= — the same as in most host languages.

:= — for example p_customerLevel := 'PLATINUM';

SET — for example SET p_customerLevel = 'PLATINUM';

== — borrowed from C.

GROUP BY clause

FROM clause with only one table

HAVING clause

DISTINCT

Both OLD and NEW, exactly as in an UPDATE trigger.

Neither — INSERT triggers cannot see row data.

Only NEW.col_name — there is no old row on INSERT.

Only OLD.col_name — there is no new row yet.

They are interchangeable.

WHERE filters groups, whereas HAVING filters individual rows.

WHERE can use aggregate functions, whereas HAVING cannot.

WHERE filters individual rows, whereas HAVING filters groups.

Deletes all rows from Viewing; the table itself remains.

Deletes the Viewing table from the database.

Deletes nothing, because WHERE is mandatory.

Deletes only rows containing null values.

SELECT department_id, COUNT(last_name) FROM employees;

SELECT AVG(salary) FROM employees;

SELECT COUNT(last_name) FROM employees;

SELECT department_id, COUNT(last_name) FROM employees GROUP BY department_id;

When they return the same number of columns with compatible data types.

When they select from the same tables.

When their column names are identical.

When they return the same number of rows.

The column stores exactly the value 7.2.

7 decimal places with 2 significant digits.

Precision 7 means 7 digits in total and scale 2 means 2 of them are after the decimal point — it stores values up to 99,999.99.

7 digits before the decimal point and 2 digits after it (9 digits total).

Use a subquery.

Use HAVING instead of WHERE.

Add DISTINCT inside the inner aggregate.

Place both aggregates in the GROUP BY clause.

Alphabetical order of the column names.

Any order, as long as the data types match.

The original CREATE TABLE column order.

Primary-key columns first, then the rest.