Note: The attached image is a screenshot of page 31 of Dr. Charles Severance’s book, Python for Everybody: Exploring Data Using Python 3 (2024-01-01 Revision).
I thought =
was a mathematical operator, not a logical operator; why does Python use
=
instead of ==
, or
<=
instead of <==
, or
!=
instead of !==
?
Thanks in advance for any clarification. I would have posted this in the help forums of FreeCodeCamp, but I wasn’t sure if this question was too…unspecified(?) for that domain.
Cheers!
>=
and<
= match the mathematical operators. The question you want to ask is why doesn’t it use=
for equality, and the answer is that=
is already used for assignment (inherited from C among other languages).In theory a language could use
=
for assignment and equality but it might be a bit confusing and error prone. Maybe not though. Someone try it and report back.I think what I’m most confused about is I cannot for the life of me seem to wrap my head around the difference between “assignment” and “equality”. They seem exactly the same to me: when a variable is assigned a value, it’s equal to that value now.
Even if I were write the program
it would still print
40
. Because x is equal to 20. Because it was assigned the value of 20.Hell, I’ve even heard Dr. Severance say “equal to” in this context in earlier videos multiple times.
Yeah it’s confusing because in maths they are the same and use the same symbol but they are 100% not the same in programming, yet they confusingly used the same symbol. In fact they even used the mathematical equality symbol (
=
) for the thing that is least like equality (i.e. assignment).To be fair not all languages made that mistake. There are a fair few where assignment is like
Or
which is probably the most logical option because it really conveys the “store 20 in x” meaning.
Anyway on to your actual question… They definitely aren’t the same in programming. Probably the simplest way to think of it is that assignment is a command: make these things equal! and equality is a question: are these things equal?
So for example equality will never mutate it’s arguments.
x == y
will never changex
ory
because you’re just asking “are they equal?”. The value of that equality expression is a bool (true or false) so you can do something like:x == y
asks if they are equal and becomes a bool with the answer, and then the = stores that answer insidea
.In contrast
=
always mutates something. You can do this:And it will print 4. If you do this:
It will (if the language doesn’t complain at you for this mistake) print 3 because the == doesn’t actually change
a
.I’ve written code before in some hardware-specific languages before (I think it was for programming a stepper motor or something?) that used
=
for both assignment and comparison. If I recall correctly, the language was vaguely C-like, but assignment was not permitted in the context of a comparison. So something likeif( a = (b+c) )
would not assign a value toa
, it would just do the comparison.Rust does an interesting thing in this regard. It does still have
==
for checking if two values are equal, but well, it actually doesn’t have a traditional assignment operator. Instead, it has a unification operator, which programmers usually call “pattern matching”.And then you can use pattern matching for what’s effectively an assignment and to some degree also for equivalence comparison.
See a few examples here: https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=1268682eb8642af925db9a499a6d587a
This reminds me on the niche tool in Mathematica I’ve been using, which has four different assignment oparators for that purpose.
It does still have a traditional assignment operator. You can assign values to mutable variables.
Also I would say let-binds are still pretty much assignment; they just support destructuring. Plenty of languages support that to some extent (JavaScript for example) and you wouldn’t say they don’t have assignment.
I don’t think it affects the ability to overload
=
anyway. I think there aren’t any situations in Rust where it would be ambiguous which one you meant. Certainly none of the examples you gave compile with both=
and==
. Maybe there’s some obscure case we haven’t thought of.