Comparisons chain, and that can surprise you

Python lets you write a range check the way you'd write it in maths: 36.1 <= temp <= 37.2. The rule behind it is simple, and it applies to every comparison operator, which is where the surprises come from.

The nice part

temp = 37.4
print(36.1 <= temp <= 37.2)

hour = 14
print(9 <= hour < 17)

False
True

No and, no repeating temp. The middle value is also evaluated only once, which matters when it's a function call.

The surprise

The rule is: a op1 b op2 c means a op1 b and b op2 c. It applies to ==, !=, is and in too, not just <:

print(1 < 2 == 2)
print((1 < 2) == 2)

x, y, z = 5, 9, 3
print(x < y > z)

print(False == False in [False])

True
False
True
True

• 1 < 2 == 2 is 1 < 2 and 2 == 2: true. With brackets, (1 < 2) is True, and True == 2 is false.
• x < y > z is 5 < 9 and 9 > 3. It says nothing about how x compares with z.
• False == False in [False] is False == False and False in [False], not "is True in the list".

Why it works

Chaining is part of the grammar, not a trick of <: every comparison operator has the same precedence, and a run of them expands into pairs joined by and, stopping at the first false one.

When not to use it

Keep chains to one direction (a <= b <= c, or a == b == c to check three values match). Mixed directions like x < y > z, or chains with in and is, read like something they're not: split them with and so the next reader doesn't have to work out the rule.