Make calculations incremental

This commit is contained in:
2025-02-09 13:35:44 -05:00
parent 2a6a0d0a87
commit 843ce71506
9 changed files with 935 additions and 1654 deletions

View File

@@ -4,19 +4,50 @@ const { getConnection } = require('./utils/db');
async function calculateBrandMetrics(startTime, totalProducts, processedCount = 0, isCancelled = false) {
const connection = await getConnection();
let success = false;
let processedOrders = 0;
const BATCH_SIZE = 5000;
try {
// Get last calculation timestamp
const [lastCalc] = await connection.query(`
SELECT last_calculation_timestamp
FROM calculate_status
WHERE module_name = 'brand_metrics'
`);
const lastCalculationTime = lastCalc[0]?.last_calculation_timestamp || '1970-01-01';
// Get total count of brands needing updates
const [brandCount] = await connection.query(`
SELECT COUNT(DISTINCT p.brand) as count
FROM products p
LEFT JOIN orders o ON p.pid = o.pid AND o.updated > ?
WHERE p.brand IS NOT NULL
AND (
p.updated > ?
OR o.id IS NOT NULL
)
`, [lastCalculationTime, lastCalculationTime]);
const totalBrands = brandCount[0].count;
if (totalBrands === 0) {
console.log('No brands need metric updates');
return {
processedProducts: 0,
processedOrders: 0,
processedPurchaseOrders: 0,
success: true
};
}
if (isCancelled) {
outputProgress({
status: 'cancelled',
operation: 'Brand metrics calculation cancelled',
current: processedCount,
total: totalProducts,
total: totalBrands,
elapsed: formatElapsedTime(startTime),
remaining: null,
rate: calculateRate(startTime, processedCount),
percentage: ((processedCount / totalProducts) * 100).toFixed(1),
percentage: ((processedCount / totalBrands) * 100).toFixed(1),
timing: {
start_time: new Date(startTime).toISOString(),
end_time: new Date().toISOString(),
@@ -31,23 +62,15 @@ async function calculateBrandMetrics(startTime, totalProducts, processedCount =
};
}
// Get order count that will be processed
const [orderCount] = await connection.query(`
SELECT COUNT(*) as count
FROM orders o
WHERE o.canceled = false
`);
processedOrders = orderCount[0].count;
outputProgress({
status: 'running',
operation: 'Starting brand metrics calculation',
current: processedCount,
total: totalProducts,
total: totalBrands,
elapsed: formatElapsedTime(startTime),
remaining: estimateRemaining(startTime, processedCount, totalProducts),
remaining: estimateRemaining(startTime, processedCount, totalBrands),
rate: calculateRate(startTime, processedCount),
percentage: ((processedCount / totalProducts) * 100).toFixed(1),
percentage: ((processedCount / totalBrands) * 100).toFixed(1),
timing: {
start_time: new Date(startTime).toISOString(),
end_time: new Date().toISOString(),
@@ -55,237 +78,144 @@ async function calculateBrandMetrics(startTime, totalProducts, processedCount =
}
});
// Calculate brand metrics with optimized queries
await connection.query(`
INSERT INTO brand_metrics (
brand,
product_count,
active_products,
total_stock_units,
total_stock_cost,
total_stock_retail,
total_revenue,
avg_margin,
growth_rate
)
WITH filtered_products AS (
SELECT
p.*,
CASE
WHEN p.stock_quantity <= 5000 AND p.stock_quantity >= 0
THEN p.pid
END as valid_pid,
CASE
WHEN p.visible = true
AND p.stock_quantity <= 5000
AND p.stock_quantity >= 0
THEN p.pid
END as active_pid,
CASE
WHEN p.stock_quantity IS NULL
OR p.stock_quantity < 0
OR p.stock_quantity > 5000
THEN 0
ELSE p.stock_quantity
END as valid_stock
// Process in batches
let lastBrand = '';
while (true) {
if (isCancelled) break;
const [batch] = await connection.query(`
SELECT DISTINCT p.brand
FROM products p
WHERE p.brand IS NOT NULL
),
sales_periods AS (
SELECT
p.brand,
SUM(o.quantity * (o.price - COALESCE(o.discount, 0))) as period_revenue,
SUM(o.quantity * (o.price - COALESCE(o.discount, 0) - p.cost_price)) as period_margin,
COUNT(DISTINCT DATE(o.date)) as period_days,
CASE
WHEN o.date >= DATE_SUB(CURRENT_DATE, INTERVAL 3 MONTH) THEN 'current'
WHEN o.date BETWEEN DATE_SUB(CURRENT_DATE, INTERVAL 15 MONTH)
AND DATE_SUB(CURRENT_DATE, INTERVAL 12 MONTH) THEN 'previous'
END as period_type
FROM filtered_products p
JOIN orders o ON p.pid = o.pid
WHERE o.canceled = false
AND o.date >= DATE_SUB(CURRENT_DATE, INTERVAL 15 MONTH)
GROUP BY p.brand, period_type
),
brand_data AS (
SELECT
p.brand,
COUNT(DISTINCT p.valid_pid) as product_count,
COUNT(DISTINCT p.active_pid) as active_products,
SUM(p.valid_stock) as total_stock_units,
SUM(p.valid_stock * p.cost_price) as total_stock_cost,
SUM(p.valid_stock * p.price) as total_stock_retail,
COALESCE(SUM(o.quantity * (o.price - COALESCE(o.discount, 0))), 0) as total_revenue,
CASE
WHEN SUM(o.quantity * o.price) > 0
THEN GREATEST(
-100.0,
LEAST(
100.0,
(
SUM(o.quantity * o.price) - -- Use gross revenue (before discounts)
SUM(o.quantity * COALESCE(p.cost_price, 0)) -- Total costs
) * 100.0 /
NULLIF(SUM(o.quantity * o.price), 0) -- Divide by gross revenue
AND p.brand > ?
AND (
p.updated > ?
OR EXISTS (
SELECT 1 FROM orders o
WHERE o.pid = p.pid
AND o.updated > ?
)
)
ORDER BY p.brand
LIMIT ?
`, [lastBrand, lastCalculationTime, lastCalculationTime, BATCH_SIZE]);
if (batch.length === 0) break;
// Update brand metrics for this batch
await connection.query(`
INSERT INTO brand_metrics (
brand,
product_count,
active_products,
total_stock_units,
total_stock_cost,
total_stock_retail,
total_revenue,
avg_margin,
growth_rate,
last_calculated_at
)
WITH product_stats AS (
SELECT
p.brand,
COUNT(DISTINCT p.pid) as product_count,
COUNT(DISTINCT CASE WHEN p.visible = true THEN p.pid END) as active_products,
SUM(p.stock_quantity) as total_stock_units,
SUM(p.stock_quantity * p.cost_price) as total_stock_cost,
SUM(p.stock_quantity * p.price) as total_stock_retail,
SUM(pm.total_revenue) as total_revenue,
AVG(pm.avg_margin_percent) as avg_margin
FROM products p
LEFT JOIN product_metrics pm ON p.pid = pm.pid
WHERE p.brand IN (?)
AND (
p.updated > ?
OR EXISTS (
SELECT 1 FROM orders o
WHERE o.pid = p.pid
AND o.updated > ?
)
)
ELSE 0
END as avg_margin
FROM filtered_products p
LEFT JOIN orders o ON p.pid = o.pid AND o.canceled = false
GROUP BY p.brand
)
SELECT
bd.brand,
bd.product_count,
bd.active_products,
bd.total_stock_units,
bd.total_stock_cost,
bd.total_stock_retail,
bd.total_revenue,
bd.avg_margin,
CASE
WHEN MAX(CASE WHEN sp.period_type = 'previous' THEN sp.period_revenue END) = 0
AND MAX(CASE WHEN sp.period_type = 'current' THEN sp.period_revenue END) > 0
THEN 100.0
WHEN MAX(CASE WHEN sp.period_type = 'previous' THEN sp.period_revenue END) = 0
THEN 0.0
ELSE GREATEST(
-100.0,
LEAST(
((MAX(CASE WHEN sp.period_type = 'current' THEN sp.period_revenue END) -
MAX(CASE WHEN sp.period_type = 'previous' THEN sp.period_revenue END)) /
NULLIF(ABS(MAX(CASE WHEN sp.period_type = 'previous' THEN sp.period_revenue END)), 0)) * 100.0,
999.99
)
)
END as growth_rate
FROM brand_data bd
LEFT JOIN sales_periods sp ON bd.brand = sp.brand
GROUP BY bd.brand, bd.product_count, bd.active_products, bd.total_stock_units,
bd.total_stock_cost, bd.total_stock_retail, bd.total_revenue, bd.avg_margin
ON DUPLICATE KEY UPDATE
product_count = VALUES(product_count),
active_products = VALUES(active_products),
total_stock_units = VALUES(total_stock_units),
total_stock_cost = VALUES(total_stock_cost),
total_stock_retail = VALUES(total_stock_retail),
total_revenue = VALUES(total_revenue),
avg_margin = VALUES(avg_margin),
growth_rate = VALUES(growth_rate),
last_calculated_at = CURRENT_TIMESTAMP
`);
processedCount = Math.floor(totalProducts * 0.97);
outputProgress({
status: 'running',
operation: 'Brand metrics calculated, starting time-based metrics',
current: processedCount,
total: totalProducts,
elapsed: formatElapsedTime(startTime),
remaining: estimateRemaining(startTime, processedCount, totalProducts),
rate: calculateRate(startTime, processedCount),
percentage: ((processedCount / totalProducts) * 100).toFixed(1),
timing: {
start_time: new Date(startTime).toISOString(),
end_time: new Date().toISOString(),
elapsed_seconds: Math.round((Date.now() - startTime) / 1000)
}
});
if (isCancelled) return {
processedProducts: processedCount,
processedOrders,
processedPurchaseOrders: 0,
success
};
// Calculate brand time-based metrics with optimized query
await connection.query(`
INSERT INTO brand_time_metrics (
brand,
year,
month,
product_count,
active_products,
total_stock_units,
total_stock_cost,
total_stock_retail,
total_revenue,
avg_margin
)
WITH filtered_products AS (
GROUP BY p.brand
),
sales_periods AS (
SELECT
p.brand,
SUM(CASE
WHEN o.date >= DATE_SUB(CURRENT_DATE, INTERVAL 30 DAY)
THEN o.quantity * o.price
ELSE 0
END) as current_period_sales,
SUM(CASE
WHEN o.date BETWEEN DATE_SUB(CURRENT_DATE, INTERVAL 60 DAY) AND DATE_SUB(CURRENT_DATE, INTERVAL 30 DAY)
THEN o.quantity * o.price
ELSE 0
END) as previous_period_sales
FROM products p
INNER JOIN orders o ON p.pid = o.pid
AND o.canceled = false
AND o.date >= DATE_SUB(CURRENT_DATE, INTERVAL 60 DAY)
AND o.updated > ?
WHERE p.brand IN (?)
GROUP BY p.brand
)
SELECT
p.*,
CASE WHEN p.stock_quantity <= 5000 THEN p.pid END as valid_pid,
CASE WHEN p.visible = true AND p.stock_quantity <= 5000 THEN p.pid END as active_pid,
CASE
WHEN p.stock_quantity IS NULL OR p.stock_quantity < 0 OR p.stock_quantity > 5000 THEN 0
ELSE p.stock_quantity
END as valid_stock
FROM products p
WHERE p.brand IS NOT NULL
),
monthly_metrics AS (
SELECT
p.brand,
YEAR(o.date) as year,
MONTH(o.date) as month,
COUNT(DISTINCT p.valid_pid) as product_count,
COUNT(DISTINCT p.active_pid) as active_products,
SUM(p.valid_stock) as total_stock_units,
SUM(p.valid_stock * p.cost_price) as total_stock_cost,
SUM(p.valid_stock * p.price) as total_stock_retail,
SUM(o.quantity * o.price) as total_revenue,
CASE
WHEN SUM(o.quantity * o.price) > 0
THEN GREATEST(
-100.0,
LEAST(
100.0,
(
SUM(o.quantity * o.price) - -- Use gross revenue (before discounts)
SUM(o.quantity * COALESCE(p.cost_price, 0)) -- Total costs
) * 100.0 /
NULLIF(SUM(o.quantity * o.price), 0) -- Divide by gross revenue
)
)
ELSE 0
END as avg_margin
FROM filtered_products p
LEFT JOIN orders o ON p.pid = o.pid AND o.canceled = false
WHERE o.date >= DATE_SUB(CURRENT_DATE, INTERVAL 12 MONTH)
GROUP BY p.brand, YEAR(o.date), MONTH(o.date)
)
SELECT *
FROM monthly_metrics
ON DUPLICATE KEY UPDATE
product_count = VALUES(product_count),
active_products = VALUES(active_products),
total_stock_units = VALUES(total_stock_units),
total_stock_cost = VALUES(total_stock_cost),
total_stock_retail = VALUES(total_stock_retail),
total_revenue = VALUES(total_revenue),
avg_margin = VALUES(avg_margin)
`);
ps.brand,
COALESCE(ps.product_count, 0) as product_count,
COALESCE(ps.active_products, 0) as active_products,
COALESCE(ps.total_stock_units, 0) as total_stock_units,
COALESCE(ps.total_stock_cost, 0) as total_stock_cost,
COALESCE(ps.total_stock_retail, 0) as total_stock_retail,
COALESCE(ps.total_revenue, 0) as total_revenue,
COALESCE(ps.avg_margin, 0) as avg_margin,
CASE
WHEN COALESCE(sp.previous_period_sales, 0) = 0 AND COALESCE(sp.current_period_sales, 0) > 0 THEN 100
WHEN COALESCE(sp.previous_period_sales, 0) = 0 THEN 0
ELSE LEAST(999.99, GREATEST(-100,
((COALESCE(sp.current_period_sales, 0) / sp.previous_period_sales) - 1) * 100
))
END as growth_rate,
NOW() as last_calculated_at
FROM product_stats ps
LEFT JOIN sales_periods sp ON ps.brand = sp.brand
ON DUPLICATE KEY UPDATE
product_count = VALUES(product_count),
active_products = VALUES(active_products),
total_stock_units = VALUES(total_stock_units),
total_stock_cost = VALUES(total_stock_cost),
total_stock_retail = VALUES(total_stock_retail),
total_revenue = VALUES(total_revenue),
avg_margin = VALUES(avg_margin),
growth_rate = VALUES(growth_rate),
last_calculated_at = NOW()
`, [
batch.map(row => row.brand), // For first IN clause
lastCalculationTime, // For p.updated > ?
lastCalculationTime, // For o.updated > ? in EXISTS
lastCalculationTime, // For o.updated > ? in sales_periods
batch.map(row => row.brand) // For second IN clause
]);
processedCount = Math.floor(totalProducts * 0.99);
outputProgress({
status: 'running',
operation: 'Brand time-based metrics calculated',
current: processedCount,
total: totalProducts,
elapsed: formatElapsedTime(startTime),
remaining: estimateRemaining(startTime, processedCount, totalProducts),
rate: calculateRate(startTime, processedCount),
percentage: ((processedCount / totalProducts) * 100).toFixed(1),
timing: {
start_time: new Date(startTime).toISOString(),
end_time: new Date().toISOString(),
elapsed_seconds: Math.round((Date.now() - startTime) / 1000)
}
});
lastBrand = batch[batch.length - 1].brand;
processedCount += batch.length;
outputProgress({
status: 'running',
operation: 'Processing brand metrics batch',
current: processedCount,
total: totalBrands,
elapsed: formatElapsedTime(startTime),
remaining: estimateRemaining(startTime, processedCount, totalBrands),
rate: calculateRate(startTime, processedCount),
percentage: ((processedCount / totalBrands) * 100).toFixed(1),
timing: {
start_time: new Date(startTime).toISOString(),
end_time: new Date().toISOString(),
elapsed_seconds: Math.round((Date.now() - startTime) / 1000)
}
});
}
// If we get here, everything completed successfully
success = true;
@@ -299,7 +229,7 @@ async function calculateBrandMetrics(startTime, totalProducts, processedCount =
return {
processedProducts: processedCount,
processedOrders,
processedOrders: 0,
processedPurchaseOrders: 0,
success
};