app.py

import base64
import json
import os
import uuid
from datetime import datetime
from pathlib import Path

import pandas as pd
import pytz
import streamlit as st
from datasets import Dataset, load_dataset, Features, Value
from huggingface_hub import CommitScheduler

# File paths as constants
USERS_JSON = 'leaders/users.json'
MATCHES_JSON = 'matches.json'
OUTCOMES_JSON = 'match_outcomes.json'
OUTCOMES = 'outcomes/match_outcomes.json'
PLAYERS_JSON = 'players.json'
image_path = 'ipl_image.png'


PREDICTIONS_FOLDER = Path("predictions")
PREDICTIONS_FOLDER.mkdir(parents=True, exist_ok=True)

users_file = Path("leaders") / f"users.json"
USERS_FOLDER = users_file.parent
USERS_FOLDER.mkdir(parents=True, exist_ok=True)

outcomes_file = Path("outcomes") / f"match_outcomes.json"
OUTCOMES_FOLDER = outcomes_file.parent
OUTCOMES_FOLDER.mkdir(parents=True, exist_ok=True)

# Initialize CommitScheduler
scheduler = CommitScheduler(
    repo_id="DIS_IPL_Preds",
    repo_type="dataset",
    folder_path=PREDICTIONS_FOLDER,  # Local folder where predictions are saved temporarily
    path_in_repo="predictions",  # Path in dataset repo where predictions will be saved
    every=720,  # Push every 240 minutes (4 hours)
)

# Initialize CommitScheduler
scheduler = CommitScheduler(
    repo_id="DIS_IPL_Leads",
    repo_type="dataset",
    folder_path=USERS_FOLDER,  # Local folder where users are saved temporarily
    path_in_repo="leaders",  # Path in dataset repo where predictions will be saved
    every=720,  # Push every 240 minutes (4 hours)
)

# Initialize CommitScheduler
scheduler = CommitScheduler(
    repo_id="DIS_IPL_Outcomes",
    repo_type="dataset",
    folder_path=OUTCOMES_FOLDER,  # Local folder where users are saved temporarily
    path_in_repo="outcomes",  # Path in dataset repo where predictions will be saved
    every=720,  # Push every 240 minutes (4 hours)
)


def load_data(file_path):
    """
    Load data from a JSON or CSV file.
    
    Args:
    file_path (str): The path to the file to load.
    
    Returns:
    pd.DataFrame or dict: The loaded data.
    """
    try:
        if file_path.endswith('.json'):
            with open(file_path, 'r') as file:
                return json.load(file)
        elif file_path.endswith('.csv'):
            return pd.read_csv(file_path)
    except FileNotFoundError:
        if file_path.endswith('.json'):
            return {}
        elif file_path.endswith('.csv'):
            return pd.DataFrame()


def get_base64_of_image(path):
    with open(path, "rb") as image_file:
        return base64.b64encode(image_file.read()).decode()


# Get today's date in IST to load today's match
def get_current_date_ist():
    tz_IST = pytz.timezone('Asia/Kolkata')
    datetime_ist = datetime.now(tz_IST)
    return datetime_ist.strftime('%Y-%m-%d')


# Function to get matches for today
def get_today_matches():
    today = get_current_date_ist()
    matches = load_data(MATCHES_JSON)
    today_matches = [match for match in matches if match['date'] == today]
    return today_matches


# Function to check if prediction submission is allowed
def is_submission_allowed(match_id):
    matches = load_data(MATCHES_JSON)  # This loads matches correctly with IST times

    for match in matches:
        if match["match_id"] == match_id:
            # Parse the match start time in IST
            tz_IST = pytz.timezone('Asia/Kolkata')
            match_datetime_str = f'{match["date"]} {match["time"]}'
            # The match time string is like "2024-03-21 7:30 PM"
            match_datetime = datetime.strptime(match_datetime_str, "%Y-%m-%d %I:%M %p")
            match_datetime = tz_IST.localize(match_datetime)  # Set the timezone to IST

            # Get the current time in IST
            current_datetime = datetime.now(tz_IST)

            if current_datetime > match_datetime:
                return False
            else:
                return True
    return False  # If match_id not found, default to False


# Submit prediction function
def submit_prediction(
        user_name,
        match_id,
        predicted_winner,
        predicted_motm,
        bid_points,
        max_bid_points
        ):

    # Validation for user selection
    if user_name == "Select a user...":
        st.warning("Please select a valid user.")
        return

    # Check if prediction submission is allowed for the match
    if not is_submission_allowed(match_id):
        st.error("Prediction submission time has passed. Predictions can't be submitted after match start.")
        return

    if bid_points > max_bid_points or bid_points < 100:
        st.error(f"Oops, Invalid bid! 🙈 You must bid at least 100 points and not exceed the 20% limit of your total points. Maximum allowed bid points: {max_bid_points}.")
        # st.error(f"Oops! 🙈 Looks like you're going overboard with your bid points! Your bid points cannot exceed your total points. Maximum allowed bid points: {max_bid_points} 😱")
        return

    prediction_id = uuid.uuid4().hex
    prediction_time = datetime.now().strftime('%Y-%m-%d')

    prediction_data = {
        'prediction_id': prediction_id,
        'user_name': user_name,
        'match_id': match_id,
        'predicted_winner': predicted_winner,
        'predicted_motm': predicted_motm,
        'bid_points': bid_points,
        'prediction_date': prediction_time  # Include the prediction time
    }
    
    # Construct the filename to include match_id for easier retrieval
    prediction_file_name = f"prediction_{match_id}_{user_name}.json"
    prediction_file = PREDICTIONS_FOLDER / prediction_file_name
    
    # Load existing predictions for the user and match, if any
    existing_predictions = []
    if prediction_file.exists():
        with prediction_file.open("r") as file:
            for line in file:
                existing_predictions.append(json.loads(line.strip()))

    # Update existing prediction if it exists for the same user and match
    prediction_updated = False
    for existing_prediction in existing_predictions:
        if existing_prediction['user_name'] == user_name and existing_prediction['match_id'] == match_id:
            existing_prediction.update(prediction_data)
            prediction_updated = True
            break

    # Save the updated predictions back to the file
    with scheduler.lock:
        if not prediction_updated:
            # Append the new prediction if it doesn't already exist
            with prediction_file.open("a") as file:
                file.write(json.dumps(prediction_data))
                file.write("\n")
        else:
            with prediction_file.open("w") as file:
                for prediction in existing_predictions:
                    file.write(json.dumps(prediction))
                    file.write("\n")

    st.success("Prediction submitted successfully!")


def get_user_total_points(user_name):
    # users_dataset = load_dataset("Jay-Rajput/DIS_IPL_Leads", split="train")
    # users = users_dataset.to_dict()
    users = load_users(USERS_JSON)
    return users.get(user_name, {}).get('points')


def calculate_max_bid_points(user_name):
    total_points = get_user_total_points(user_name)
    max_bid_points = int(total_points * 0.20)  # 20% of total points
    return max_bid_points


def load_users(USERS_JSON):
    try:
        with open(USERS_JSON, 'r') as file:
            return json.load(file)
    except FileNotFoundError:
        return {}


def user_selection_and_prediction():
    users = list(load_data(USERS_JSON))
    user_name = st.selectbox("Select User", ["Select a user..."] + users)

    max_bid_points = None
    if user_name != "Select a user...":
        max_bid_points = calculate_max_bid_points(user_name)
        st.write(f"Maximum bid points you can submit: {max_bid_points}")

    matches = get_today_matches()
    if matches:
        match_choice = st.selectbox("Select Today's Match", matches, format_func=lambda match: f"{match['teams'][0]} vs {match['teams'][1]}")
        match_id = match_choice['match_id']
        teams = match_choice['teams']

        predicted_winner = st.selectbox("Predicted Winner", teams)

        player_list = load_data(PLAYERS_JSON)
        predicted_motm = ""
        if predicted_winner in player_list:
            players = player_list[predicted_winner]
            predicted_motm = st.selectbox("Predicted Man of the Match", players)

        bid_points = st.number_input("Bid Points", min_value=0, value=100, format="%d")

        if st.button("Submit Prediction"):
            submit_prediction(user_name, match_id, predicted_winner, predicted_motm, bid_points, max_bid_points)
    else:
        st.write("No matches are scheduled for today.")


def display_predictions():
    if st.button("Show Predictions"):
        all_predictions = []

        # Check if the directory exists
        if not os.path.exists(PREDICTIONS_FOLDER):
            st.write("No predictions directory found.")
            return

        # List all JSON files in the directory
        for filename in os.listdir(PREDICTIONS_FOLDER):
            if filename.endswith('.json'):
                file_path = os.path.join(PREDICTIONS_FOLDER, filename)
                # Read each JSON file and append its contents to the list
                with open(file_path, 'r') as file:
                    prediction = json.load(file)
                    all_predictions.append(prediction)

        # Convert the list of dictionaries to a DataFrame
        predictions_df = pd.DataFrame(all_predictions)

        if not predictions_df.empty:
            predictions_df['prediction_date'] = predictions_df.apply(lambda x: datetime.strptime(x['prediction_date'], '%Y-%m-%d'), axis=1)

            # Filter for today's predictions
            today_str = datetime.now().strftime('%Y-%m-%d')
            todays_predictions = predictions_df[predictions_df['prediction_date'] == today_str]

            # Remove the 'prediction_id' column if it exists
            if 'prediction_id' in todays_predictions.columns:
                todays_predictions = todays_predictions.drop(columns=['prediction_id', 'prediction_date'])

            st.dataframe(todays_predictions, hide_index=True)
        else:
            st.write("No predictions for today's matches yet.")


def display_leaderboard():
    if st.button("Show Leaderboard"):
        try:
            # # Load the 'leaders' configuration
            dataset = load_dataset("Jay-Rajput/DIS_IPL_Leads", split='train')

            users_data = []
            if dataset:
                for user, points_dict in dataset[0].items():
                    points = points_dict.get("points", 0)
                    last_5_results = " ".join(points_dict.get("last_5_results", ["⚪"] * 5))  # Default: 5 white circles
                    users_data.append({'User': user, 'Points': points, "Last 5 Bids": last_5_results})
            else:
                data = load_users(USERS_JSON)
                for user, points_dict in data.items():
                    points = points_dict.get("points", 0)
                    last_5_results = " ".join(points_dict.get("last_5_results", ["⚪"] * 5))  # Default: 5 white circles
                    users_data.append({'User': user, 'Points': points, "Last 5 Bids": last_5_results})

            leaderboard = pd.DataFrame(users_data)
            
            # Sort DataFrame by points in descending order
            leaderboard = leaderboard.sort_values(by='Points', ascending=False)
            
            # Add a 'Rank' column starting from 1
            leaderboard['Rank'] = range(1, len(leaderboard) + 1)
            
            # Select and order the columns for display
            leaderboard = leaderboard[['Rank', 'User', 'Points', 'Last 5 Bids']]
            
            st.dataframe(leaderboard, hide_index=True)
        except Exception as e:
            st.write("Failed to load leaderboard data: ", str(e))


# Streamlit UI
encoded_image = get_base64_of_image(image_path)
custom_css = f"""
<style>
.header {{
    font-size: 50px;
    color: #FFD700; /* Gold */
    text-shadow: -1px -1px 0 #000, 1px -1px 0 #000, -1px 1px 0 #000, 1px 1px 0 #000; /* Black text shadow */
    text-align: center;
    padding: 10px;
    background-image: url('data:image/png;base64,{encoded_image}');
    background-size: cover;
}}
</style>
"""
# Apply custom CSS
st.markdown(custom_css, unsafe_allow_html=True)
# Use the custom class in a div with your title
st.markdown('<div class="header">DIS IPL Match Predictions</div>', unsafe_allow_html=True)

st.write("🏆 Predict, Compete, and Win 🏏 - Where Every Guess Counts! 🏆")

user_guide_content = """
### 📘 User Guide

#### Submitting Predictions
- **Match Selection**: Choose the match you want to predict from today's available matches.
- **Team and Player Prediction**: Select the team you predict will win and the "Man of the Match".
- **Bid Points**: Enter the number of points you wish to bid on your prediction. Remember, the maximum you can bid is capped at **20% of your total points**.

#### Scoring System
- **Winning Team Prediction**:  
  - ✅ **Correct Prediction**: You earn **2000 points** plus your bid amount.  
  - ❌ **Incorrect Prediction**: You lose **200 points** plus your bid amount.  
- **Man of the Match Prediction**:  
  - ✅ **Correct Prediction**: You earn **an additional 500 points**.  
  - ❌ **Incorrect Prediction**: No penalty.  
- **No Prediction Submitted**:  
  - ❌ **You lose 1000 points** automatically for not submitting a prediction. 

#### Bid Point Constraints
- You cannot bid more than 20% of your current total points.
- Bid points will be doubled if your prediction is correct, and deducted if incorrect.

#### Rules for Submission
- **Predictions must be submitted before the match starts**.
- **Only one prediction per match is allowed**.
- **Review your prediction carefully before submission, as it cannot be changed once submitted**.

#### 🔴🟢⚪ Match Performance Tracking
- After each match, your last **5 predictions will be tracked** and displayed on the leaderboard:  
  - 🟢 **Green** → Correct prediction.  
  - 🔴 **Red** → Wrong prediction.  
  - ⚪ **White** → No prediction submitted.  

🚀 **Compete, strategize, and climb the leaderboard!**
"""

# User Guide as an expander
with st.expander("User Guide 📘"):
    st.markdown(user_guide_content)

with st.expander("Submit Prediction 📝"):
    user_selection_and_prediction()

with st.expander("Predictions 🔍"):
    display_predictions()

with st.expander("Leaderboard 🏆"):
    display_leaderboard()


############################# Admin Panel ##################################
ADMIN_PASSPHRASE = "admin123"


def fetch_latest_predictions(match_id):
    # Load the dataset. Adjust "split" to "train" or appropriate if "predictions" is a configuration.
    dataset = load_dataset("Jay-Rajput/DIS_IPL_Preds", split="train")
    # Convert the dataset to a pandas DataFrame
    df = pd.DataFrame(dataset)
    # Ensure the DataFrame is not empty and contains the required columns
    if not df.empty and {'user_name', 'match_id'}.issubset(df.columns):
        # Filter rows by 'match_id'
        filtered_df = df[df['match_id'] == match_id]
        # Drop duplicate rows based on 'user_name'
        unique_df = filtered_df.drop_duplicates(subset=['user_name'])
        return unique_df
    else:
        return pd.DataFrame()


def update_leaderboard_and_outcomes(match_id, winning_team, man_of_the_match, outcome_only=False):
    features = Features({
        'match_id': Value('string'),
        'man_of_the_match': Value('string'),
        'winning_team': Value('string'),
    })

    # Load existing match outcomes
    outcomes = load_dataset("Jay-Rajput/DIS_IPL_Outcomes", split="train")  
    outcomes_df = pd.DataFrame(outcomes)

    # Directly update or add the match outcome
    outcome_exists = False
    for idx, outcome in outcomes_df.iterrows():
        if outcome['match_id'] == match_id:
            outcomes_df.at[idx, 'winning_team'] = winning_team
            outcomes_df.at[idx, 'man_of_the_match'] = man_of_the_match
            outcome_exists = True
            break
    if not outcome_exists:
        new_outcome = {"match_id": match_id, "winning_team": winning_team, "man_of_the_match": man_of_the_match}
        outcomes_df = pd.concat([outcomes_df, pd.DataFrame([new_outcome])], ignore_index=True)
        outcomes = Dataset.from_pandas(outcomes_df)

    if not outcome_only:  # Update user scores only if outcome_only is False
        # Load predictions only if necessary
        predictions = fetch_latest_predictions(match_id)

        # Load users' data only if necessary
        users = load_dataset("Jay-Rajput/DIS_IPL_Leads", split="train")
        users_df = pd.DataFrame(users)

        # Update user points based on prediction accuracy
        users_with_predictions = set(predictions['user_name'])
        for user_name in users_df.columns:
            user_points = users_df[user_name][0]['points']
            if user_name in users_with_predictions:
                prediction = predictions[predictions['user_name'] == user_name].iloc[0]
                predicted_winner = prediction['predicted_winner']
                predicted_motm = prediction['predicted_motm']
                bid_points = prediction['bid_points']

                # Update points based on prediction accuracy
                if predicted_winner == winning_team:
                    user_points += 2000 + bid_points
                    result_indicator = "🟢"  # Correct Prediction
                    if predicted_motm == man_of_the_match:
                        user_points += 500
                else:
                    user_points -= 200 + bid_points
                    result_indicator = "🔴"  # Wrong Prediction
            else:
                # Deduct 1000 points for not submitting a prediction
                user_points -= 1000
                result_indicator = "⚪"  # No Prediction
                
            # Ensure user_points is never negative
            user_points = max(user_points, 0)

            # Update user's points in the DataFrame
            users_df[user_name][0]['points'] = user_points
            users[user_name][0]['points'] = user_points
            
            # Maintain last 5 prediction results
            if "last_5_results" not in users_df[user_name][0]:
                users_df[user_name][0]["last_5_results"] = []
            users_df[user_name][0]["last_5_results"].insert(0, result_indicator)  # Insert at beginning
            users_df[user_name][0]["last_5_results"] = users_df[user_name][0]["last_5_results"][:5]  # Keep only last 5
            
            if "last_5_results" not in users[user_name][0]:
                users[user_name][0]["last_5_results"] = []
            users[user_name][0]["last_5_results"].insert(0, result_indicator)  # Insert at beginning
            users[user_name][0]["last_5_results"] = users[user_name][0]["last_5_results"][:5]  # Keep only last 5

        users.to_json(USERS_JSON)
        updated_dataset = Dataset.from_pandas(users_df)
        updated_dataset.push_to_hub("Jay-Rajput/DIS_IPL_Leads", split="train")

    outcomes.to_json(OUTCOMES)
    outcomes.push_to_hub("Jay-Rajput/DIS_IPL_Outcomes", split="train")


# Function to fetch matches for a given date
def fetch_matches_by_date(matches, selected_date):
    return [match for match in matches if datetime.strptime(match['date'], '%Y-%m-%d').date() == selected_date]


with st.sidebar:
    expander = st.expander("Admin Panel", expanded=False)
    admin_pass = expander.text_input("Enter admin passphrase:", type="password", key="admin_pass")

    if admin_pass == ADMIN_PASSPHRASE:
        expander.success("Authenticated")

        all_matches = load_data(MATCHES_JSON)
        match_outcomes = load_dataset("Jay-Rajput/DIS_IPL_Outcomes", split="train")
        submitted_match_ids = [outcome["match_id"] for outcome in match_outcomes]

        # Filter matches to those that do not have outcomes submitted yet
        matches_without_outcomes = [match for match in all_matches if match["match_id"] not in submitted_match_ids]

        # If matches are available, let the admin select one
        if matches_without_outcomes:
            # Optional: Allow the admin to filter matches by date
            selected_date = expander.date_input("Select Match Date", key="match_date")
            if selected_date:
                filtered_matches = fetch_matches_by_date(matches_without_outcomes, selected_date)
            else:
                filtered_matches = matches_without_outcomes

            if filtered_matches:
                match_selection = expander.selectbox("Select Match", filtered_matches, format_func=lambda match: f"{match['teams'][0]} vs {match['teams'][1]} (Match ID: {match['match_id']})", key="match_selection")
                selected_match_id = match_selection['match_id']
                teams = match_selection['teams']

                # Let admin select the winning team
                winning_team = expander.selectbox("Winning Team", teams, key="winning_team")

                # Fetch and display players for the selected winning team
                player_list = load_data(PLAYERS_JSON)
                if winning_team in player_list:
                    players = player_list[winning_team]
                    man_of_the_match = expander.selectbox("Man of the Match", players, key="man_of_the_match")
                else:
                    players = []
                    man_of_the_match = expander.text_input("Man of the Match (Type if not listed)", key="man_of_the_match_fallback")

                # Add checkbox for outcome only submission
                outcome_only = expander.checkbox("Submit Outcome Only", key="outcome_only_checkbox")

                if expander.button("Submit Match Outcome", key="submit_outcome"):
                    if outcome_only:
                        # Submit match outcome without updating user scores
                        update_leaderboard_and_outcomes(selected_match_id, winning_team, man_of_the_match, outcome_only=True)
                        expander.success("Match outcome submitted!")
                    else:
                        # Submit match outcome and update user scores
                        update_leaderboard_and_outcomes(selected_match_id, winning_team, man_of_the_match)
                        expander.success("Match outcome submitted and leaderboard updated!")
            else:
                expander.write("No matches available for the selected date.")
        else:
            expander.write("No matches are available for today.")
    else:
        if admin_pass:  # Show error only if something was typed
            expander.error("Not authenticated")