From 4e221610882d7f3a41972e723a5ad9d80655f399 Mon Sep 17 00:00:00 2001
From: VivianDee <115420678+VivianDee@users.noreply.github.com>
Date: Tue, 9 Sep 2025 10:57:47 +0100
Subject: [PATCH] [add]: refactor/clean up

---
 app/__init__.py                             | 26 ---------
 app/api/__init__.py                         | 21 +++++++
 app/models/__init__.py                      |  4 ++
 app/salary_analytics/__init__.py            | 40 +++++++++++++
 app/salary_analytics/core/state.py          | 64 +++++++++++++++++++++
 app/salary_analytics/events/lifecycle.py    | 36 ++++++++++++
 app/salary_analytics/helpers/data_checks.py | 12 ++++
 app/salary_analytics/routes/base.py         | 28 +++++++++
 app/salary_analytics/routes/train.py        | 33 +++++++++++
 docker-compose.yml                          |  2 +
 10 files changed, 240 insertions(+), 26 deletions(-)
 delete mode 100644 app/__init__.py
 create mode 100644 app/api/__init__.py
 create mode 100644 app/models/__init__.py
 create mode 100644 app/salary_analytics/__init__.py
 create mode 100644 app/salary_analytics/core/state.py
 create mode 100644 app/salary_analytics/events/lifecycle.py
 create mode 100644 app/salary_analytics/helpers/data_checks.py
 create mode 100644 app/salary_analytics/routes/base.py
 create mode 100644 app/salary_analytics/routes/train.py

diff --git a/app/__init__.py b/app/__init__.py
deleted file mode 100644
index 32f1b84..0000000
--- a/app/__init__.py
+++ /dev/null
@@ -1,26 +0,0 @@
-from flask import Flask
-import os
-from .extensions import db, migrate
-
-
-"""
-Salary Analytics Package
-A package for analyzing and predicting salary patterns from transaction data.
-"""
-
-__version__ = "0.1.0" 
-
-
-def create_app():
-    app = Flask(__name__)
-    app.config.from_object('salary_analytics.config')
-
-    # Initialize extensions
-    db.init_app(app)
-    migrate.init_app(app, db)
-
-    # Register blueprints or CLI commands here if needed
-    from app.analytics.commands import commands
-    app.cli.add_command(commands.upload_xls_cli)
-
-    return app 
\ No newline at end of file
diff --git a/app/api/__init__.py b/app/api/__init__.py
new file mode 100644
index 0000000..dde0fd0
--- /dev/null
+++ b/app/api/__init__.py
@@ -0,0 +1,21 @@
+from flask import Flask
+import os
+from app.extensions import db, migrate
+
+
+
+def create_app():
+    app = Flask(__name__)
+    
+    # Load configuration from config.py
+    app.config.from_object('app.config')
+
+    # Initialize extensions
+    db.init_app(app)
+    migrate.init_app(app, db)
+
+    # Register blueprints or CLI commands here if needed
+    from app.api.commands import commands
+    app.cli.add_command(commands.upload_xls_cli)
+
+    return app 
\ No newline at end of file
diff --git a/app/models/__init__.py b/app/models/__init__.py
new file mode 100644
index 0000000..62baa90
--- /dev/null
+++ b/app/models/__init__.py
@@ -0,0 +1,4 @@
+from .raw_transaction import RawTransaction
+
+
+__all__ = ['RawTransaction']
\ No newline at end of file
diff --git a/app/salary_analytics/__init__.py b/app/salary_analytics/__init__.py
new file mode 100644
index 0000000..babbabb
--- /dev/null
+++ b/app/salary_analytics/__init__.py
@@ -0,0 +1,40 @@
+from fastapi import FastAPI
+from app.salary_analytics.routes import analysis, reports, pipeline, load, base, train
+from app.salary_analytics.middlewares.middleware import add_middlewares
+from app.salary_analytics.events.lifecycle import register_events
+from app.utils.logger import logger
+import socket
+
+"""
+Salary Analytics Package
+A package for analyzing and predicting salary patterns from transaction data.
+"""
+
+__version__ = "0.1.0" 
+
+
+def create_app() -> FastAPI:
+    app = FastAPI(
+        title="Salary Analytics API",
+        description="API for analyzing and predicting salary patterns from transaction data",
+        version="1.0.0"
+    )
+
+    # Middlewares
+    add_middlewares(app)
+
+    # Events
+    register_events(app)
+
+    # Routers
+    app.include_router(base.router, tags=["Base"])
+    app.include_router(analysis.router, prefix="/analyze", tags=["Analysis"])
+    app.include_router(reports.router, tags=["Reports"])
+    app.include_router(pipeline.router, tags=["Pipeline"])
+    app.include_router(load.router, tags=["Data"])
+    app.include_router(train.router, tags=["Model Training"])
+
+
+    return app
+
+app = create_app()
diff --git a/app/salary_analytics/core/state.py b/app/salary_analytics/core/state.py
new file mode 100644
index 0000000..4e59c2f
--- /dev/null
+++ b/app/salary_analytics/core/state.py
@@ -0,0 +1,64 @@
+from app.salary_analytics.services.main import SalaryAnalyticsPipeline
+from app.salary_analytics.services.data_loader import DataLoader
+
+
+class GlobalState:
+    def __init__(self):
+        self._pipeline = None
+        self._data_loader = None
+        self.df = None
+        self.salary_predictor = None
+        self.salary_earner_analyzer = None
+
+    # ---- Pipeline ----
+    @property
+    def pipeline(self):
+        if self._pipeline is None:
+            self._pipeline = SalaryAnalyticsPipeline()
+        return self._pipeline
+
+    @pipeline.setter
+    def pipeline(self, value):
+        self._pipeline = value
+
+    # ---- Data Loader ----
+    @property
+    def data_loader(self):
+        if self._data_loader is None:
+            self._data_loader = DataLoader()
+        return self._data_loader
+
+    @data_loader.setter
+    def data_loader(self, value):
+        self._data_loader = value
+
+    # ---- DataFrame ----
+    @property
+    def df(self):
+        return self._df
+
+    @df.setter
+    def df(self, value):
+        self._df = value
+
+    # ---- Salary Predictor ----
+    @property
+    def salary_predictor(self):
+        return self._salary_predictor
+
+    @salary_predictor.setter
+    def salary_predictor(self, value):
+        self._salary_predictor = value
+
+    # ---- Salary Earner Analyzer ----
+    @property
+    def salary_earner_analyzer(self):
+        return self._salary_earner_analyzer
+
+    @salary_earner_analyzer.setter
+    def salary_earner_analyzer(self, value):
+        self._salary_earner_analyzer = value
+
+
+state = GlobalState()
+
diff --git a/app/salary_analytics/events/lifecycle.py b/app/salary_analytics/events/lifecycle.py
new file mode 100644
index 0000000..635ab73
--- /dev/null
+++ b/app/salary_analytics/events/lifecycle.py
@@ -0,0 +1,36 @@
+import socket
+from fastapi import FastAPI
+from app.salary_analytics.integrations.salary_detect import SalaryDetect
+from app.utils.logger import logger
+
+salary_detect = SalaryDetect()
+
+
+def register_events(app: FastAPI):
+    @app.on_event("startup")
+    async def startup_event():
+        """Initialize the pipeline on startup."""
+        try:
+            logger.info("Initializing pipeline...")
+            
+            # Start autonomous salary detection loop
+            salary_detect.start()
+            logger.info("Started autonomous salary detection loop.")
+            
+            # Print network information
+            hostname = socket.gethostname()
+            ip_address = socket.gethostbyname(hostname)
+            logger.info(f"Server running on hostname: {hostname}")
+            logger.info(f"Server IP address: {ip_address}")
+            logger.info(f"Server is accessible at:")
+            logger.info(f"- http://localhost:8000")
+            logger.info(f"- http://127.0.0.1:8000")
+            logger.info(f"- http://{ip_address}:8000")
+            logger.info("Pipeline initialized successfully")
+        except Exception as e:
+            logger.error(f"Error during startup: {str(e)}")
+            raise
+
+    @app.on_event("shutdown")
+    async def shutdown_event():
+        logger.info("Shutting down Salary Analytics API...")
diff --git a/app/salary_analytics/helpers/data_checks.py b/app/salary_analytics/helpers/data_checks.py
new file mode 100644
index 0000000..47caa4a
--- /dev/null
+++ b/app/salary_analytics/helpers/data_checks.py
@@ -0,0 +1,12 @@
+from fastapi import HTTPException
+from app.salary_analytics.core.state import state
+
+
+def check_data_loaded():
+    """Raise HTTP 400 if no data is loaded into the pipeline."""
+    if state.pipeline.df is None:
+        raise HTTPException(
+            status_code=400,
+            detail="No data loaded. Please load data first using the /load-data endpoint."
+        )
+    return True
diff --git a/app/salary_analytics/routes/base.py b/app/salary_analytics/routes/base.py
new file mode 100644
index 0000000..c1319fb
--- /dev/null
+++ b/app/salary_analytics/routes/base.py
@@ -0,0 +1,28 @@
+from fastapi import APIRouter
+from app.utils.logger import logger
+import time
+
+
+router = APIRouter()
+
+@router.get("/")
+async def root():
+    """Root endpoint."""
+    start_time = time.time()
+    logger.info("Root endpoint accessed")
+    response = {"message": "Welcome to Salary Analytics API"}
+    logger.info(f"Root endpoint completed in {time.time() - start_time:.2f} seconds")
+    return response
+
+
+@router.get("/health")
+async def health_check():
+    """Health check endpoint."""
+    start_time = time.time()
+    logger.info("Health check endpoint accessed")
+    response = {"status": "healthy"}
+    logger.info(f"Health check completed in {time.time() - start_time:.2f} seconds")
+    return response
+
+
+
diff --git a/app/salary_analytics/routes/train.py b/app/salary_analytics/routes/train.py
new file mode 100644
index 0000000..94320cb
--- /dev/null
+++ b/app/salary_analytics/routes/train.py
@@ -0,0 +1,33 @@
+import time
+import logging
+from fastapi import APIRouter, HTTPException
+from app.salary_analytics.services.main import SalaryAnalyticsPipeline
+from app.salary_analytics.helpers.data_checks import check_data_loaded
+from app.salary_analytics.helpers.response_helpers import AnalysisResponse
+from app.salary_analytics.core.state import state
+from app.utils.logger import logger
+
+
+
+router = APIRouter()
+
+
+@router.post("/train/models", response_model=AnalysisResponse)
+async def train_models():
+    """Train salary prediction models."""
+    start_time = time.time()
+    try:
+        check_data_loaded()
+        logger.info("Starting model training...")
+        state.pipeline.train_salary_prediction_models()
+        logger.info("Models trained successfully")
+        response = AnalysisResponse(
+            message="Models trained successfully"
+        )
+        logger.info(f"Model training endpoint completed in {time.time() - start_time:.2f} seconds")
+        return response
+    except Exception as e:
+        logger.error(f"Error in model training: {str(e)}")
+        logger.info(f"Model training endpoint failed after {time.time() - start_time:.2f} seconds")
+        raise HTTPException(status_code=500, detail=str(e))
+
diff --git a/docker-compose.yml b/docker-compose.yml
index 2f701b2..9205cb3 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -1,6 +1,8 @@
 services:
   digifi-analytics:
     build: .
+    env_file:
+      - .env
     ports:
       - "${APP_PORT:-4800}:8000"
     environment: