ullr-data-cli 1.0.1__tar.gz → 1.0.2__tar.gz

{ullr_data_cli-1.0.1 → ullr_data_cli-1.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ullr-data-cli
-Version: 1.0.1
+Version: 1.0.2
 Summary: An offline, AI-free Data Analytics and Dashboard Auditing CLI
 Home-page: https://github.com/jshlydnzl/ullr-data-cli
 Author: jshlydnzl

{ullr_data_cli-1.0.1 → ullr_data_cli-1.0.2}/README.md

@@ -36,6 +36,24 @@ Turn raw numbers into a full dashboard layout in seconds. Ullr calculates your k
 *   **Dynamic Chart Logic:** Automatically recommends Line Charts for dates, Bar Charts for categories, and Donut Charts for binary metrics.
 *   **Multi-Platform Mentor:** Generates step-by-step click instructions on how to build the recommended dashboard in **Microsoft Excel, Power BI, Google Looker Studio, and Tableau**.
 
+### 3. 🧹 Auto-Clean Engine (`ullr clean`)
+Automatically fixes the errors found during the audit phase. 
+*   **Ghost Purge:** Instantly deletes 100% empty columns and "Unnamed" structural columns.
+*   **Regex Stripping:** Acts as a Python `=TRIM()`, vaporizing invisible leading/trailing spaces.
+*   **Math Fixer:** Hunts down numbers trapped as text (currencies `$`, `£`, commas), rips out the symbols, and converts them to pure math-ready floats.
+*   **Multi-Format Export:** Pauses after cleaning to ask if you want to save the output as `.xlsx` (Dashboards), `.csv` (PostgreSQL bulk imports), or `.json` (Web Apps).
+
+### 4. 🗃️ Markdown Data Dictionary & Job Simulator (`ullr map`)
+Automatically parses your dataset to identify column data types, find Primary Keys, and generate a beautiful Markdown schema dictionary. 
+*   **Virtual Manager Brief:** Reads your specific columns and generates a **Simulated Stakeholder Brief** giving you exact instructions on what KPIs to calculate and what layout to build.
+*   **Auto-Documentation:** Saves the `.md` dictionary file directly next to your dataset so you never lose context.
+
+### 5. 🏭 Practice Data Engine (`ullr generate`)
+A built-in data synthesizer for building your portfolio.
+*   **Synthetic Generation:** Uses the `Faker` library to instantly generate up to 10,000+ rows of realistic E-commerce, Healthcare, or Real Estate data.
+*   **The Sabotage Engine:** Intentionally breaks the perfect data by injecting NULL values, invisible spaces, and numbers trapped as text, giving you the perfect dirty dataset to practice cleaning in Excel or SQL.
+*   **Dynamic Stakeholder Brief:** Automatically analyzes the generated dataset and prints a System Requirements Document, giving you specific KPIs and Business Questions to answer for your project scope.
+
 ---
 
 ## 📖 Quick Start

{ullr_data_cli-1.0.1 → ullr_data_cli-1.0.2}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name='ullr-data-cli',
-    version='1.0.1',
+    version='1.0.2',
     author='jshlydnzl',
     author_email='jshlydnzl@users.noreply.github.com',
     description='An offline, AI-free Data Analytics and Dashboard Auditing CLI',

{ullr_data_cli-1.0.1 → ullr_data_cli-1.0.2}/ullr/main.py

@@ -21,8 +21,11 @@ def print_banner():
     """
     console.print(banner, style="bold cyan")
 
+import pathlib
+
 def find_file_globally(filename):
-    home_dir = os.path.expanduser('~')
+    # pathlib.Path.home() natively handles Windows (C:\Users\Name), macOS (/Users/Name), and Linux (/home/Name)
+    home_dir = str(pathlib.Path.home())
     
     with console.status(f"[yellow]Scanning your computer for '{filename}'...[/]", spinner="dots"):
         for root, dirs, files in os.walk(home_dir):
@@ -317,6 +320,347 @@ def analyze_data(filepath):
                  console.print(f"      [dim]↳ Steps: Click 'Add a control' -> Select 'Drop-down list' -> Control field: your category.[/dim]")
     console.print("")
 
+def clean_data(filepath):
+    if not os.path.exists(filepath):
+        found_path = find_file_globally(os.path.basename(filepath))
+        if found_path:
+            filepath = found_path
+        else:
+            console.print(f"[bold red]Error:[/] Could not find '{os.path.basename(filepath)}' anywhere on your computer.")
+            return 
+        
+    df = load_dataframe(filepath)
+    if df is None:
+        return
+        
+    console.print("\n[bold yellow]🧹 INITIATING AUTO-CLEAN ENGINE...[/]")
+    time.sleep(0.5)
+    
+    # 1. Drop Ghost / Unnamed Columns
+    unnamed_cols = [col for col in df.columns if str(col).lower().startswith('unnamed')]
+    if unnamed_cols:
+        df = df.drop(columns=unnamed_cols)
+        console.print(f"  [green]✔[/] Dropped {len(unnamed_cols)} 'Unnamed' ghost columns.")
+        
+    # 2. Drop Completely Empty Columns
+    empty_cols = df.columns[df.isnull().all()].tolist()
+    if empty_cols:
+        df = df.drop(columns=empty_cols)
+        console.print(f"  [green]✔[/] Dropped {len(empty_cols)} completely empty columns.")
+        
+    # 3. Fix Text Columns (Invisible Spaces & Text Numbers)
+    text_cols = df.select_dtypes(include=['object']).columns
+    space_fixed = 0
+    dirty_num_fixed = 0
+    
+    for col in text_cols:
+        # Check if column looks like a dirty number (has currency/commas and digits)
+        curr_mask = df[col].notna() & df[col].astype(str).str.contains(r'[\$£€,]', regex=True) & df[col].astype(str).str.contains(r'\d', regex=True)
+        if curr_mask.any():
+            df[col] = df[col].astype(str).str.replace(r'[^\d.-]', '', regex=True)
+            df[col] = pd.to_numeric(df[col], errors='coerce')
+            dirty_num_fixed += 1
+            console.print(f"  [green]✔[/] Converted '{col}' from text to pure numbers.")
+            continue # Moved to numeric, skip string stripping
+            
+        # Strip invisible spaces for remaining text columns
+        mask = df[col].notna() & df[col].astype(str).str.contains(r'^\s+|\s+$', regex=True)
+        if mask.any():
+            df[col] = df[col].astype(str).str.strip()
+            space_fixed += 1
+            console.print(f"  [green]✔[/] Stripped invisible spaces from '{col}'.")
+            
+    # Prompt user for output format
+    from rich.prompt import Prompt
+    console.print("\n[bold cyan]💾 How would you like to save the cleaned data?[/]")
+    console.print("  [bold green]1.[/] Excel Workbook (.xlsx) - Best for Dashboards & formatting")
+    console.print("  [bold blue]2.[/] Raw CSV (.csv) - Best for PostgreSQL Bulk Imports")
+    console.print("  [bold magenta]3.[/] JSON (.json) - Best for Web Apps & APIs")
+    
+    save_choice = Prompt.ask("Select output format", choices=["1", "2", "3"], default="1")
+    base, ext = os.path.splitext(filepath)
+    
+    if save_choice == "1":
+        new_filepath = f"{base}_cleaned.xlsx"
+        with console.status(f"[cyan]● Saving to '{os.path.basename(new_filepath)}'...[/]", spinner="bouncingBar"):
+            df.to_excel(new_filepath, index=False)
+            time.sleep(0.6)
+    elif save_choice == "2":
+        new_filepath = f"{base}_cleaned.csv"
+        with console.status(f"[cyan]● Saving to '{os.path.basename(new_filepath)}'...[/]", spinner="bouncingBar"):
+            df.to_csv(new_filepath, index=False)
+            time.sleep(0.6)
+    else:
+        new_filepath = f"{base}_cleaned.json"
+        with console.status(f"[cyan]● Saving to '{os.path.basename(new_filepath)}'...[/]", spinner="bouncingBar"):
+            df.to_json(new_filepath, orient="records", indent=4)
+            time.sleep(0.6)
+            
+    console.print(f"\n[bold green]🎉 SUCCESS![/] Cleaned dataset saved to:\n[cyan]{new_filepath}[/]")
+
+
+
+def map_data(filepath):
+    if not os.path.exists(filepath):
+        found_path = find_file_globally(os.path.basename(filepath))
+        if found_path:
+            filepath = found_path
+        else:
+            console.print(f"[bold red]Error:[/] Could not find '{os.path.basename(filepath)}' anywhere on your computer.")
+            return 
+        
+    df = load_dataframe(filepath)
+    if df is None:
+        return
+        
+    console.print("\n[bold yellow]🗺️ INITIATING MARKDOWN DATA DICTIONARY (ULLR MAP)...[/]")
+    time.sleep(0.5)
+    
+    # 1. Identify Columns and Types
+    schema_data = []
+    primary_keys = []
+    
+    for col in df.columns:
+        dtype = str(df[col].dtype)
+        is_unique = df[col].is_unique
+        is_id = 'id' in str(col).lower()
+        
+        # Auto-detect Primary Key
+        pk_flag = ""
+        if is_unique and is_id:
+            pk_flag = "🔑 **PRIMARY KEY**"
+            primary_keys.append(col)
+        elif is_unique:
+            pk_flag = "Candidate Key"
+            
+        # Map pandas dtypes to business logic
+        business_type = dtype
+        if "object" in dtype:
+            business_type = "Text / Categorical"
+        elif "int" in dtype:
+            business_type = "Integer"
+        elif "float" in dtype:
+            business_type = "Decimal / Float"
+        elif "datetime" in dtype:
+            business_type = "Datetime"
+            
+        schema_data.append(f"| `{col}` | {business_type} | {pk_flag} |   |")
+        
+    # Generate Markdown Content
+    base_name = os.path.splitext(os.path.basename(filepath))[0]
+    md_content = f"# 🗃️ Data Dictionary: {base_name}\n\n"
+    md_content += f"**Source File:** `{os.path.basename(filepath)}`\n"
+    md_content += f"**Total Rows:** {len(df):,}\n"
+    md_content += f"**Total Columns:** {len(df.columns)}\n\n"
+    
+    if primary_keys:
+         md_content += f"**Detected Primary Key(s):** `{', '.join(primary_keys)}`\n\n"
+         
+    # Combined AI Stakeholder Brief Logic
+    real_cols = [col for col in df.columns if not str(col).lower().startswith('unnamed')]
+    numeric_cols = [c for c in df[real_cols].select_dtypes(include=['number']).columns.tolist() if 'id' not in c.lower() and 'zip' not in c.lower()]
+    text_cols = df[real_cols].select_dtypes(include=['object', 'category']).columns.tolist()
+    date_cols = [col for col in real_cols if 'date' in col.lower() or 'time' in col.lower() or 'year' in col.lower()]
+    
+    brief = "## 👔 Simulated Stakeholder Brief\n"
+    brief += "> *\"Hey team, I just dropped the new dataset in the folder. Before the weekly standup, I need you to clean this up and build a dashboard for me. Specifically, I want to see:\"\n>\n"
+    
+    req_num = 1
+    if numeric_cols:
+        brief += f"> **{req_num}. High-Level KPIs**: I need the total sum and average of `{numeric_cols[0]}`.\n"
+        req_num += 1
+    if date_cols and numeric_cols:
+        brief += f"> **{req_num}. Trend Analysis**: Build a line chart showing `{numeric_cols[0]}` tracking over time using `{date_cols[0]}`.\n"
+        req_num += 1
+    if text_cols and numeric_cols:
+        brief += f"> **{req_num}. Category Breakdown**: I need a visual showing the top performers in `{text_cols[0]}` based on `{numeric_cols[-1]}`.\n"
+        req_num += 1
+    if len(text_cols) > 1:
+        valid_slicers = [f'`{c}`' for c in text_cols if df[c].nunique() <= 20][:3]
+        if valid_slicers:
+            brief += f"> **{req_num}. Interactivity**: Make sure you include Slicers for {', '.join(valid_slicers)} so I can filter the dashboard myself.\n"
+            
+    brief += "> \n> **Recommended Dashboard Layout:**\n"
+    
+    if numeric_cols:
+        kpis = " and ".join([f"`Total {c}`" for c in numeric_cols[:2]])
+        brief += f"> - **Top (KPI Cards):** Put large, bold text showing {kpis} at the very top.\n"
+    if date_cols and numeric_cols:
+        brief += f"> - **Center (Line Chart):** Show the trend of `{numeric_cols[0]}` over `{date_cols[0]}`.\n"
+    if text_cols:
+        small_cats = [c for c in text_cols if 1 < df[c].nunique() <= 5]
+        if small_cats:
+            brief += f"> - **Bottom Left (Donut Chart):** Break down the percentage share of `{small_cats[0]}`.\n"
+        med_cats = [c for c in text_cols if 5 < df[c].nunique() <= 15]
+        if med_cats and numeric_cols:
+            brief += f"> - **Bottom Right (Bar Chart):** Rank the top `{med_cats[0]}` by `{numeric_cols[0]}`.\n"
+        valid_slicers = [f'`{c}`' for c in text_cols if df[c].nunique() <= 20][:3]
+        if valid_slicers:
+            brief += f"> - **Left Sidebar (Slicers):** Add clickable filters for {', '.join(valid_slicers)}.\n"
+            
+    brief += "> \n> *\"Let me know when the Excel file is ready for review!\"*\n\n"
+
+    md_content += brief
+    md_content += "## 📊 Schema Map\n\n"
+    md_content += "| Column Name | Data Type | Key Type | Description/Notes |\n"
+    md_content += "|---|---|---|---|\n"
+    for row in schema_data:
+        md_content += row + "\n"
+        
+    # Save next to the original file
+    file_dir = os.path.dirname(os.path.abspath(filepath))
+    out_path = os.path.join(file_dir, f"{base_name}_schema.md")
+    
+    with open(out_path, "w", encoding="utf-8") as f:
+        f.write(md_content)
+        
+    console.print(f"\n[bold green]🎉 SUCCESS![/] Markdown Data Dictionary generated and saved to:\n[cyan]{out_path}[/]")
+
+
+
+def generate_data():
+    from rich.prompt import Prompt
+    try:
+        from faker import Faker
+    except ImportError:
+        console.print("[bold red]Faker library not installed. Run 'pip install faker' first.[/]")
+        return
+        
+    import random
+    from datetime import datetime
+    import numpy as np
+    
+    fake = Faker()
+    
+    console.print("\n[bold yellow]🏭 INITIATING PRACTICE DATA ENGINE (ULLR GENERATE)...[/]")
+    time.sleep(0.5)
+    
+    console.print("\n[bold cyan]Select an Industry Context:[/]")
+    console.print("  [bold green]1.[/] E-commerce (Sales, Customers, Products)")
+    console.print("  [bold blue]2.[/] Healthcare (Patients, Treatments, Costs)")
+    console.print("  [bold magenta]3.[/] Real Estate (Properties, Agents, Prices)")
+    
+    ind_choice = Prompt.ask("Select an option", choices=["1", "2", "3"], default="1")
+    rows_choice = Prompt.ask("How many rows of dirty practice data do you need?", default="1000")
+    
+    try:
+        rows = int(rows_choice)
+    except:
+        rows = 1000
+        
+    with console.status(f"[cyan]● Fabricating {rows:,} rows of beautifully dirty data...[/]", spinner="dots"):
+        data = []
+        if ind_choice == "1":
+            industry = "ecommerce"
+            for i in range(rows):
+                data.append({
+                    "Transaction_ID": fake.uuid4()[:8],
+                    "Customer_Name": fake.name(),
+                    "Purchase_Date": fake.date_between(start_date='-1y', end_date='today').strftime('%Y-%m-%d'),
+                    "Product_Category": random.choice(["Electronics", "Clothing", "Home", "Sports"]),
+                    "Quantity": random.randint(1, 10),
+                    "Unit_Price": round(random.uniform(10.0, 500.0), 2)
+                })
+        elif ind_choice == "2":
+            industry = "healthcare"
+            for i in range(rows):
+                data.append({
+                    "Patient_ID": f"PT-{fake.random_int(min=1000, max=9999)}",
+                    "Patient_Name": fake.name(),
+                    "Admission_Date": fake.date_between(start_date='-2y', end_date='today').strftime('%Y-%m-%d'),
+                    "Department": random.choice(["Cardiology", "Neurology", "Oncology", "Pediatrics"]),
+                    "Length_of_Stay": random.randint(1, 30),
+                    "Treatment_Cost": round(random.uniform(500.0, 15000.0), 2)
+                })
+        else:
+            industry = "realestate"
+            for i in range(rows):
+                data.append({
+                    "Property_ID": f"RE-{fake.random_int(min=100, max=999)}",
+                    "Agent_Name": fake.name(),
+                    "Listing_Date": fake.date_between(start_date='-6m', end_date='today').strftime('%Y-%m-%d'),
+                    "Property_Type": random.choice(["House", "Condo", "Townhouse", "Commercial"]),
+                    "Square_Feet": random.randint(800, 5000),
+                    "Listing_Price": round(random.uniform(150000.0, 2000000.0), 2)
+                })
+                
+        df = pd.DataFrame(data)
+        
+        # --- INJECT DIRTY DATA ---
+        cols = list(df.columns)
+        
+        # 1. Nulls (Missing Values)
+        for col in cols:
+            mask = np.random.rand(len(df)) < 0.05  # 5% missing
+            df.loc[mask, col] = np.nan
+            
+        # 2. Invisible Spaces (Messy Text)
+        text_cols = [c for c in cols if 'Name' in c or 'Category' in c or 'Type' in c or 'Department' in c]
+        for col in text_cols:
+            mask = np.random.rand(len(df)) < 0.15 # 15% messy text
+            def add_spaces(x):
+                if pd.isna(x): return x
+                return ("   " + str(x)) if random.random() > 0.5 else (str(x) + "   ")
+            
+            df.loc[mask, col] = df.loc[mask, col].apply(add_spaces)
+            
+        # 3. Numbers Trapped as Text
+        num_cols = [c for c in cols if 'Price' in c or 'Cost' in c]
+        for col in num_cols:
+            df[col] = df[col].astype('object')
+            mask = np.random.rand(len(df)) < 0.20 # 20% trapped numbers
+            def add_currency(x):
+                if pd.isna(x): return x
+                return f"${x:,.2f}" if random.random() > 0.5 else f"£{x:,.2f}"
+                
+            df.loc[mask, col] = df.loc[mask, col].apply(add_currency)
+            
+        # 4. Ghost Column
+        df["Unnamed: 4"] = np.nan
+        
+        # Save File
+        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+        filename = f"{industry}_raw_{timestamp}.csv"
+        
+        out_dir = os.path.join(pathlib.Path.home(), "Projects", "eksel-praktis")
+        os.makedirs(out_dir, exist_ok=True)
+        out_path = os.path.join(out_dir, filename)
+        
+        df.to_csv(out_path, index=False)
+        time.sleep(1)
+        
+    console.print(f"\n[bold green]🎉 SUCCESS![/] {rows:,} rows of beautifully broken {industry} data generated.")
+    console.print(f"[cyan]Saved to: {out_path}[/]")
+    
+    # --- DYNAMIC STAKEHOLDER BRIEF ---
+    real_cols = [col for col in df.columns if not str(col).lower().startswith('unnamed')]
+    numeric_cols = [c for c in df[real_cols].select_dtypes(include=['number', 'float64', 'int64']).columns.tolist() if 'id' not in c.lower() and 'zip' not in c.lower()]
+    # Text cols excluding specific names that usually act as unique IDs (like Patient_Name)
+    text_cols = [c for c in df[real_cols].select_dtypes(include=['object', 'string']).columns.tolist() if 'name' not in c.lower() and 'id' not in c.lower()]
+    date_cols = [col for col in real_cols if 'date' in col.lower() or 'time' in col.lower() or 'year' in col.lower()]
+
+    console.print("\n[bold magenta]👔 STAKEHOLDER BRIEF: SYSTEM REQUIREMENTS[/]")
+    console.print("[bold cyan]KPIs (The Top-Level Cards)[/]")
+    if numeric_cols:
+        for num_col in numeric_cols[:2]:
+            console.print(f"  * Total {num_col.replace('_', ' ')}")
+            console.print(f"  * Avg. {num_col.replace('_', ' ')}")
+    else:
+        console.print("  * Total Records")
+
+    console.print("\n[bold cyan]Business Questions (The Charts & Visuals)[/]")
+    if date_cols and numeric_cols:
+        console.print(f"  * Time-Series Trend: How is {numeric_cols[0].replace('_', ' ')} trending based on {date_cols[0].replace('_', ' ')}? (Line Chart)")
+    if text_cols and numeric_cols:
+        console.print(f"  * Financial/Metric Breakdown: Which {text_cols[0].replace('_', ' ')} drives the highest {numeric_cols[0].replace('_', ' ')}? (Bar Chart)")
+    if text_cols:
+        console.print(f"  * Volume Breakdown: What is the distribution of total volume by {text_cols[0].replace('_', ' ')}? (Donut Chart)")
+    if len(text_cols) > 1 and numeric_cols:
+        console.print(f"  * Secondary Ranking: Compare {text_cols[1].replace('_', ' ')} by {numeric_cols[-1].replace('_', ' ')}? (Bar Chart)")
+
+    console.print("\n[bold yellow]Your training mission is ready. Clean the data and build the Engine to answer these exact questions![/]")
+
+
 def interactive_mode():
     from rich.prompt import Prompt
     
@@ -325,21 +669,40 @@ def interactive_mode():
     
     while True:
         console.print("[bold cyan]What would you like to do?[/]")
-        console.print("  [bold green]1.[/] Audit raw data (Data Quality Check)")
-        console.print("  [bold blue]2.[/] Analyze cleaned data (Dashboard Blueprint)")
-        console.print("  [bold red]3.[/] Exit")
+        console.print("  [bold green]1.[/] Audit Data")
+        console.print("  [bold blue]2.[/] Analyze Data")
+        console.print("  [bold magenta]3.[/] Auto-Clean")
+        console.print("  [bold bright_cyan]4.[/] Map Schema")
+        console.print("  [bold yellow]5.[/] Generate Data")
+        console.print("  [bold white]6.[/] Help Menu")
+        console.print("  [bold red]7.[/] Exit")
         
-        choice = Prompt.ask("\nSelect an option", choices=["1", "2", "3"], default="1")
+        choice = Prompt.ask("\nSelect an option", choices=["1", "2", "3", "4", "5", "6", "7"], default="1")
         
-        if choice == "3":
+        if choice == "7":
             console.print("[yellow]Goodbye![/]")
             break
+        elif choice == "6":
+            console.print("\n[bold underline]Ullr Help Menu[/]")
+            console.print("[bold green]Audit Data:[/] Scans a raw CSV/Excel file for missing values, duplicates, and formatting errors without changing the file.")
+            console.print("[bold blue]Analyze Data:[/] Reads a cleaned dataset and provides a blueprint/layout for building a Dashboard in BI tools.")
+            console.print("[bold magenta]Auto-Clean:[/] Automatically drops empty columns, strips invisible spaces, and converts currency text back to pure numbers.")
+            console.print("[bold bright_cyan]Map Schema:[/] Generates a Markdown Data Dictionary of your dataset and saves it in the same folder as your original file.")
+            console.print("[bold yellow]Generate Data:[/] Creates synthetic, dirty practice datasets (e.g. E-commerce, Healthcare) with intentional errors for you to practice cleaning.")
+        elif choice == "5":
+            generate_data()
+        elif choice == "4":
+            filepath = Prompt.ask("\n[bold yellow]Enter the filename or path to map (e.g., data.csv)[/]")
+            map_data(filepath.strip())
         elif choice == "1":
             filepath = Prompt.ask("\n[bold yellow]Enter the filename or path to audit (e.g., data.csv)[/]")
             audit_data(filepath.strip())
         elif choice == "2":
             filepath = Prompt.ask("\n[bold yellow]Enter the filename or path to analyze (e.g., data.csv)[/]")
             analyze_data(filepath.strip())
+        elif choice == "3":
+            filepath = Prompt.ask("\n[bold yellow]Enter the filename or path to clean (e.g., data.csv)[/]")
+            clean_data(filepath.strip())
         
         console.print("\n" + "-"*50 + "\n")
 
@@ -353,6 +716,14 @@ def cli():
     analyze_parser = subparsers.add_parser("analyze", help="Generate dashboard blueprints for cleaned data.")
     analyze_parser.add_argument("file", help="Path to the cleaned CSV file")
     
+    clean_parser = subparsers.add_parser("clean", help="Auto-clean data (remove empty columns, strip spaces, fix text numbers).")
+    clean_parser.add_argument("file", help="Path to the raw CSV file")
+    
+    map_parser = subparsers.add_parser("map", help="Auto-generate a Markdown Data Dictionary for Obsidian.")
+    map_parser.add_argument("file", help="Path to the CSV/Excel file")
+    
+    generate_parser = subparsers.add_parser("generate", help="Generate synthetic dirty datasets for Excel practice.")
+    
     args = parser.parse_args()
     
     if not args.command:
@@ -364,6 +735,12 @@ def cli():
         audit_data(args.file)
     elif args.command == "analyze":
         analyze_data(args.file)
+    elif args.command == "clean":
+        clean_data(args.file)
+    elif args.command == "map":
+        map_data(args.file)
+    elif args.command == "generate":
+        generate_data()
 
 if __name__ == "__main__":
     cli()

{ullr_data_cli-1.0.1 → ullr_data_cli-1.0.2}/ullr_data_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ullr-data-cli
-Version: 1.0.1
+Version: 1.0.2
 Summary: An offline, AI-free Data Analytics and Dashboard Auditing CLI
 Home-page: https://github.com/jshlydnzl/ullr-data-cli
 Author: jshlydnzl