Initial work on db validation

schemas/account_validator.json

@@ -0,0 +1,42 @@
+{
+	"$jsonSchema": {
+		"bsonType": "object",
+		"title": "Account validator",
+		"required": [
+			"name",
+			"role",
+			"created_at",
+			"updated_at"
+		],
+		"properties": {
+			"name": {
+				"bsonType": "string",
+				"description": "'name' must be a string is required."
+			},
+			"role": {
+				"enum": [
+					"Doctor",
+					"Healthcare Prof",
+					"Nurse",
+					"Caregiver",
+					"Physicion",
+					"Medical Student",
+					"Other"
+				],
+				"description": "'role' must be one of: [Doctor, Healthcare Prof, Nurse, Caregiver, Physicion, Medical Student, Other] and is required."
+			},
+			"specialty": {
+				"bsonType": "string",
+				"description": "'speciality' must be a string and is optional."
+			},
+			"created_at": {
+				"bsonType": "date",
+				"description": "'created_at' must be a date and is required."
+			},
+			"updated_at": {
+				"bsonType": "date",
+				"description": "'updated_at' must be a date and is required."
+			}
+		}
+	}
+}

schemas/patient_validator.json

@@ -0,0 +1,59 @@
+{
+	"$jsonSchema": {
+		"bsonType": "object",
+		"title": "Patient validator",
+		"required": [
+			"name",
+			"age",
+			"sex",
+			"assigned_doctor_id",
+			"created_at",
+			"updated_at"
+		],
+		"properties": {
+			"assigned_doctor_id": {
+				"bsonType": ""
+			},
+			"name": {
+				"bsonType": "string",
+				"description": "'name' must be a string is required."
+			},
+			"age": {
+				"bsonType": "uint",
+				"description": "'description' must be an unsigned int and is required."
+			},
+			"sex": {
+				"bsonType": "string",
+				"description": "sex must be string and is required."
+			},
+			"address": {
+				"bsonType": "string",
+				"description": "'address' must be string and is optional."
+			},
+			"phone": {
+				"bsonType": "string",
+				"description": "'phone' must be a string and is optional."
+			},
+			"email": {
+				"bsonType": "string",
+				"description": "'email' must be a string and is optional."
+			},
+			"medications": {
+				"bsonType": "string",
+				"description": "'medication' must be a  and is optional."
+			},
+			"past_assessment_summary": {
+				"bsonType": "string",
+				"description": "'past_assessment_summary' must be a string and is optional."
+			},
+			"created_at": {
+				"bsonType": "date",
+				"description": "'created_at' must be a date and is required."
+			},
+			"updated_at": {
+				"bsonType": "date",
+				"description": "'updated_at' must be a date and is required."
+			}
+		}
+	}
+}

schemas/session_validator.json

@@ -0,0 +1,35 @@
+{
+	"$jsonSchema": {
+		"bsonType": "object",
+		"title": "Session validator",
+		"required": [
+			"user_id",
+			"title",
+			"created_at",
+			"updated_at"
+		],
+		"properties": {
+			"user_id": {
+				"bsonType": "string",
+				"description": "'user_id' must be a string is required."
+			},
+			"title": {
+				"bsonType": "string",
+				"description": "'title' must be a string and is optional."
+			},
+			"created_at": {
+				"bsonType": "date",
+				"description": "'created_at' must be a date and is required."
+			},
+			"updated_at": {
+				"bsonType": "date",
+				"description": "'updated_at' must be a date and is required."
+			},
+			"messages": {
+				"bsonType": "array",
+				"items": {"bsonType": "string"},
+				"description": "'messages' must be a string and is optional."
+			}
+		}
+	}
+}

src/data/connection.py

@@ -1,5 +1,6 @@
 # data/repositories/base.py
 
+import json
 import os
 
 from pymongo import MongoClient
@@ -14,7 +15,9 @@ from src.utils.logger import logger
 
 class ActionFailed(Exception):
 	"""Raised when a database action fails."""
-	pass
+
+class EntryNotFound(Exception):
+	"""Raised when an entry cannot be found in the database."""
 
 _mongo_client: MongoClient | None = None
 
@@ -41,6 +44,23 @@ def close_connection():
 		_mongo_client = None
 
 def get_collection(name: str) -> Collection:
-	"""Retrieves a MongoDB collection by name."""
-	db = get_database()
-	return db.get_collection(name)
+	"""Retrieves a MongoDB collection by name. Create it if it does not exist."""
+	return get_database().get_collection(name)
+
+def does_collection_exist(name: str) -> bool:
+	return True if name in get_database().list_collection_names() else False
+
+def create_collection(collection_name: str, validator_path: str):
+	#get_collection(collection_name).drop()
+	if does_collection_exist(collection_name):
+		raise ActionFailed("Collection already exists")
+
+	with open(validator_path, "r", encoding="utf-8") as f:
+		validator = json.load(f)
+	get_database().create_collection(
+		collection_name,
+		validator=validator,
+		validationLevel="moderate"
+	)
+	#logger().info(validator)
+	logger().info(collection_name + " database initialised")

src/data/repositories/account.py

@@ -1,6 +1,16 @@
 # data/repositories/account.py
 """
 User account management operations for MongoDB.
+Each account represents a doctor.
+
+## Fields
+	_id: index
+	name: the name attached to the account
+	role:
+	specialty:
+	medical_roles:
+	created_at: the timestamp when the account was created
+	updated_at: the timestamp when the account data was last modified
 """
 
 import re
@@ -12,14 +22,15 @@ from pymongo import ASCENDING
 from pymongo.errors import (ConnectionFailure, DuplicateKeyError,
                             OperationFailure, PyMongoError)
 
-from src.data.connection import ActionFailed, get_collection
+from src.data.connection import (ActionFailed, EntryNotFound,
+                                 create_collection, get_collection)
 from src.utils.logger import logger
 
 ACCOUNTS_COLLECTION = "accounts"
 
-class UserNotFound(Exception):
-	"""Raised when a user is not found in the database."""
-	pass
+def create():
+	#get_collection(ACCOUNTS_COLLECTION).drop()
+	create_collection(ACCOUNTS_COLLECTION, "schemas/account_validator.json")
 
 def get_account_frame(
 	*,
@@ -120,13 +131,13 @@ def set_user_preferences(
 			}
 		)
 		if result.matched_count == 0:
-			raise UserNotFound(f"User with ID '{user_id}' not found.")
+			raise EntryNotFound(f"User with ID '{user_id}' not found.")
 
 		return result.modified_count > 0
 	except PyMongoError as e:
 		logger().error(f"An error occurred with the database operation: {e}")
 		return False
-	except UserNotFound as e:
+	except EntryNotFound as e:
 		logger().error(e)
 		return False
 

src/data/repositories/medical.py

@@ -1,6 +1,8 @@
 # data/repositories/medical.py
 """
 Medical records and memory management operations for MongoDB.
+
+@Note Could this be split into two? One for records and one for memory.
 """
 
 from datetime import datetime, timezone

src/data/repositories/message.py

@@ -1,8 +1,17 @@
 # data/message/operations.py
 """
 Message management operations for MongoDB.
-
-@TODO Review and revise.
+A message is owned by a session, and is sent by either the user or the ai.
+
+## Fields
+	_id: index
+	session_id:
+	patient_id:
+	doctor_id:
+	role:
+	content:
+	timestamp:
+	created_at:
 """
 
 from datetime import datetime, timezone
@@ -10,13 +19,19 @@ from typing import Any
 
 from bson import ObjectId
 from pymongo import ASCENDING
+from pymongo.errors import (ConnectionFailure, DuplicateKeyError,
+                            OperationFailure, PyMongoError)
 
-from src.data.connection import get_collection
+from src.data.connection import ActionFailed, create_collection, get_collection
 from src.utils.logger import logger
 
 CHAT_SESSIONS_COLLECTION = "chat_sessions"
 CHAT_MESSAGES_COLLECTION = "chat_messages"
 
+def create():
+	#get_collection(CHAT_MESSAGES_COLLECTION).drop()
+	create_collection(CHAT_MESSAGES_COLLECTION, "schemas/message_validator.json")
+
 def add_message(
 	session_id: str,
 	message_data: dict[str, Any],

src/data/repositories/patient.py

@@ -1,6 +1,21 @@
 # data/patient/operations.py
 """
 Patient management operations for MongoDB.
+A patient is a person who has been assigned to a doctor for treatment.
+
+## Fields
+	_id: index
+	name:
+	age:
+	sex:
+	address:
+	phone:
+	email:
+	medications:
+	past_assessment_summary:
+	assigned_doctor_id:
+	created_at:
+	updated_at:
 """
 
 import re
@@ -8,93 +23,99 @@ from datetime import datetime, timezone
 from typing import Any
 
 from pymongo import ASCENDING
+from pymongo.errors import (ConnectionFailure, DuplicateKeyError,
+                            OperationFailure, PyMongoError)
 
-from src.data.connection import get_collection
+from src.data.connection import ActionFailed, create_collection, get_collection
 from src.utils.logger import logger
 
 PATIENTS_COLLECTION = "patients"
 
+def create():
+	#get_collection(PATIENTS_COLLECTION).drop()
+	create_collection(PATIENTS_COLLECTION, "schemas/patient_validator.json")
+
 def _generate_patient_id() -> str:
-    """Generate zero-padded 8-digit ID"""
-    import random
-    return f"{random.randint(0, 99999999):08d}"
+	"""Generate zero-padded 8-digit ID"""
+	import random
+	return f"{random.randint(0, 99999999):08d}"
 
 
 def get_patient_by_id(patient_id: str) -> dict[str, Any] | None:
-    collection = get_collection(PATIENTS_COLLECTION)
-    return collection.find_one({"patient_id": patient_id})
+	collection = get_collection(PATIENTS_COLLECTION)
+	return collection.find_one({"patient_id": patient_id})
 
 
 def create_patient(
-    *,
-    name: str,
-    age: int,
-    sex: str,
-    address: str | None = None,
-    phone: str | None = None,
-    email: str | None = None,
-    medications: list[str] | None = None,
-    past_assessment_summary: str | None = None,
-    assigned_doctor_id: str | None = None
+	*,
+	name: str,
+	age: int,
+	sex: str,
+	address: str | None = None,
+	phone: str | None = None,
+	email: str | None = None,
+	medications: list[str] | None = None,
+	past_assessment_summary: str | None = None,
+	assigned_doctor_id: str | None = None
 ) -> dict[str, Any]:
-    collection = get_collection(PATIENTS_COLLECTION)
-    now = datetime.now(timezone.utc)
-    # Ensure unique 8-digit id
-    for _ in range(10):
-        pid = _generate_patient_id()
-        if not collection.find_one({"patient_id": pid}):
-            break
-    else:
-        raise RuntimeError("Failed to generate unique patient ID")
-    doc = {
-        "patient_id": pid,
-        "name": name,
-        "age": age,
-        "sex": sex,
-        "address": address,
-        "phone": phone,
-        "email": email,
-        "medications": medications or [],
-        "past_assessment_summary": past_assessment_summary or "",
-        "assigned_doctor_id": assigned_doctor_id,
-        "created_at": now,
-        "updated_at": now
-    }
-    collection.insert_one(doc)
-    return doc
+	collection = get_collection(PATIENTS_COLLECTION)
+	now = datetime.now(timezone.utc)
+	# Ensure unique 8-digit id
+	for _ in range(10):
+		pid = _generate_patient_id()
+		if not collection.find_one({"patient_id": pid}):
+			break
+	else:
+		raise RuntimeError("Failed to generate unique patient ID")
+	doc = {
+		"patient_id": pid,
+		"name": name,
+		"age": age,
+		"sex": sex,
+		"address": address,
+		"phone": phone,
+		"email": email,
+		"medications": medications or [],
+		"past_assessment_summary": past_assessment_summary or "",
+		"assigned_doctor_id": assigned_doctor_id,
+		"created_at": now,
+		"updated_at": now
+	}
+	collection.insert_one(doc)
+	return doc
 
 
 def update_patient_profile(patient_id: str, updates: dict[str, Any]) -> int:
-    collection = get_collection(PATIENTS_COLLECTION)
-    updates["updated_at"] = datetime.now(timezone.utc)
-    result = collection.update_one({"patient_id": patient_id}, {"$set": updates})
-    return result.modified_count
+	collection = get_collection(PATIENTS_COLLECTION)
+	updates["updated_at"] = datetime.now(timezone.utc)
+	result = collection.update_one({"patient_id": patient_id}, {"$set": updates})
+	return result.modified_count
 
 
 def search_patients(query: str, limit: int = 10) -> list[dict[str, Any]]:
-    """Search patients by name (case-insensitive starts-with/contains) or partial patient_id."""
-    collection = get_collection(PATIENTS_COLLECTION)
-    if not query:
-        return []
-
-    logger().info(f"Searching patients with query: '{query}', limit: {limit}")
-
-    # Build a regex for name search and patient_id partial match
-    pattern = re.compile(re.escape(query), re.IGNORECASE)
-
-    try:
-        cursor = collection.find({
-            "$or": [
-                {"name": {"$regex": pattern}},
-                {"patient_id": {"$regex": pattern}}
-            ]
-        }).sort("name", ASCENDING).limit(limit)
-        results = []
-        for p in cursor:
-            p["_id"] = str(p.get("_id")) if p.get("_id") else None
-            results.append(p)
-        logger().info(f"Found {len(results)} patients matching query")
-        return results
-    except Exception as e:
-        logger().error(f"Error in search_patients: {e}")
-        return []
+	"""Search patients by name (case-insensitive starts-with/contains) or partial patient_id."""
+	collection = get_collection(PATIENTS_COLLECTION)
+	if not query:
+		return []
+
+	logger().info(f"Searching patients with query: '{query}', limit: {limit}")
+
+	# Build a regex for name search and patient_id partial match
+	pattern = re.compile(re.escape(query), re.IGNORECASE)
+
+	try:
+		cursor = collection.find({
+			"$or": [
+				{"name": {"$regex": pattern}},
+				{"patient_id": {"$regex": pattern}}
+			]
+		}).sort("name", ASCENDING).limit(limit)
+		results = []
+		for p in cursor:
+			p["_id"] = str(p.get("_id")) if p.get("_id") else None
+			results.append(p)
+		logger().info(f"Found {len(results)} patients matching query")
+		return results
+	except Exception as e:
+		logger().error(f"Error in search_patients: {e}")
+		return []

src/data/repositories/session.py

@@ -1,8 +1,16 @@
-# data/repositories/chat.py
+# data/repositories/session.py
 """
 Chat session management operations for MongoDB.
-
-@TODO Review and revise.
+A session is owned by an account and is related to a patient.
+A session contains many messages.
+
+## Fields
+	_id: index
+	user_id:
+	title:
+	created_at:
+	updated_at:
+	messages:
 """
 
 import uuid
@@ -13,12 +21,16 @@ from pymongo import DESCENDING
 from pymongo.errors import (ConnectionFailure, DuplicateKeyError,
                             OperationFailure, PyMongoError)
 
-from src.data.connection import ActionFailed, get_collection
+from src.data.connection import ActionFailed, create_collection, get_collection
 from src.utils.logger import logger
 
 CHAT_SESSIONS_COLLECTION = "chat_sessions"
 CHAT_MESSAGES_COLLECTION = "chat_messages"
 
+def create():
+	#get_collection(CHAT_SESSIONS_COLLECTION).drop()
+	create_collection(CHAT_SESSIONS_COLLECTION, "schemas/session_validator.json")
+
 def create_session(
 	user_id: str,
 	title: str,

src/main.py

@@ -24,9 +24,20 @@ except Exception as e:
 	logger(tag="env").warning(f"Error loading .env file: {e}")
 
 # Import project modules after trying to load environment variables
-from src.api.routes import (audio, chat, doctors, patients, session, static,
-                            system, user)
+from src.api.routes import audio as audio_route
+from src.api.routes import chat as chat_route
+from src.api.routes import doctors as doctors_route
+from src.api.routes import patients as patients_route
+from src.api.routes import session as session_route
+from src.api.routes import static as static_route
+from src.api.routes import system as system_route
+from src.api.routes import user as user_route
 from src.core.state import MedicalState, get_state
+from src.data.repositories import account as account_repo
+from src.data.repositories import medical as medical_repo
+from src.data.repositories import message as message_repo
+from src.data.repositories import patient as patient_repo
+from src.data.repositories import session as session_repo
 
 
 def startup_event(state: MedicalState):
@@ -68,6 +79,13 @@ def startup_event(state: MedicalState):
 
 	logger(tag="startup").info("Medical AI Assistant startup complete")
 
+	# TODO On first startup, create all repositories if they don't exist
+	#account_repo.create()
+	#patient_repo.create()
+	#session_repo.create()
+	#message_repo.create()
+	#medical_repo.create()
+
 def shutdown_event():
 	"""Cleanup on shutdown"""
 	logger(tag="shutdown").info("Shutting down Medical AI Assistant...")
@@ -105,14 +123,14 @@ app.add_middleware(
 app.mount("/static", StaticFiles(directory="static"), name="static")
 
 # Include routers
-app.include_router(chat.router)
-app.include_router(user.router)
-app.include_router(session.router)
-app.include_router(patients.router)
-app.include_router(doctors.router)
-app.include_router(system.router)
-app.include_router(static.router)
-app.include_router(audio.router)
+app.include_router(chat_route.router)
+app.include_router(user_route.router)
+app.include_router(session_route.router)
+app.include_router(patients_route.router)
+app.include_router(doctors_route.router)
+app.include_router(system_route.router)
+app.include_router(static_route.router)
+app.include_router(audio_route.router)
 
 @app.get("/api/info")
 async def get_api_info():

tests/mongo_test.py

@@ -1,6 +1,8 @@
 """MongoDB Integration Tests
 
 To run this file, execute `python -m tests.mongo_test` from the project root directory.
+
+@TODO This file is outdated and will need to be updated after the validators have been implemented.
 """
 
 import unittest