SExI System SQL Scripts and data loading implementation

calculate_largest_expensors.sql

@@ -0,0 +1,31 @@
+-- This query identifies employees who have exceeded the spending threshold
+-- It calculates total expenses for each employee and compares against the limit
+-- The results include the employee's manager for approval chain visibility
+-- This analysis helps identify potential policy violations
+
+USE memory.default;
+
+WITH employee_expenses AS (
+    -- Calculate total expenses for each employee
+    -- Multiply unit price by quantity to get the total amount for each expense
+    -- Group by employee to get their total spending
+    SELECT
+        employee_id,
+        SUM(unit_price * quantity) AS total_spent
+    FROM EXPENSE
+    GROUP BY employee_id
+)
+-- Join with employee data to get names and manager information
+-- Filter for employees who exceeded the spending threshold
+-- Order results by total amount spent in descending order
+SELECT
+    ee.employee_id,
+    CONCAT(e.first_name, ' ', e.last_name) AS employee_name,
+    e.manager_id,
+    CONCAT(m.first_name, ' ', m.last_name) AS manager_name,
+    ee.total_spent AS total_expensed_amount
+FROM employee_expenses ee
+JOIN EMPLOYEE e ON ee.employee_id = e.employee_id
+LEFT JOIN EMPLOYEE m ON e.manager_id = m.employee_id
+WHERE ee.total_spent > 1000
+ORDER BY ee.total_spent DESC;

create_employees.sql

@@ -0,0 +1,22 @@
+-- This file creates the EMPLOYEE table and loads our employee data
+USE memory.default;
+
+CREATE TABLE EMPLOYEE (
+    employee_id TINYINT,
+    first_name VARCHAR,
+    last_name VARCHAR,
+    job_title VARCHAR,
+    manager_id TINYINT
+);
+
+-- Loading data from hr/employee_index.csv
+INSERT INTO EMPLOYEE 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);

create_expenses.sql

@@ -0,0 +1,18 @@
+-- This file creates the EXPENSE table and loads all our receipts
+USE memory.default;
+
+CREATE TABLE EXPENSE (
+    employee_id TINYINT,
+    unit_price DECIMAL(8, 2),
+    quantity TINYINT
+);
+
+-- Loading data from finance/receipts_from_last_night/*.txt
+INSERT INTO EXPENSE VALUES
+    (3, 13.0, 75),
+    (3, 22.0, 18),
+    (4, 40.0, 9),
+    (9, 300.0, 1),
+    (2, 17.5, 4),
+    (3, 11.0, 20),
+    (3, 6.5, 14);

create_invoices.sql

@@ -0,0 +1,32 @@
+-- This file creates the SUPPLIER and INVOICE tables and loads our invoice data
+USE memory.default;
+
+-- First create the SUPPLIER table
+CREATE TABLE SUPPLIER (
+    supplier_id TINYINT,
+    name VARCHAR
+);
+
+-- Load supplier data from finance/invoices_due/*.txt
+INSERT INTO SUPPLIER VALUES
+    (1, 'Catering Plus'),
+    (2, 'Dave''s Discos'),
+    (3, 'Entertainment tonight'),
+    (4, 'Ice Ice Baby'),
+    (5, 'Party Animals');
+
+-- Now create the INVOICE table
+CREATE TABLE INVOICE (
+    supplier_id TINYINT,
+    invoice_amount DECIMAL(8, 2),
+    due_date DATE
+);
+
+-- Load invoice data from finance/invoices_due/*.txt
+INSERT INTO INVOICE VALUES
+    (1, 1500.0, DATE '2025-07-31'),
+    (2, 500.0, DATE '2025-07-31'),
+    (4, 4000.0, DATE '2025-07-31'),
+    (1, 2000.0, DATE '2025-07-31'),
+    (3, 6000.0, DATE '2025-07-31'),
+    (5, 6000.0, DATE '2025-07-31');

find_employee_expenses.sql

@@ -0,0 +1,22 @@
+-- This query retrieves all expenses for a specific employee
+-- It joins the EXPENSE table with the EMPLOYEE table to get complete information
+-- This provides visibility into employee spending patterns
+-- It also shows who approved each expense for accountability purposes
+
+USE memory.default;
+
+SELECT
+    e.expense_id,
+    e.employee_id,
+    emp.first_name,
+    emp.last_name,
+    e.amount,
+    e.description,
+    e.date,
+    e.approved_by,
+    approver.first_name as approver_first_name,
+    approver.last_name as approver_last_name
+FROM EXPENSE e
+JOIN EMPLOYEE emp ON e.employee_id = emp.employee_id
+LEFT JOIN EMPLOYEE approver ON e.approved_by = approver.employee_id
+WHERE e.employee_id = 1;  -- We can replace with the employee ID based on which we want to check!

find_employee_managers.sql

@@ -0,0 +1,40 @@
+-- This query retrieves the complete management chain for a specific employee
+-- It uses a recursive CTE to traverse the entire management hierarchy
+-- This provides visibility into the organizational structure
+-- It shows all managers in the approval chain for an employee
+
+USE memory.default;
+
+WITH RECURSIVE management_chain AS (
+    -- Base case: the employee's direct manager
+    -- Start with the immediate supervisor
+    SELECT
+        e.employee_id,
+        e.first_name,
+        e.last_name,
+        e.manager_id,
+        1 as level
+    FROM EMPLOYEE e
+    WHERE e.employee_id = 1   -- We can replace with the employee ID based on which we want to check!
+
+    UNION ALL
+
+    -- Recursive case: continue up the management chain
+    -- For each manager, find their manager, and so on
+    -- This builds the complete approval hierarchy
+    SELECT
+        m.employee_id,
+        m.first_name,
+        m.last_name,
+        m.manager_id,
+        mc.level + 1
+    FROM EMPLOYEE m
+    JOIN management_chain mc ON m.employee_id = mc.manager_id
+)
+SELECT
+    employee_id,
+    first_name,
+    last_name,
+    level
+FROM management_chain
+ORDER BY level;

find_manager_cycles.sql

@@ -0,0 +1,42 @@
+-- This query identifies cycles in the management hierarchy
+-- A cycle occurs when an employee reports to someone who eventually reports back to them
+-- These cycles can cause issues with expense approval workflows
+-- This analysis helps identify organizational structure problems
+
+USE memory.default;
+
+WITH RECURSIVE manager_chain(employee_id, manager_id, path, depth) AS (
+    -- Start with each employee and their direct manager
+    -- Initialize the recursive traversal of the management hierarchy
+    SELECT
+        employee_id,
+        manager_id,
+        CAST(employee_id AS VARCHAR) as path,
+        1 as depth
+    FROM EMPLOYEE
+
+    UNION ALL
+
+    -- Continue traversing the management chain upward
+    -- For each employee, examine their manager's manager, and so on
+    -- This builds a complete path of the management chain
+    SELECT
+        e.employee_id,
+        e.manager_id,
+        mc.path || ',' || CAST(e.employee_id AS VARCHAR),
+        mc.depth + 1
+    FROM EMPLOYEE e
+    JOIN manager_chain mc ON e.employee_id = mc.manager_id
+    WHERE mc.depth < 10  -- Limit recursion depth to prevent infinite loops
+    -- Maximum depth of 10 levels is sufficient for most organizational structures
+)
+-- Identify paths that contain the same employee more than once
+-- This indicates a cycle in the management hierarchy
+-- These cycles need to be addressed to ensure proper expense approval workflows
+SELECT DISTINCT
+    employee_id,
+    path as cycle
+FROM manager_chain
+WHERE path LIKE '%,' || CAST(employee_id AS VARCHAR) || ',%'  -- Employee appears in middle of path
+   OR path LIKE CAST(employee_id AS VARCHAR) || ',%'          -- Employee appears at start of path
+   OR path LIKE '%,' || CAST(employee_id AS VARCHAR);         -- Employee appears at end of path

generate_supplier_payment_plans.sql

@@ -0,0 +1,88 @@
+-- This query generates a payment plan for each supplier
+-- It calculates monthly payment amounts to clear invoices before due dates
+-- The plan ensures all invoices are paid in full by their respective due dates
+-- This analysis helps with cash flow management and vendor relationships
+
+USE memory.default;
+
+WITH supplier_totals AS (
+    -- Calculate total amount owed to each supplier
+    -- This is the starting point for our payment plan
+    SELECT
+        supplier_id,
+        SUM(invoice_amount) AS total_amount,
+        MAX(due_date) AS latest_due_date
+    FROM INVOICE
+    GROUP BY supplier_id
+),
+payment_months AS (
+    -- Determine how many months we have to pay each supplier
+    -- Calculate from current date to the latest due date
+    SELECT
+        supplier_id,
+        total_amount,
+        latest_due_date,
+        -- Calculate months between now and due date, rounded up
+        CEIL(DATEDIFF('month', CURRENT_DATE, latest_due_date)) AS months_to_pay
+    FROM supplier_totals
+),
+monthly_payments AS (
+    -- Generate a row for each payment month
+    -- This creates the payment schedule for each supplier
+    SELECT
+        supplier_id,
+        total_amount,
+        latest_due_date,
+        months_to_pay,
+        -- Calculate payment amount for each month
+        -- For the last month, pay the remaining balance
+        CASE
+            WHEN payment_month = months_to_pay THEN
+                total_amount - (FLOOR(total_amount / months_to_pay) * (months_to_pay - 1))
+            ELSE
+                FLOOR(total_amount / months_to_pay)
+        END AS payment_amount,
+        -- Calculate the payment date (last day of each month)
+        -- Start from current month and increment for each payment
+        DATE_ADD('month', payment_month - 1,
+            DATE_ADD('day',
+                -1,
+                DATE_ADD('month', 1,
+                    DATE_TRUNC('month', CURRENT_DATE)
+                )
+            )
+        ) AS payment_date,
+        -- Calculate remaining balance after each payment
+        total_amount - (FLOOR(total_amount / months_to_pay) * payment_month) AS balance_outstanding
+    FROM payment_months
+    CROSS JOIN (
+        -- Generate a sequence of payment months
+        SELECT payment_month
+        FROM (
+            SELECT 1 AS payment_month
+            UNION ALL SELECT 2
+            UNION ALL SELECT 3
+            UNION ALL SELECT 4
+            UNION ALL SELECT 5
+            UNION ALL SELECT 6
+            UNION ALL SELECT 7
+            UNION ALL SELECT 8
+            UNION ALL SELECT 9
+            UNION ALL SELECT 10
+            UNION ALL SELECT 11
+            UNION ALL SELECT 12
+        ) months
+        WHERE payment_month <= (SELECT MAX(months_to_pay) FROM payment_months)
+    ) payment_sequence
+)
+-- Join with supplier information and format the results
+-- This provides a complete payment schedule for each supplier
+SELECT
+    mp.supplier_id,
+    s.name AS supplier_name,
+    mp.payment_amount,
+    mp.balance_outstanding,
+    mp.payment_date
+FROM monthly_payments mp
+JOIN SUPPLIER s ON mp.supplier_id = s.supplier_id
+ORDER BY mp.supplier_id, mp.payment_date;