extract score calculation to new script

app.py

@@ -10,453 +10,9 @@ import pandas as pd
 from dotenv import load_dotenv
 from fastapi import FastAPI
 
-load_dotenv()
-
-# Constants
-DATA_DIR = "data"
-PARTICIPANTS_CSV = os.path.join(DATA_DIR, "participants.csv")
-EQUIPOS_CSV = os.path.join(DATA_DIR, "equipos.csv")
-LEADERBOARD_PERSONAL_CSV = "leaderboard_personal.csv"
-LEADERBOARD_EQUIPOS_CSV = "leaderboard_equipos.csv"
-
-# Column mappings for participants info
-COLUMN_MAP = {
-    "gmail": "Dirección de correo electrónico",
-    "discord": "¿Cuál es tu nombre en Discord?",
-    "hf_username": "¿Cuál es tu nombre en el Hub de Hugging Face?",
-    "contact_email": "Email de contacto",
-}
-
-# Column mappings for teams info
-TEAM_COLUMNS = {
-    "team_name": "Nombre del equipo",
-    "email_1": "Email 1",
-    "email_2": "Email 2",
-    "email_3": "Email 3",
-    "email_4": "Email 4",
-    "email_5": "Email 5",
-}
-
-# Initialize Argilla client
-try:
-    client = rg.Argilla(
-        api_url=os.getenv("ARGILLA_API_URL", ""),
-        api_key=os.getenv("ARGILLA_API_KEY", ""),
-    )
-except Exception as e:
-    print(f"Error initializing Argilla client: {e}")
-    client = None
-
-# Countries data
-countries = {
-    "Argentina": {"iso": "ARG", "emoji": "🇦🇷"},
-    "Bolivia": {"iso": "BOL", "emoji": "🇧🇴"},
-    "Chile": {"iso": "CHL", "emoji": "🇨🇱"},
-    "Colombia": {"iso": "COL", "emoji": "🇨🇴"},
-    "Costa Rica": {"iso": "CRI", "emoji": "🇨🇷"},
-    "Cuba": {"iso": "CUB", "emoji": "🇨🇺"},
-    "Ecuador": {"iso": "ECU", "emoji": "🇪🇨"},
-    "El Salvador": {"iso": "SLV", "emoji": "🇸🇻"},
-    "España": {"iso": "ESP", "emoji": "🇪🇸"},
-    "Guatemala": {"iso": "GTM", "emoji": "🇬🇹"},
-    "Honduras": {"iso": "HND", "emoji": "🇭🇳"},
-    "México": {"iso": "MEX", "emoji": "🇲🇽"},
-    "Nicaragua": {"iso": "NIC", "emoji": "🇳🇮"},
-    "Panamá": {"iso": "PAN", "emoji": "🇵🇦"},
-    "Paraguay": {"iso": "PRY", "emoji": "🇵🇾"},
-    "Perú": {"iso": "PER", "emoji": "🇵🇪"},
-    "Puerto Rico": {"iso": "PRI", "emoji": "🇵🇷"},
-    "República Dominicana": {"iso": "DOM", "emoji": "🇩🇴"},
-    "Uruguay": {"iso": "URY", "emoji": "🇺🇾"},
-    "Venezuela": {"iso": "VEN", "emoji": "🇻🇪"},
-}
-
-
-@lru_cache(maxsize=1)
-def get_user_mapping():
-    """Get cached mapping of emails and hf_usernames to discord usernames."""
-    if not os.path.exists(PARTICIPANTS_CSV):
-        return {}, {}
-
-    try:
-        df = pd.read_csv(PARTICIPANTS_CSV)
-        email_to_discord = {}
-        hf_to_discord = {}
-
-        for _, row in df.iterrows():
-            discord = row.get(COLUMN_MAP["discord"], "")
-            if pd.notna(discord) and discord != "NA":
-                discord_lower = discord.lower()
-
-                # Map email to discord
-                gmail = row.get(COLUMN_MAP["gmail"], "")
-                if pd.notna(gmail):
-                    email_to_discord[gmail.lower()] = discord_lower
-
-                # Map hf_username to discord
-                hf_username = row.get(COLUMN_MAP["hf_username"], "")
-                if pd.notna(hf_username):
-                    hf_to_discord[hf_username.lower()] = discord_lower
-
-        return email_to_discord, hf_to_discord
-    except Exception as e:
-        print(f"Error loading {PARTICIPANTS_CSV}: {e}")
-        return {}, {}
-
-
-def get_discord_username(identifier):
-    """Get discord username from email or hf_username."""
-    email_to_discord, hf_to_discord = get_user_mapping()
-
-    if "@" in identifier:
-        return email_to_discord.get(identifier.lower(), identifier.split("@")[0])
-
-    return hf_to_discord.get(identifier.lower(), identifier)
-
-
-def get_participant_info():
-    """Get participant information from CSV."""
-    if not os.path.exists(PARTICIPANTS_CSV):
-        return {}
-
-    try:
-        df = pd.read_csv(PARTICIPANTS_CSV)
-        participant_info = {}
-
-        for _, row in df.iterrows():
-            discord_username = row.get(COLUMN_MAP["discord"], "")
-            if pd.notna(discord_username) and discord_username != "NA":
-                participant_info[discord_username.lower()] = {
-                    "gmail": row.get(COLUMN_MAP["gmail"], ""),
-                    "discord_username": discord_username,
-                    "hf_username": row.get(COLUMN_MAP["hf_username"], ""),
-                    "email": row.get(COLUMN_MAP["contact_email"], ""),
-                }
-
-        return participant_info
-    except Exception as e:
-        print(f"Error loading participant info: {e}")
-        return {}
-
-
-def get_team_leaderboard(personal_leaderboard_df):
-    """Calculate team leaderboard based on personal scores."""
-    if not os.path.exists(EQUIPOS_CSV):
-        return pd.DataFrame()
-
-    try:
-        teams_df = pd.read_csv(EQUIPOS_CSV)
-        team_leaderboard = []
-
-        for _, team_row in teams_df.iterrows():
-            team_name = team_row.get(TEAM_COLUMNS["team_name"], "")
-            if not team_name:
-                continue
-
-            # Get team member emails
-            team_emails = []
-            for i in range(1, 6):
-                email_col = TEAM_COLUMNS[f"email_{i}"]
-                email = team_row.get(email_col, "")
-                if pd.notna(email) and email.strip():
-                    team_emails.append(email.lower())
-
-            if not team_emails:
-                continue
-
-            # Map emails to Discord usernames and get scores
-            discord_usernames = []
-            team_scores = {"arena": 0, "blend_es": 0, "estereotipos": 0, "include": 0}
-
-            for email in team_emails:
-                # Get Discord username from email
-                discord_username = get_discord_username(email)
-                discord_usernames.append(discord_username)
-
-                # Find this user in the personal leaderboard
-                user_scores = personal_leaderboard_df[
-                    personal_leaderboard_df["Username"].str.lower()
-                    == discord_username.lower()
-                ]
-
-                if not user_scores.empty:
-                    team_scores["arena"] += user_scores.iloc[0]["Arena"]
-                    team_scores["blend_es"] += user_scores.iloc[0]["Blend-ES"]
-                    team_scores["estereotipos"] += user_scores.iloc[0]["Estereotipos"]
-                    team_scores["include"] += user_scores.iloc[0]["INCLUDE"]
-
-            # Pad Discord usernames list to 5 elements
-            while len(discord_usernames) < 5:
-                discord_usernames.append("")
-
-            # Create team row
-            team_row_data = {
-                "team_name": team_name,
-                "discord_1": discord_usernames[0],
-                "discord_2": discord_usernames[1],
-                "discord_3": discord_usernames[2],
-                "discord_4": discord_usernames[3],
-                "discord_5": discord_usernames[4],
-                "total_arena": team_scores["arena"],
-                "ptos_arena": 0,  # Set to 0 for now as requested
-                "total_blend_es": team_scores["blend_es"],
-                "ptos_blend_es": 0,  # Set to 0 for now as requested
-                "total_estereotipos": team_scores["estereotipos"],
-                "ptos_estereotipos": 0,  # Set to 0 for now as requested
-                "total_include": team_scores["include"],
-                "ptos_include": 0,  # Set to 0 for now as requested
-                "ptos_total": 0,  # Set to 0 for now as requested
-            }
-
-            team_leaderboard.append(team_row_data)
-
-        # Create DataFrame and sort by total_arena
-        if team_leaderboard:
-            team_df = pd.DataFrame(team_leaderboard)
-            team_df.sort_values("total_arena", ascending=False, inplace=True)
-            return team_df
-        else:
-            return pd.DataFrame()
-
-    except Exception as e:
-        print(f"Error calculating team leaderboard: {e}")
-        return pd.DataFrame()
-
-
-def get_blend_es_data():
-    """Get blend-es data from Argilla."""
-    if not client:
-        return []
-
-    data = []
-    for country, info in countries.items():
-        dataset_name = f"{info['emoji']} {country} - {info['iso']} - Responder"
-
-        try:
-            dataset = client.datasets(dataset_name)
-            records = list(dataset.records(with_responses=True))
-
-            user_counts = defaultdict(int)
-            user_mapping = {}
-
-            for record in records:
-                if "answer_1" in record.responses:
-                    for answer in record.responses["answer_1"]:
-                        if answer.user_id:
-                            user_id = answer.user_id
-                            user_counts[user_id] += 1
-
-                            if user_id not in user_mapping:
-                                try:
-                                    user = client.users(id=user_id)
-                                    user_mapping[user_id] = user.username
-                                except:
-                                    user_mapping[user_id] = f"User-{user_id[:8]}"
-
-            for user_id, count in user_counts.items():
-                hf_username = user_mapping.get(user_id, f"User-{user_id[:8]}")
-                username = get_discord_username(hf_username)
-                data.append(
-                    {"source": "blend-es", "username": username, "count": count}
-                )
-
-        except Exception as e:
-            print(f"Error processing {dataset_name}: {e}")
-
-    return data
-
-
-def get_include_data():
-    """Get include data from CSV."""
-    csv_path = os.path.join(DATA_DIR, "include.csv")
-    if not os.path.exists(csv_path):
-        return []
-
-    try:
-        df = pd.read_csv(csv_path)
-        username_col = "Nombre en Discord / username"
-        questions_col = "Total preguntas hackathon"
-
-        if username_col not in df.columns or questions_col not in df.columns:
-            return []
-
-        user_counts = defaultdict(int)
-        for _, row in df.iterrows():
-            username = row[username_col][1:] if pd.notna(row[username_col]) else ""
-            questions = row[questions_col] if pd.notna(row[questions_col]) else 0
-            if username and questions:
-                user_counts[username.lower()] += int(questions)
-
-        return [
-            {"source": "include", "username": username, "count": count}
-            for username, count in user_counts.items()
-        ]
-    except Exception as e:
-        print(f"Error loading include data: {e}")
-        return []
-
-
-def get_estereotipos_data():
-    """Get estereotipos data from CSV."""
-    csv_path = os.path.join(DATA_DIR, "stereotypes.csv")
-    if not os.path.exists(csv_path):
-        return []
-
-    try:
-        df = pd.read_csv(csv_path)
-        if "token_id" not in df.columns or "count" not in df.columns:
-            return []
-
-        user_counts = defaultdict(int)
-        for _, row in df.iterrows():
-            mail = row.get("token_id", "")
-            count = row.get("count", 0)
-            if pd.notna(mail) and pd.notna(count):
-                user_counts[mail.lower()] += int(count)
-
-        return [
-            {
-                "source": "include",
-                "username": get_discord_username(mail),
-                "count": count,
-            }
-            for mail, count in user_counts.items()
-        ]
-    except Exception as e:
-        print(f"Error loading estereotipos data: {e}")
-        return []
-
-
-def get_arena_data():
-    """Get arena data from JSON."""
-    json_path = os.path.join(DATA_DIR, "arena.json")
-    if not os.path.exists(json_path):
-        return []
-
-    try:
-        with open(json_path, "r", encoding="utf-8") as f:
-            arena_data = json.load(f)
-
-        user_counts = defaultdict(int)
-        for conversations in arena_data.values():
-            for conversation in conversations:
-                if username := conversation.get("username"):
-                    user_counts[username.lower()] += 1
-
-        return [
-            {"source": "arena", "username": get_discord_username(mail), "count": count}
-            for mail, count in user_counts.items()
-        ]
-    except Exception as e:
-        print(f"Error loading arena data: {e}")
-        return []
-
-
-def consolidate_all_data():
-    """Consolidate all data sources and create leaderboard."""
-    # Collect all data
-    all_data = (
-        get_blend_es_data()
-        + get_include_data()
-        + get_estereotipos_data()
-        + get_arena_data()
-    )
-
-    # Get participant info
-    participant_info = get_participant_info()
-
-    # Aggregate user contributions
-    user_contributions = defaultdict(
-        lambda: {
-            "username": "",
-            "gmail": "",
-            "discord_username": "",
-            "hf_username": "",
-            "email": "",
-            "blend_es": 0,
-            "include": 0,
-            "estereotipos": 0,
-            "arena": 0,
-        }
-    )
-
-    for item in all_data:
-        source = item["source"]
-        username = item["username"]
-        count = item["count"]
-        user_key = username.lower()
-
-        if not user_contributions[user_key]["username"]:
-            user_contributions[user_key]["username"] = username
-            if username.lower() in participant_info:
-                info = participant_info[username.lower()]
-                user_contributions[user_key].update(
-                    {
-                        "gmail": info["gmail"],
-                        "discord_username": info["discord_username"],
-                        "hf_username": info["hf_username"],
-                        "email": info["email"],
-                    }
-                )
-
-        if source == "blend-es":
-            user_contributions[user_key]["blend_es"] += count
-        elif source == "include":
-            user_contributions[user_key]["include"] += count
-        elif source == "estereotipos":
-            user_contributions[user_key]["estereotipos"] += count
-        elif source == "arena":
-            user_contributions[user_key]["arena"] += count
-
-    # Create dataframes
-    full_rows = []
-    display_rows = []
-
-    for data in user_contributions.values():
-        # Full data for CSV
-        full_rows.append(
-            {
-                "Username": data["username"],
-                "Gmail": data["gmail"],
-                "Discord_Username": data["discord_username"],
-                "HF_Username": data["hf_username"],
-                "Email": data["email"],
-                "Arena": data["arena"],
-                "Blend-ES": data["blend_es"],
-                "Estereotipos": data["estereotipos"],
-                "INCLUDE": data["include"],
-            }
-        )
-
-        # Display data for UI (public)
-        display_rows.append(
-            {
-                "Username": data["username"],
-                "Arena": data["arena"],
-                "Blend-ES": data["blend_es"],
-                "Estereotipos": data["estereotipos"],
-                "INCLUDE": data["include"],
-            }
-        )
-
-    # Save full data to CSV
-    full_df = pd.DataFrame(full_rows)
-    if not full_df.empty:
-        full_df.sort_values("Arena", ascending=False, inplace=True)
-        full_df.to_csv(LEADERBOARD_PERSONAL_CSV, index=False, encoding="utf-8")
-
-    # Generate and save team leaderboard
-    team_df = get_team_leaderboard(full_df)
-    if not team_df.empty:
-        team_df.to_csv(LEADERBOARD_EQUIPOS_CSV, index=False, encoding="utf-8")
-
-    # Return display dataframe for UI
-    display_df = pd.DataFrame(display_rows)
-    if not display_df.empty:
-        display_df.sort_values("Arena", ascending=False, inplace=True)
-
-    return display_df
+from calculate_scores import calculate_scores
 
+load_dotenv()
 
 # FastAPI app
 app = FastAPI()
@@ -474,7 +30,7 @@ def create_leaderboard_ui():
     if cached_data is not None and current_time - last_update_time < 300:
         df = cached_data
     else:
-        df = consolidate_all_data()
+        df = calculate_scores()
         cached_data = df
         last_update_time = current_time
 

calculate_scores.py

@@ -0,0 +1,462 @@
+import json
+import os
+import time
+from collections import defaultdict
+from functools import lru_cache
+
+import argilla as rg
+import gradio as gr
+import pandas as pd
+from dotenv import load_dotenv
+from fastapi import FastAPI
+
+load_dotenv()
+
+# Constants
+DATA_DIR = "data"
+PARTICIPANTS_CSV = os.path.join(DATA_DIR, "participants.csv")
+EQUIPOS_CSV = os.path.join(DATA_DIR, "equipos.csv")
+LEADERBOARD_PERSONAL_CSV = "leaderboard_personal.csv"
+LEADERBOARD_EQUIPOS_CSV = "leaderboard_equipos.csv"
+
+# Column mappings for participants info
+COLUMN_MAP = {
+    "gmail": "Dirección de correo electrónico",
+    "discord": "¿Cuál es tu nombre en Discord?",
+    "hf_username": "¿Cuál es tu nombre en el Hub de Hugging Face?",
+    "contact_email": "Email de contacto",
+}
+
+# Column mappings for teams info
+TEAM_COLUMNS = {
+    "team_name": "Nombre del equipo",
+    "email_1": "Email 1",
+    "email_2": "Email 2",
+    "email_3": "Email 3",
+    "email_4": "Email 4",
+    "email_5": "Email 5",
+}
+
+# Initialize Argilla client
+try:
+    client = rg.Argilla(
+        api_url=os.getenv("ARGILLA_API_URL", ""),
+        api_key=os.getenv("ARGILLA_API_KEY", ""),
+    )
+except Exception as e:
+    print(f"Error initializing Argilla client: {e}")
+    client = None
+
+# Countries data
+countries = {
+    "Argentina": {"iso": "ARG", "emoji": "🇦🇷"},
+    "Bolivia": {"iso": "BOL", "emoji": "🇧🇴"},
+    "Chile": {"iso": "CHL", "emoji": "🇨🇱"},
+    "Colombia": {"iso": "COL", "emoji": "🇨🇴"},
+    "Costa Rica": {"iso": "CRI", "emoji": "🇨🇷"},
+    "Cuba": {"iso": "CUB", "emoji": "🇨🇺"},
+    "Ecuador": {"iso": "ECU", "emoji": "🇪🇨"},
+    "El Salvador": {"iso": "SLV", "emoji": "🇸🇻"},
+    "España": {"iso": "ESP", "emoji": "🇪🇸"},
+    "Guatemala": {"iso": "GTM", "emoji": "🇬🇹"},
+    "Honduras": {"iso": "HND", "emoji": "🇭🇳"},
+    "México": {"iso": "MEX", "emoji": "🇲🇽"},
+    "Nicaragua": {"iso": "NIC", "emoji": "🇳🇮"},
+    "Panamá": {"iso": "PAN", "emoji": "🇵🇦"},
+    "Paraguay": {"iso": "PRY", "emoji": "🇵🇾"},
+    "Perú": {"iso": "PER", "emoji": "🇵🇪"},
+    "Puerto Rico": {"iso": "PRI", "emoji": "🇵🇷"},
+    "República Dominicana": {"iso": "DOM", "emoji": "🇩🇴"},
+    "Uruguay": {"iso": "URY", "emoji": "🇺🇾"},
+    "Venezuela": {"iso": "VEN", "emoji": "🇻🇪"},
+}
+
+
+@lru_cache(maxsize=1)
+def get_user_mapping():
+    """Get cached mapping of emails and hf_usernames to discord usernames."""
+    if not os.path.exists(PARTICIPANTS_CSV):
+        return {}, {}
+
+    try:
+        df = pd.read_csv(PARTICIPANTS_CSV)
+        email_to_discord = {}
+        hf_to_discord = {}
+
+        for _, row in df.iterrows():
+            discord = row.get(COLUMN_MAP["discord"], "")
+            if pd.notna(discord) and discord != "NA":
+                discord_lower = discord.lower()
+
+                # Map email to discord
+                gmail = row.get(COLUMN_MAP["gmail"], "")
+                if pd.notna(gmail):
+                    email_to_discord[gmail.lower()] = discord_lower
+
+                # Map hf_username to discord
+                hf_username = row.get(COLUMN_MAP["hf_username"], "")
+                if pd.notna(hf_username):
+                    hf_to_discord[hf_username.lower()] = discord_lower
+
+        return email_to_discord, hf_to_discord
+    except Exception as e:
+        print(f"Error loading {PARTICIPANTS_CSV}: {e}")
+        return {}, {}
+
+
+def get_discord_username(identifier):
+    """Get discord username from email or hf_username."""
+    email_to_discord, hf_to_discord = get_user_mapping()
+
+    if "@" in identifier:
+        return email_to_discord.get(identifier.lower(), identifier.split("@")[0])
+
+    return hf_to_discord.get(identifier.lower(), identifier)
+
+
+def get_participant_info():
+    """Get participant information from CSV."""
+    if not os.path.exists(PARTICIPANTS_CSV):
+        return {}
+
+    try:
+        df = pd.read_csv(PARTICIPANTS_CSV)
+        participant_info = {}
+
+        for _, row in df.iterrows():
+            discord_username = row.get(COLUMN_MAP["discord"], "")
+            if pd.notna(discord_username) and discord_username != "NA":
+                participant_info[discord_username.lower()] = {
+                    "gmail": row.get(COLUMN_MAP["gmail"], ""),
+                    "discord_username": discord_username,
+                    "hf_username": row.get(COLUMN_MAP["hf_username"], ""),
+                    "email": row.get(COLUMN_MAP["contact_email"], ""),
+                }
+
+        return participant_info
+    except Exception as e:
+        print(f"Error loading participant info: {e}")
+        return {}
+
+
+def get_team_leaderboard(personal_leaderboard_df):
+    """Calculate team leaderboard based on personal scores."""
+    if not os.path.exists(EQUIPOS_CSV):
+        return pd.DataFrame()
+
+    try:
+        teams_df = pd.read_csv(EQUIPOS_CSV)
+        team_leaderboard = []
+
+        for _, team_row in teams_df.iterrows():
+            team_name = team_row.get(TEAM_COLUMNS["team_name"], "")
+            if not team_name:
+                continue
+
+            # Get team member emails
+            team_emails = []
+            for i in range(1, 6):
+                email_col = TEAM_COLUMNS[f"email_{i}"]
+                email = team_row.get(email_col, "")
+                if pd.notna(email) and email.strip():
+                    team_emails.append(email.lower())
+
+            if not team_emails:
+                continue
+
+            # Map emails to Discord usernames and get scores
+            discord_usernames = []
+            team_scores = {"arena": 0, "blend_es": 0, "estereotipos": 0, "include": 0}
+
+            for email in team_emails:
+                # Get Discord username from email
+                discord_username = get_discord_username(email)
+                discord_usernames.append(discord_username)
+
+                # Find this user in the personal leaderboard
+                user_scores = personal_leaderboard_df[
+                    personal_leaderboard_df["Username"].str.lower()
+                    == discord_username.lower()
+                ]
+
+                if not user_scores.empty:
+                    team_scores["arena"] += user_scores.iloc[0]["Arena"]
+                    team_scores["blend_es"] += user_scores.iloc[0]["Blend-ES"]
+                    team_scores["estereotipos"] += user_scores.iloc[0]["Estereotipos"]
+                    team_scores["include"] += user_scores.iloc[0]["INCLUDE"]
+
+            # Pad Discord usernames list to 5 elements
+            while len(discord_usernames) < 5:
+                discord_usernames.append("")
+
+            # Create team row
+            team_row_data = {
+                "team_name": team_name,
+                "discord_1": discord_usernames[0],
+                "discord_2": discord_usernames[1],
+                "discord_3": discord_usernames[2],
+                "discord_4": discord_usernames[3],
+                "discord_5": discord_usernames[4],
+                "total_arena": team_scores["arena"],
+                "ptos_arena": 0,  # Set to 0 for now as requested
+                "total_blend_es": team_scores["blend_es"],
+                "ptos_blend_es": 0,  # Set to 0 for now as requested
+                "total_estereotipos": team_scores["estereotipos"],
+                "ptos_estereotipos": 0,  # Set to 0 for now as requested
+                "total_include": team_scores["include"],
+                "ptos_include": 0,  # Set to 0 for now as requested
+                "ptos_total": 0,  # Set to 0 for now as requested
+            }
+
+            team_leaderboard.append(team_row_data)
+
+        # Create DataFrame and sort by total_arena
+        if team_leaderboard:
+            team_df = pd.DataFrame(team_leaderboard)
+            team_df.sort_values("total_arena", ascending=False, inplace=True)
+            return team_df
+        else:
+            return pd.DataFrame()
+
+    except Exception as e:
+        print(f"Error calculating team leaderboard: {e}")
+        return pd.DataFrame()
+
+
+def get_blend_es_data():
+    """Get blend-es data from Argilla."""
+    if not client:
+        return []
+
+    data = []
+    for country, info in countries.items():
+        dataset_name = f"{info['emoji']} {country} - {info['iso']} - Responder"
+
+        try:
+            dataset = client.datasets(dataset_name)
+            records = list(dataset.records(with_responses=True))
+
+            user_counts = defaultdict(int)
+            user_mapping = {}
+
+            for record in records:
+                if "answer_1" in record.responses:
+                    for answer in record.responses["answer_1"]:
+                        if answer.user_id:
+                            user_id = answer.user_id
+                            user_counts[user_id] += 1
+
+                            if user_id not in user_mapping:
+                                try:
+                                    user = client.users(id=user_id)
+                                    user_mapping[user_id] = user.username
+                                except:
+                                    user_mapping[user_id] = f"User-{user_id[:8]}"
+
+            for user_id, count in user_counts.items():
+                hf_username = user_mapping.get(user_id, f"User-{user_id[:8]}")
+                username = get_discord_username(hf_username)
+                data.append(
+                    {"source": "blend-es", "username": username, "count": count}
+                )
+
+        except Exception as e:
+            print(f"Error processing {dataset_name}: {e}")
+
+    return data
+
+
+def get_include_data():
+    """Get include data from CSV."""
+    csv_path = os.path.join(DATA_DIR, "include.csv")
+    if not os.path.exists(csv_path):
+        return []
+
+    try:
+        df = pd.read_csv(csv_path)
+        username_col = "Nombre en Discord / username"
+        questions_col = "Total preguntas hackathon"
+
+        if username_col not in df.columns or questions_col not in df.columns:
+            return []
+
+        user_counts = defaultdict(int)
+        for _, row in df.iterrows():
+            username = row[username_col][1:] if pd.notna(row[username_col]) else ""
+            questions = row[questions_col] if pd.notna(row[questions_col]) else 0
+            if username and questions:
+                user_counts[username.lower()] += int(questions)
+
+        return [
+            {"source": "include", "username": username, "count": count}
+            for username, count in user_counts.items()
+        ]
+    except Exception as e:
+        print(f"Error loading include data: {e}")
+        return []
+
+
+def get_estereotipos_data():
+    """Get estereotipos data from CSV."""
+    csv_path = os.path.join(DATA_DIR, "stereotypes.csv")
+    if not os.path.exists(csv_path):
+        return []
+
+    try:
+        df = pd.read_csv(csv_path)
+        if "token_id" not in df.columns or "count" not in df.columns:
+            return []
+
+        user_counts = defaultdict(int)
+        for _, row in df.iterrows():
+            mail = row.get("token_id", "")
+            count = row.get("count", 0)
+            if pd.notna(mail) and pd.notna(count):
+                user_counts[mail.lower()] += int(count)
+
+        return [
+            {
+                "source": "include",
+                "username": get_discord_username(mail),
+                "count": count,
+            }
+            for mail, count in user_counts.items()
+        ]
+    except Exception as e:
+        print(f"Error loading estereotipos data: {e}")
+        return []
+
+
+def get_arena_data():
+    """Get arena data from JSON."""
+    json_path = os.path.join(DATA_DIR, "arena.json")
+    if not os.path.exists(json_path):
+        return []
+
+    try:
+        with open(json_path, "r", encoding="utf-8") as f:
+            arena_data = json.load(f)
+
+        user_counts = defaultdict(int)
+        for conversations in arena_data.values():
+            for conversation in conversations:
+                if username := conversation.get("username"):
+                    user_counts[username.lower()] += 1
+
+        return [
+            {"source": "arena", "username": get_discord_username(mail), "count": count}
+            for mail, count in user_counts.items()
+        ]
+    except Exception as e:
+        print(f"Error loading arena data: {e}")
+        return []
+
+
+def calculate_scores():
+    """Consolidate all data sources and create leaderboard."""
+    # Collect all data
+    all_data = (
+        get_blend_es_data()
+        + get_include_data()
+        + get_estereotipos_data()
+        + get_arena_data()
+    )
+
+    # Get participant info
+    participant_info = get_participant_info()
+
+    # Aggregate user contributions
+    user_contributions = defaultdict(
+        lambda: {
+            "username": "",
+            "gmail": "",
+            "discord_username": "",
+            "hf_username": "",
+            "email": "",
+            "blend_es": 0,
+            "include": 0,
+            "estereotipos": 0,
+            "arena": 0,
+        }
+    )
+
+    for item in all_data:
+        source = item["source"]
+        username = item["username"]
+        count = item["count"]
+        user_key = username.lower()
+
+        if not user_contributions[user_key]["username"]:
+            user_contributions[user_key]["username"] = username
+            if username.lower() in participant_info:
+                info = participant_info[username.lower()]
+                user_contributions[user_key].update(
+                    {
+                        "gmail": info["gmail"],
+                        "discord_username": info["discord_username"],
+                        "hf_username": info["hf_username"],
+                        "email": info["email"],
+                    }
+                )
+
+        if source == "blend-es":
+            user_contributions[user_key]["blend_es"] += count
+        elif source == "include":
+            user_contributions[user_key]["include"] += count
+        elif source == "estereotipos":
+            user_contributions[user_key]["estereotipos"] += count
+        elif source == "arena":
+            user_contributions[user_key]["arena"] += count
+
+    # Create dataframes
+    full_rows = []
+    display_rows = []
+
+    for data in user_contributions.values():
+        # Full data for CSV
+        full_rows.append(
+            {
+                "Username": data["username"],
+                "Gmail": data["gmail"],
+                "Discord_Username": data["discord_username"],
+                "HF_Username": data["hf_username"],
+                "Email": data["email"],
+                "Arena": data["arena"],
+                "Blend-ES": data["blend_es"],
+                "Estereotipos": data["estereotipos"],
+                "INCLUDE": data["include"],
+            }
+        )
+
+        # Display data for UI (public)
+        display_rows.append(
+            {
+                "Username": data["username"],
+                "Arena": data["arena"],
+                "Blend-ES": data["blend_es"],
+                "Estereotipos": data["estereotipos"],
+                "INCLUDE": data["include"],
+            }
+        )
+
+    # Save full data to CSV
+    full_df = pd.DataFrame(full_rows)
+    if not full_df.empty:
+        full_df.sort_values("Arena", ascending=False, inplace=True)
+        full_df.to_csv(LEADERBOARD_PERSONAL_CSV, index=False, encoding="utf-8")
+
+    # Generate and save team leaderboard
+    team_df = get_team_leaderboard(full_df)
+    if not team_df.empty:
+        team_df.to_csv(LEADERBOARD_EQUIPOS_CSV, index=False, encoding="utf-8")
+
+    # Return display dataframe for UI
+    display_df = pd.DataFrame(display_rows)
+    if not display_df.empty:
+        display_df.sort_values("Arena", ascending=False, inplace=True)
+
+    return display_df
+
+
+if __name__ == "__main__":
+    calculate_scores()