SELECT basics

SELECT is the verb you’ll use more than any other in SQL. It returns rows from one or more tables. In its simplest form:

SELECT column1, column2
FROM   table_name
WHERE  some_condition
ORDER BY column1
LIMIT 10;

But here’s the part most tutorials skip: SQL doesn’t run that statement top-to-bottom. The logical order is:

1. FROM — pick the source table(s)
2. WHERE — filter rows
3. GROUP BY — collapse into groups (next lesson)
4. HAVING — filter groups
5. SELECT — choose the columns to keep
6. ORDER BY — sort
7. LIMIT — keep only N rows

That’s why you can’t use a column alias defined in SELECT inside the WHERE clause: the SELECT hasn’t run yet.

Try it

The schema below has users, products, and orders. Run any query against it.

SELECT * returns every column. In real production code, always list columns explicitly — SELECT * is unstable (schema changes silently add/remove columns) and slow (every column is shipped over the wire).

Filtering with WHERE

BETWEEN a AND b is inclusive of both ends. Other useful comparisons:

WHERE country IN ('IN', 'US')          -- match any in list
WHERE country NOT IN ('CN')
WHERE name LIKE 'A%'                    -- A and anything after
WHERE name ILIKE 'a%'                   -- case-insensitive (Postgres)
WHERE signup_date IS NULL
WHERE signup_date IS NOT NULL

Counting and aggregating

COUNT(*) counts rows. COUNT(column) counts non-NULL values in that column. COUNT(DISTINCT column) counts unique non-NULL values. The difference matters when columns have nulls.

ORDER BY and LIMIT

DESC = descending, ASC = ascending (default). You can order by multiple columns: ORDER BY country, signup_date DESC.

Column aliases — the small thing that scales

In most dialects, AS is optional but recommended for clarity. Table aliases (FROM users u) are the same idea — they keep multi-table queries short.

Putting it together

Quick check