if (x := some_function()):
do_something(x)
else:
# x is None or False or something, consider it invalid
The only thing I wish was different is adding a scope, which would make x invalid outside the block. But Python’s scoping rules are too dumb to handle this case.
In lots of functional languages, you’d have some_function() return an Option type and then you’d use .map() or similar on it to only do something when the value is defined. I think, that can be done in Python’s syntax, but you need a library for the Option type…
Python has an “Optional” type, but it’s merely an alias for T | None. I wish Python had better support for FP, but each time it gets close, it doesn’t quite go far enough.
For example, it now has match blocks, but there’s no error if the match is exhaustive, which hurts formal proofs of correctness. Likewise, tons of other features fall a bit short, but in general it’s workable with some discipline.
I think my most common use case is with dictionary lookups.
if (val := dct.get(key)) is not None:
# do something with val
I’ve also found some cases where the walrus is useful in something like a list comprehension. I suppose expanding on the above example, you you make one that looks up several keys in a dict and gives you their corresponding values where available.
vals = [val for key in (key1, key2, key3) if (val := dct.get(key)) is not None]
Walrus operator does an inline assignment to a variable and resolves to the value assigned. If it is in a condition statement, like “if x := y:”, it assigns the value of y to x then interprets the expression of the condition as of it just said “if x:”. Functionally, that means the assignment happens regardless of the value of y, but the condition only passes if the value of y is “truthy”, i.e. if it’s not None, an empty collection, numerically equal to zero, or just False.
Walrus operator. What did I just read?
I love the walrus operator:
if (x := some_function()): do_something(x) else: # x is None or False or something, consider it invalidThe only thing I wish was different is adding a scope, which would make
xinvalid outside the block. But Python’s scoping rules are too dumb to handle this case.Walrus is more useful in a while loop.
while (data := f.read(1024)): passIn lots of functional languages, you’d have
some_function()return anOptiontype and then you’d use.map()or similar on it to only do something when the value is defined. I think, that can be done in Python’s syntax, but you need a library for theOptiontype…Python has an “Optional” type, but it’s merely an alias for
T | None. I wish Python had better support for FP, but each time it gets close, it doesn’t quite go far enough.For example, it now has
matchblocks, but there’s no error if the match is exhaustive, which hurts formal proofs of correctness. Likewise, tons of other features fall a bit short, but in general it’s workable with some discipline.I think my most common use case is with dictionary lookups.
I’ve also found some cases where the walrus is useful in something like a list comprehension. I suppose expanding on the above example, you you make one that looks up several keys in a dict and gives you their corresponding values where available.
Walrus operator does an inline assignment to a variable and resolves to the value assigned. If it is in a condition statement, like “if x := y:”, it assigns the value of y to x then interprets the expression of the condition as of it just said “if x:”. Functionally, that means the assignment happens regardless of the value of y, but the condition only passes if the value of y is “truthy”, i.e. if it’s not None, an empty collection, numerically equal to zero, or just False.