Understanding WHERE

The WHERE clause is one of the fundamental building blocks of a SQL SELECT statement. In this article we'll explain how it works.

-- Generic SQL SELECT statement
select
  -- (some columns)
from
  -- (some tables)
where
  -- (A 'predicate' expression to be used as a filter)
group by
  -- (some columns)

The WHERE clause is formed of the reserved 'where' keyword followed by what is known as a predicate expression - simply an expression which returns true or false.

If, for a given row, the predicate returns true, that row will be included in the output of the query.

If the predicate returns false, that row will be excluded.

Let's look at some examples.

In this article the code snippets are written in the Google BigQuery Standard SQL syntax.

Simple WHERE expressions

In these examples, the predicate expression in the WHERE clause is a simple boolean expression - an expression that returns true or false.

Inequality

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  numbers > 2 -- This is the predicate expression

Equality

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  numbers = 2

IS NOT NULL

with my_table as (select * from unnest([null,2,3,4,null]) as numbers)
select numbers from my_table
where
  numbers is not null

IS NULL

with my_table as (select * from unnest([null,2,3,4,null]) as numbers)
select numbers from my_table
where
  numbers is null

IN

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  numbers in (1,2)

BETWEEN

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  numbers between 2 and 4 -- Note - these limits are inclusive

Compound WHERE expressions

In the following examples, the WHERE clause contains more complex predicate expressions, using the building blocks of the AND and OR logical operators.

AND / OR

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  -- Use brackets to control precedence
  numbers > 2 and (numbers < 3 or numbers > 4)

Functions

with my_table as (select * from unnest(['a', 'b', 'ab', 'aa', 'bb', 'cc']) as strings)
select strings from my_table
where
  -- Predicate expressions can contain functions
  (strings like 'a%' or strings like 'b%') and length(strings) = 2

SELECT statements

with my_table as (select * from unnest([1,2,3,4,5]) as numbers)
select numbers from my_table
where
  -- Predicate expressions can contain other select statements
  numbers > (select 1+1)

WHERE expressions on dates

When using dates and times, all of the same rules apply. The predicate expression in the WHERE clause must still return true or false, but you are free to use functions and compound logical expressions.

Equality

with my_table as (
  select * from unnest(generate_date_array('2020-01-01', '2020-01-10')) as dates
)
select dates from my_table
where
  dates = '2020-01-01'

Inequality

with my_table as (
  select * from unnest(generate_date_array('2020-01-01', '2020-01-10')) as dates
)
select dates from my_table
where
  dates > '2020-01-05'

BETWEEN

with my_table as (
  select * from unnest(generate_date_array('2020-01-01', '2020-01-10')) as dates
)
select dates from my_table
where
  -- Note - between is inclusive
  dates between '2020-01-02' and '2020-01-04'

AND / OR

with my_table as (
  select * from unnest(generate_date_array('2020-01-01', '2020-01-10')) as dates
)
select dates from my_table
where
  extract(day from dates) = 5 and extract(month from dates) = 1