[add]: code refractoring and cleanup

Dockerfile

@@ -1,20 +1,23 @@
 FROM python:3.11-slim
 
+# Set the working directory in the container
 WORKDIR /app
 
+# Copy the current directory contents into the container at /app
+COPY . /app
+
 RUN apt-get update && apt-get install -y libpq-dev && rm -rf /var/lib/apt/lists/*
 
 COPY requirements.txt .
 RUN pip install -r requirements.txt
 
-COPY salary_analytics/ ./salary_analytics/
 
 RUN mkdir -p output/csv output/plots output/models
 
-ENV PYTHONPATH=/app
+
-ENV HOST=0.0.0.0
+ENV FLASK_APP=wsgi.py
-ENV PORT=8000
+ENV FLASK_RUN_HOST=0.0.0.0
 
 EXPOSE 8000
 
-CMD ["uvicorn", "salary_analytics.api:app", "--host", "0.0.0.0", "--port", "8000", "--reload"] 
+CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:8000", "wsgi:wsgi_app"]

app/__init__.py

@@ -2,6 +2,15 @@ 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')
@@ -11,7 +20,7 @@ def create_app():
     migrate.init_app(app, db)
 
     # Register blueprints or CLI commands here if needed
-    from . import commands
+    from .commands import commands
     app.cli.add_command(commands.upload_xls_cli)
 
     return app 

app/analytics/integrations/rac_check.py

@@ -1,10 +1,8 @@
 from django.conf import settings
 import httpx
 import json
-from salary_analytics.config import SIMBRELLA_BASE_URL, SIMBRELLA_ENDPOINT_RAC_CHECKS
+from app.config import SIMBRELLA_BASE_URL, SIMBRELLA_ENDPOINT_RAC_CHECKS
-import logging
+from app.utils.logger import logger
-
-logger = logging.getLogger(__name__)
 
 class SimbrellaIntegration:
     BASE_URL = SIMBRELLA_BASE_URL

app/analytics/integrations/salary_detect.py

@@ -1,11 +1,8 @@
 import time
-import logging
 import threading
 import requests
-from .config import SALARY_DETECT_URL, SALARY_DETECT_HEADERS, get_random_salary_payload
+from ...config import SALARY_DETECT_URL, SALARY_DETECT_HEADERS, get_random_salary_payload
-
+from app.utils.logger import logger
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
 
 class SalaryDetect:
     def __init__(self):

app/analytics/services/__init__.py

@@ -0,0 +1,24 @@
+from .main import SalaryAnalyticsPipeline
+from .data_loader import DataLoader
+from .keyword_analyzer import KeywordAnalyzer
+from .consistent_amount_analyzer import ConsistentAmountAnalyzer
+from .transaction_type_analyzer import TransactionTypeAnalyzer
+from .salary_earner_analyzer import SalaryEarnerAnalyzer
+from .salary_predictor import SalaryPredictor
+
+
+"""
+Salary Analytics Package
+A package for analyzing and predicting salary patterns from transaction data.
+"""
+
+__version__ = "0.1.0" 
+__all__ = [
+    "SalaryAnalyticsPipeline",
+    "DataLoader",
+    "KeywordAnalyzer",
+    "ConsistentAmountAnalyzer",
+    "TransactionTypeAnalyzer",
+    "SalaryEarnerAnalyzer",
+    "SalaryPredictor"
+]

app/analytics/services/data_loader.py

@@ -8,8 +8,7 @@ from datetime import datetime
 import logging
 import os
 from .config import DB_CONFIG, TABLE_NAME
-
+from app.utils.logger import logger
-logger = logging.getLogger(__name__)
 
 class DataLoader:
     def __init__(self):

app/analytics/services/main.py

@@ -9,8 +9,7 @@ from .consistent_amount_analyzer import ConsistentAmountAnalyzer
 from .transaction_type_analyzer import TransactionTypeAnalyzer
 from .salary_earner_analyzer import SalaryEarnerAnalyzer
 from .salary_predictor import SalaryPredictor
-
+from app.utils.logger import logger
-logger = logging.getLogger(__name__)
 
 class SalaryAnalyticsPipeline:
     def __init__(self):

app/analytics/services/salary_earner_analyzer.py

@@ -6,15 +6,8 @@ import pandas as pd
 import matplotlib.pyplot as plt
 from matplotlib_venn import venn3
 from datetime import datetime, timedelta
-import logging
 from .config import MODEL_CONFIG, OUTPUT_PATHS
-
+from app.utils.logger import logger
-# Configure logging
-logging.basicConfig(
-    level=logging.INFO,
-    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
-)
-logger = logging.getLogger(__name__)
 
 class SalaryEarnerAnalyzer:
     def __init__(self, df):

app/api.py

@@ -17,20 +17,15 @@ from sqlalchemy import text, Table, Column, Integer, String, Float, DateTime, Me
 import numpy as np
 import warnings
 import time
-from .main import SalaryAnalyticsPipeline
+from .analytics.services.main import SalaryAnalyticsPipeline
 from .config import OUTPUT_PATHS, TABLE_NAME, BATCH_RESULTS_TABLE
 from .data_loader import DataLoader
 from .salary_predictor import SalaryPredictor
 from .salary_earner_analyzer import SalaryEarnerAnalyzer
 from .db_operations import DatabaseOperations
-from .salary_detect import SalaryDetect
+from .analytics.integrations.salary_detect import SalaryDetect
+from app.utils.logger import logger
 
-# Configure logging
-logging.basicConfig(
-    level=logging.INFO,
-    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
-)
-logger = logging.getLogger(__name__)
 
 # Suppress warnings
 warnings.filterwarnings('ignore', category=RuntimeWarning, module='numpy')

app/commands/commands.py

@@ -2,8 +2,8 @@ import click
 import pandas as pd
 from datetime import datetime
 from flask.cli import with_appcontext
-from salary_analytics.app.extensions import db
+from app.extensions import db
-from salary_analytics.app.models import RawTransaction
+from app.models import RawTransaction
 
 @click.group()
 def commands():

app/db_operations.py

@@ -2,12 +2,10 @@
 Database operations module for salary analytics.
 """
 
-import logging
 from sqlalchemy import text
 from .config import BATCH_RESULTS_TABLE
 from datetime import datetime
-
+from app.utils.logger import logger
-logger = logging.getLogger(__name__)
 
 class DatabaseOperations:
     def __init__(self, engine):

app/models/raw_transaction.py

@@ -1,4 +1,4 @@
-from .extensions import db
+from app.extensions import db
 
 class RawTransaction(db.Model):
     __tablename__ = 'analytics_raw_transactions'

app/utils/logger.py

@@ -0,0 +1,13 @@
+import logging
+
+# Configure logging
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s - %(levelname)s - %(message)s",
+    handlers=[
+        # logging.StreamHandler(),  
+        logging.FileHandler("app.log", mode='a')  # Log to file
+    ]
+)
+
+logger = logging.getLogger("DetectionService")

docker-compose.yml

@@ -3,14 +3,13 @@ services:
     build: .
     ports:
       - "${APP_PORT:-4800}:8000"
-    volumes:
-      - ./output:/app/output
     environment:
-      - DB_USER=salaryloan
+      - FLASK_APP=${FLASK_APP}
-      - DB_PASSWORD=salaryloan
+      - FLASK_ENV=${FLASK_ENV}
-      - DB_NAME=salaryloan
+      - DATABASE_URL=postgresql+psycopg2://${DATABASE_USER}:${DATABASE_PASSWORD}@${DATABASE_HOST}:${DATABASE_PORT}/${DATABASE_NAME}
-      - DB_PORT=10532
+    volumes:
-      - DB_HOST=dev-data.simbrellang.net
+      - .:/app
+      - ./output:/app/output
     restart: unless-stopped
     networks:
       - salary_network

migrations/env.py

@@ -19,7 +19,7 @@ if config.config_file_name is not None:
 # from myapp import Base
 # target_metadata = Base.metadata
 from flask import current_app
-from salary_analytics.app.extensions import db
+from app.extensions import db
 
 config.set_main_option('sqlalchemy.url',
                        current_app.config.get('SQLALCHEMY_DATABASE_URI'))

requirements.txt

@@ -17,4 +17,6 @@ openpyxl>=3.0.10
 Flask>=2.0.0
 Flask-SQLAlchemy>=3.0.0
 Flask-Migrate>=4.0.0
-alembic>=1.8.0
+alembic>=1.8.0
+requests>=2.26.0
+gunicorn

run.py

@@ -1,4 +0,0 @@
-import os
-from salary_analytics.app import create_app
-
-app = create_app() 

salary_analytics/__init__.py

@@ -1,6 +0,0 @@
-"""
-Salary Analytics Package
-A package for analyzing and predicting salary patterns from transaction data.
-"""
-
-__version__ = "0.1.0" 

wsgi.py

@@ -0,0 +1,7 @@
+from app import create_app
+
+app = create_app()
+
+if __name__ != "__main__":
+    # Expose WSGI app instance for Gunicorn
+    wsgi_app = app