Ricardo Camacho Submission

calculate_largest_expensors.sql

@@ -0,0 +1,24 @@
+USE memory.default;
+
+-- List of employees that expended more than 1000 units with their manager information.
+CREATE OR REPLACE VIEW largest_expensors AS (
+    SELECT  
+        emp.employee_id
+        , concat(emp.first_name, ' ', emp.last_name) AS employee_name
+        , emp.manager_id
+        , concat(man.first_name, ' ', man.last_name) AS manager_name
+        , sum(exp.unit_price * exp.quantity) AS total_expensed_amount
+    FROM EMPLOYEE AS emp
+    INNER JOIN EXPENSE AS exp
+        ON emp.employee_id = exp.employee_id
+    LEFT JOIN EMPLOYEE AS man
+        ON man.employee_id = emp.manager_id
+    GROUP BY
+        emp.employee_id
+        , concat(emp.first_name, ' ', emp.last_name)
+        , emp.manager_id
+        , concat(man.first_name, ' ', man.last_name)
+    HAVING sum(exp.unit_price * exp.quantity) > 1000
+);
+
+SELECT * FROM largest_expensors ORDER BY total_expensed_amount DESC;

create_employees.sql

@@ -0,0 +1,14 @@
+USE memory.default;
+
+-- EMPLOYEE table definition
+CREATE TABLE EMPLOYEE (
+    employee_id TINYINT NOT NULL,
+    first_name VARCHAR,
+    last_name VARCHAR,
+    job_title VARCHAR,
+    manager_id TINYINT
+);
+
+-- Populate EMPLOYEE table with manual data from hr/employee_index.csv
+INSERT INTO EMPLOYEE SELECT * FROM brz_employees
+;

create_expenses.sql

@@ -0,0 +1,20 @@
+USE memory.default;
+
+-- EXPENSE table definition
+CREATE TABLE EXPENSE (
+    employee_id TINYINT,
+    unit_price DECIMAL(8, 2),
+    quantity TINYINT
+);
+
+-- Populate EXPENSE table with data from brz_expenses and employee IDs from EMPLOYEE dimension table.
+INSERT INTO EXPENSE 
+    SELECT 
+        emp.employee_id AS employee_id
+        , exp_t.unit_price AS unit_price
+        , exp_t.quantity AS quantity
+    FROM brz_expenses AS exp_t
+    INNER JOIN EMPLOYEE AS emp
+        ON lower(exp_t.employee_full_name) = lower(concat(emp.first_name, ' ', emp.last_name))
+;
+

create_invoices.sql

@@ -0,0 +1,33 @@
+USE memory.default;
+
+-- INVOICE table definition
+CREATE TABLE INVOICE (
+    supplier_id TINYINT,
+    invoice_amount DECIMAL(8, 2),
+    due_date DATE
+);
+
+-- SUPPLIER table definition
+CREATE TABLE SUPPLIER (
+    supplier_id TINYINT,
+    name VARCHAR
+);
+
+-- Populate SUPPLIER table with 
+INSERT INTO SUPPLIER 
+    SELECT
+        row_number() over(order by company_name) AS supplier_id
+        , company_name AS supplier_name
+    FROM (SELECT DISTINCT company_name FROM brz_invoices)
+;
+
+-- Populate INVOICE table with data from brz_invoices and new supplier_id from SUPPLIER
+INSERT INTO INVOICE 
+    SELECT
+        sup.supplier_id AS supplier_id
+        , inv_t.invoice_amount AS invoice_amount
+        , last_day_of_month(date_add('month', inv_t.due_date_in_months, now())) AS due_date
+    FROM brz_invoices AS inv_t
+    INNER JOIN SUPPLIER AS sup 
+        ON inv_t.company_name = sup.name
+;

data_loading/employee_data_loading.py

@@ -0,0 +1,48 @@
+import pandas as pd
+import os
+
+def generate_sql_from_csv():
+    """
+    Reads employee data from CSV and generates SQL INSERT statements
+    to save in a .sql file.
+    """
+    # Get the absolute path to the CSV file
+    file_path = os.path.join('hr', 'employee_index.csv')
+
+    # Check if the file exists
+    if not os.path.exists(file_path):
+        print(f"Error: File not found at {file_path}")
+        print(f"Current working directory: {os.getcwd()}")
+        return False
+
+    try:
+        # Read the CSV file into a pandas DataFrame
+        df = pd.read_csv(file_path)
+
+        # Generate SQL INSERT statements
+        insert_statements = []
+        for _, row in df.iterrows():
+            values = f"({row['employee_id']}, '{row['first_name']}', '{row['last_name']}', '{row['job_title']}', {row['manager_id']})"
+            insert_statements.append(values)
+
+        # Combine all INSERT statements
+        all_values = ",\n    ".join(insert_statements)
+        sql_insert = f"""-- Insert data from CSV
+INSERT INTO brz_employees (employee_id, first_name, last_name, job_title, manager_id) VALUES
+    {all_values};
+"""
+
+        # Create or append to the SQL file
+        with open('data_loading\output_queries\brz_employees.sql', 'a') as f:
+            f.write(sql_insert)
+
+        print(f"Successfully generated SQL INSERT statements for {len(df)} employee records")
+        print(f"Appended to employee_data.sql")
+        return True
+
+    except Exception as e:
+        print(f"Error processing the data: {e}")
+        return False
+
+if __name__ == "__main__":
+    generate_sql_from_csv()

data_loading/expenses_data_loading.py

@@ -0,0 +1,63 @@
+import os
+import re
+import glob
+
+def parse_expense_file(file_path):
+    """Parse an expense receipt text file and extract relevant information."""
+    with open(file_path, 'r') as file:
+        content = file.read()
+
+    # Extract employee name
+    employee_match = re.search(r'Employee: (.+)', content)
+    employee_name = employee_match.group(1) if employee_match else ""
+
+    # Extract unit price
+    price_match = re.search(r'Unit Price: (\d+\.\d+)', content)
+    unit_price = float(price_match.group(1)) if price_match else 0.0
+
+    # Extract quantity
+    quantity_match = re.search(r'Quantity: (\d+)', content)
+    quantity = int(quantity_match.group(1)) if quantity_match else 0
+
+    return {
+        'employee_name': employee_name if employee_name is not None else "",
+        'unit_price': unit_price,
+        'quantity': quantity
+    }
+
+def generate_expenses_sql():
+    """Generate SQL for brz_expenses table."""
+    # Create table
+    sql = """-- Create brz_expenses table
+USE memory.default;
+DROP TABLE IF EXISTS brz_expenses;
+CREATE TABLE brz_expenses (
+    employee_full_name VARCHAR,
+    unit_price DECIMAL(8, 2),
+    quantity TINYINT
+);
+
+-- Insert data from receipts_from_last_night/*.txt files
+"""
+
+    # Get all receipt files
+    receipt_files = glob.glob('finance\\receipts_from_last_night\\*.txt')
+
+    # Process each file and generate INSERT statements
+    inserts = []
+    for file_path in receipt_files:
+        expense_data = parse_expense_file(file_path)
+        insert = f"INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES " \
+                f"(\'{expense_data['employee_name']}\', {expense_data['unit_price']}, {expense_data['quantity']});"
+        inserts.append(insert)
+
+    return sql + "\n".join(inserts)
+
+def main():
+    # Generate SQL for expenses
+    expenses_sql = generate_expenses_sql()
+    with open('data_loading\\output_queries\\brz_expenses.sql', 'w') as file:
+        file.write(expenses_sql)
+
+if __name__ == "__main__":
+    main()

data_loading/invoices_data_loading.py

@@ -0,0 +1,64 @@
+import os
+import re
+import glob
+def parse_invoice_file(file_path):
+    """Parse an invoice text file and extract relevant information."""
+
+    with open(file_path, 'r') as file:
+        content = file.read()
+
+    # Extract invoice amount
+    company_name_match = re.search(r'Company Name: (.+)', content)
+    company_name = str(company_name_match.group(1)) if company_name_match else ""
+    company_name = company_name.replace("\'", "\"")
+    amount_match = re.search(r'Invoice Amount: (\d+)', content)
+    invoice_amount = float(amount_match.group(1)) if amount_match else 0.0
+
+    # Extract due date in months
+    due_date_match = re.search(r'Due Date: (\d+) months', content)
+    due_date_in_months = int(due_date_match.group(1)) if due_date_match else 0
+
+    return {
+        'company_name': company_name,
+        'invoice_amount': invoice_amount,
+        'due_date_in_months': due_date_in_months
+    }
+
+def generate_invoices_sql():
+    """Generate SQL for brz_invoices table."""
+    # Create table
+    sql = """-- Create brz_invoices table
+    USE memory.default;
+DROP TABLE IF EXISTS brz_invoices;
+CREATE TABLE brz_invoices (
+    company_name VARCHAR(100),
+    invoice_amount DECIMAL(8, 2),
+    due_date_in_months INT
+);
+
+-- Insert data from invoices_due/*.txt files
+"""
+
+    # Get all invoice files
+    invoice_files = glob.glob('finance\\invoices_due\\*.txt')
+
+    # Process each file and generate INSERT statements
+    inserts = []
+    for file_path in invoice_files:
+        invoice_data = parse_invoice_file(file_path)
+        insert = f"INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES " \
+                f"('{invoice_data['company_name']}', {invoice_data['invoice_amount']}, {invoice_data['due_date_in_months']});"
+        inserts.append(insert)
+
+    return sql + "\n".join(inserts)
+
+def main():
+    """Generate SQL files for both tables."""
+    # Generate SQL for invoices
+    invoices_sql = generate_invoices_sql()
+    with open('data_loading\\output_queries\\brz_invoices.sql', 'w') as file:
+        file.write(invoices_sql)
+
+
+if __name__ == "__main__":
+    main()

data_loading/output_queries/brz_employees.sql

@@ -0,0 +1,11 @@
+-- Insert data from CSV
+INSERT INTO brz_employees (employee_id, first_name, last_name, job_title, manager_id) VALUES
+    (1, 'Ian', 'James', 'CEO', 4),
+    (2, 'Umberto', 'Torrielli', 'CSO', 1),
+    (3, 'Alex', 'Jacobson', 'MD EMEA', 2),
+    (4, 'Darren', 'Poynton', 'CFO', 2),
+    (5, 'Tim', 'Beard', 'MD APAC', 2),
+    (6, 'Gemma', 'Dodd', 'COS', 1),
+    (7, 'Lisa', 'Platten', 'CHR', 6),
+    (8, 'Stefano', 'Camisaca', 'GM Activation', 2),
+    (9, 'Andrea', 'Ghibaudi', 'MD NAM', 2);

data_loading/output_queries/brz_expenses.sql

@@ -0,0 +1,17 @@
+-- Create brz_expenses table
+USE memory.default;
+DROP TABLE IF EXISTS brz_expenses;
+CREATE TABLE brz_expenses (
+    employee_full_name VARCHAR,
+    unit_price DECIMAL(8, 2),
+    quantity TINYINT
+);
+
+-- Insert data from receipts_from_last_night/*.txt files
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Alex Jacobson', 6.5, 14);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Alex Jacobson', 11.0, 20);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Alex Jacobson', 22.0, 18);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Alex Jacobson', 13.0, 75);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Andrea Ghibaudi', 0.0, 1);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Darren Poynton', 40.0, 9);
+INSERT INTO brz_expenses (employee_full_name, unit_price, quantity) VALUES ('Umberto Torrielli', 17.5, 4);

data_loading/output_queries/brz_invoices.sql

@@ -0,0 +1,16 @@
+-- Create brz_invoices table
+    USE memory.default;
+DROP TABLE IF EXISTS brz_invoices;
+CREATE TABLE brz_invoices (
+    company_name VARCHAR(100),
+    invoice_amount DECIMAL(8, 2),
+    due_date_in_months INT
+);
+
+-- Insert data from invoices_due/*.txt files
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Party Animals', 6000.0, 3);
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Catering Plus', 2000.0, 2);
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Catering Plus', 1500.0, 3);
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Dave"s Discos', 500.0, 0);
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Entertainment tonight', 6000.0, 3);
+INSERT INTO brz_invoices (company_name, invoice_amount, due_date_in_months) VALUES ('Ice Ice Baby', 4000.0, 6);

find_manager_cycles.sql

@@ -0,0 +1,46 @@
+USE memory.default;
+
+CREATE OR REPLACE VIEW manager_cycles AS ( SELECT * FROM (
+    -- Create recursive MANAGERS_CTE with all the managers with their employee ID list.
+    WITH RECURSIVE MANAGERS_CTE (
+        manager_id
+        , employee_id
+        , employee_id_list
+        , level
+    ) AS (
+        -- Anchor: All managers
+        SELECT 
+            manager_id
+            , employee_id
+            , cast(employee_id as VARCHAR) AS employee_id_list
+            , 0 as level
+        FROM EMPLOYEE
+
+        UNION ALL
+
+        -- Recursive: Add recursively the employees of each manager's employee
+        SELECT
+            man.manager_id
+            , emp.employee_id
+            , concat(man.employee_id_list, '; ', cast(emp.employee_id as VARCHAR)) AS employee_id_list
+            , level + 1
+        FROM EMPLOYEE AS emp
+        -- Recursive over each employee if it is also a manager
+        INNER JOIN MANAGERS_CTE AS man
+            ON man.employee_id = emp.manager_id
+        -- Avoid infinite recursion
+        WHERE position(cast(emp.employee_id as VARCHAR) in man.employee_id_list) = 0
+    )
+
+    -- List of employees that form a managing loop.
+    SELECT
+        manager_id
+        , max(employee_id_list) AS employee_id_list
+        , max(level) AS subordinates_level
+    FROM MANAGERS_CTE
+    -- Filter for managers that are in their own recursive list of employees (indicating a loop).
+    WHERE position(cast(manager_id as VARCHAR) in employee_id_list) <> 0
+    GROUP BY manager_id
+));
+
+SELECT * FROM manager_cycles;