check all documents in one pass, move to next doc_id
            else:
                print(f"Failed to get status for {doc_id}. Retry {retries+1}/{MAX_RETRIES}. Error: {status}")
                time.sleep(min(sleep_time, MAX_SLEEP_TIME))
                sleep_time *= BACKOFF_FACTOR
                retries += 1
        if retries == MAX_RETRIES:
            log_event('DOCUMENT_STATUS', f"Max retries reached for document {doc_id}", 'error')

if __name__ == "__main__":
    log_event('SCRIPT_START', 'Relativity API Poller script started')
    try:
        poll_documents_for_status()
    except Exception as e:
        log_event('SCRIPT_ERROR', f'An unexpected error occurred: {e}', 'error')
    log_event('SCRIPT_END', 'Relativity API Poller script finished')