Merge pull request #20 from guillaumepot/dev-0.1.3

Dev 0.1.3

.env

@@ -1,6 +1,8 @@
 # API configuration
 UVICORN_HOST=0.0.0.0
 UVICORN_PORT=8000
+API_VERSION=0.1.3
+API_STATE=DEV
 
 # API Auth configuration
 ALGORITHM="HS256"

.github/workflows/.env

@@ -1,2 +1,2 @@
-api_version=0.1.2
-streamlit_version=0.1.2
+api_version=0.1.3
+streamlit_version=0.1.3

README.md

@@ -141,6 +141,7 @@ This is a personal project to check my bank accounts. It is currently in develop
 
 ## Changelogs
 
+[v0.1.3](./changelogs/0.1.3.md)  
 [v0.1.2](./changelogs/0.1.2.md)  
 [v0.1.1](./changelogs/0.1.1.md)  
 [v0.1.0](./changelogs/0.1.0.md)
@@ -153,28 +154,35 @@ This is a personal project to check my bank accounts. It is currently in develop
 
 **[Done]**
 ```
+0.1.3 - Mini Analytics update
+- Streamlit Analytics page now display charts based on transactions
+- Clean Streamlit code
+
 0.1.2 - Repo Update
 - Review README.MD
-0.1.0
-- First version of the app containing an API & a Streamlit interface
+
 0.1.1
 - Refactor API files (cleaning & readability)
 - Refactor Streamlit files (cleaning & readability)
+
+0.1.0
+- First version of the app containing an API & a Streamlit interface
 ```
 
 
 **[Todo]**  
 ```
-0.1.3
+0.1.4 - Api security
 - Add a simple logger (API)
 - Update API security (max requests, etc.)
+0.1.5 - Streamlit code
+- Streamlit refactorization
+0.1.6 - CI Update
 - Add Container tests (Github Actions)
 - Add more unit tests, e.g., API tests
-
 0.2.0
 - Create an ETL pipline to get transaction datas, put raw datas in a storage; transform datas for analytics
 - Add analytics charts (based on transactions) -> Analytics page ; Streamlit
-
 0.3.0
 - Use a ML pipeline to predict things (budgets, ..)
 ```

changelogs/0.1.3.md

@@ -0,0 +1,26 @@
+## Changelogs 0.1.3
+
+---
+
+
+**Repo**
+```
+-
+```
+
+
+**POSTGRES**
+```
+-
+```
+
+**API**
+```
+- Changed how vars api_version & api_state are loaded (Now from env var)
+```
+
+**Streamlit**
+```
+- Analytics page now display charts based on transactions
+- Minor code changes
+```

docker-compose.yaml

@@ -43,7 +43,7 @@ services:
     image: unclad3610/personal_bank_app_streamlit:latest
     container_name: bank_app_streamlit
     environment:
-      - API_VERSION=0.1.0
+      - API_VERSION=0.1.3
       - API_URL=http://bank_app_api:8000
     ports:
       - "8001:8001"

src/api/api_vars.py

@@ -17,8 +17,8 @@
 VARS
 """
 # API metadatas
-api_version = "0.1.2"
-current_state = "Prod"
+api_version = os.getenv("API_VERSION", "0.1.3")
+current_state = os.getenv("API_STATE", "DEV")
 
 # UUID Generation
 def generate_uuid():

src/streamlit/Dockerfile

@@ -7,7 +7,7 @@ WORKDIR /app
 # Copy files
 COPY ./requirements.txt /app
 COPY ./streamlit.py /app
-
+COPY ./api_requests_functions.py /app
 
 # Install dependencies
 RUN pip install --no-cache-dir -r requirements.txt

src/streamlit/api_requests_functions.py

@@ -0,0 +1,158 @@
+import matplotlib.pyplot as plt
+from matplotlib import cm
+import numpy as np
+import os
+import pandas as pd
+import requests
+import streamlit as st
+
+
+# API VARS
+api_version = os.getenv("API_VERSION", "0.1.2")
+api_url = os.getenv("API_URL", "http://localhost:8000")
+
+login_placeholder = st.sidebar.empty()
+
+
+
+def get_api_status() -> None:
+    """
+    Request the api to get the status, block app if API is not responsding (or bad version)
+    """
+    url = f"{api_url}/api/{api_version}/status"
+    response = requests.get(url)
+    if response.status_code != 200:
+        st.error("Error while checking API status.")
+        st.stop()
+
+
+def validate_credentials(login_data: dict) -> None:
+    """
+    This function will be called when the "Log In" button is pressed
+    """
+    # Request API
+    url = f"{api_url}/api/{api_version}/login"
+    response = requests.post(url, data=login_data)
+    if response.status_code == 200:
+        access_token = response.json().get("access_token")
+        login_placeholder.empty()
+        st.session_state.access_token = access_token
+        st.sidebar.empty()
+
+    else:
+        st.sidebar.error("Incorrect Username or Password.")
+
+
+def get_account_table() -> pd.DataFrame:
+    """
+
+    """
+    # Request API
+    url = f"{api_url}/api/{api_version}/table/account"
+    account_table = requests.get(url, headers=st.session_state.headers)
+    account_table = dict(account_table.json())
+    account_table = account_table['account table']
+
+    # Generate DF
+    df_accounts = pd.DataFrame(account_table, columns=["id", "name", "type", "balance", "owner_id", "created_at"])
+    df_accounts["balance"] = df_accounts["balance"].apply(lambda x: f"{x:.2f} €")
+    df_accounts.drop(columns=["id", "owner_id"], inplace=True)
+    df_accounts["created_at"] = df_accounts["created_at"].str.split("T").str[0]
+
+    return df_accounts
+
+def get_budget_table() -> pd.DataFrame:
+    """
+    
+    """
+    # Request API
+    url = f"{api_url}/api/{api_version}/table/budget"
+    budget_table = requests.get(url, headers=st.session_state.headers)
+    budget_table = budget_table.json()
+    budget_table = budget_table["budget table"]
+
+    # Generate DF
+    df_budgets = pd.DataFrame(budget_table, columns=["id", "name", "month", "amount","created_at"])
+    df_budgets["amount"] = df_budgets["amount"].apply(lambda x: f"{x:.2f} €")
+
+    # Create a dictionary mapping from id to name in df_budgets (used for df_transactions)
+    budget_id_to_name = pd.Series(df_budgets.name.values, index=df_budgets.id).to_dict()
+
+    df_budgets.drop(columns=["id"], inplace=True)
+    df_budgets["created_at"] = df_budgets["created_at"].str.split("T").str[0]
+
+    return df_budgets, budget_id_to_name
+
+
+def get_transaction_table(budget_id_to_name) -> pd.DataFrame:
+    """
+    
+    """
+    # Request API
+    url = f"{api_url}/api/{api_version}/table/transaction"
+    transaction_table = requests.get(url, headers=st.session_state.headers)
+    transaction_table = dict(transaction_table.json())
+    transaction_table = transaction_table['transaction table']
+    # Generate DF
+    df_transactions = pd.DataFrame(transaction_table, columns=["id", "date", "type", "amount", "origin_account", "destination_account", "budget", "recipient", "category", "description"])
+    df_transactions["date"] = df_transactions["date"].str.split("T").str[0]
+    df_transactions["amount"] = df_transactions["amount"].apply(lambda x: f"{x:.2f} €")
+
+    # Map each budget id in df_transactions to its corresponding name using the dictionary
+    df_transactions["budget_name"] = df_transactions["budget"].map(budget_id_to_name).fillna("None")
+    df_transactions.drop(columns=["id", "budget"], inplace=True)
+    df_transactions.rename(columns={"budget_name": "budget"}, inplace=True)
+    df_transactions = df_transactions[["date", "type", "amount", "origin_account", "destination_account", "budget", "recipient", "category", "description"]]
+
+    return df_transactions
+
+
+
+def get_bar_chart(df, name, amount, legend=None):
+    """
+    
+    """
+    x = df[name]
+    y = df[amount]
+    fig = plt.figure(figsize=(10, 5))
+    colors = cm.viridis(np.linspace(0, 1, len(x)))
+
+    bars = plt.bar(x, y, color=colors)
+
+    #if legend is None:
+    for bar in bars:
+        height = bar.get_height()
+        plt.text(bar.get_x() + bar.get_width() / 2, height - (height*0.05), f'{height}'+"€", ha='center', va='top', color='white') 
+
+    plt.title(f"{name} expenses")
+    plt.xlabel(name, fontweight='bold')
+    plt.ylabel(amount, fontweight='bold')
+
+    if legend is not None:
+        labels = [elt for elt in legend]
+        plt.legend(bars, labels, title="Categories")
+
+    st.pyplot(fig)
+
+
+def get_time_series(df):
+    """
+    
+    """
+    # Convert 'date' to datetime format
+    df['date'] = pd.to_datetime(df['date'])
+
+    df['year_month'] = df['date'].dt.to_period('M')
+    df = df.groupby('year_month')['amount'].sum().reset_index()
+
+    # Calculate cumulative sum
+    df["cumsum"] = df["amount"].cumsum()
+
+    # Plot
+    fig = plt.figure(figsize=(10, 5))
+    plt.plot(df['year_month'].astype(str), df["cumsum"], color="blue")  # Convert 'year_month' to string for plotting
+    plt.fill_between(df['year_month'].astype(str), df["cumsum"], color="blue", alpha=0.3)
+    plt.title("Cumulative expenses over time")
+    plt.xlabel("Date", fontweight='bold')
+    plt.ylabel("Amount", fontweight='bold')
+    st.pyplot(fig)

src/streamlit/requirements.txt

@@ -15,6 +15,7 @@ jsonschema
 jsonschema-specifications 
 markdown-it-py 
 MarkupSafe 
+matplotlib
 mdurl 
 numexpr 
 numpy