re.match().group() fails despite if check

[jamesmcm]

The following code fails the mypy check:

   df.loc[:, "col"] = df.loc[:, "col"].apply(
          lambda x: r.match(x).group(1) if r.match(x) else "0"
      )

With error:

error: Item "None" of "Optional[Match[str]]" has no attribute "group"

Despite the explicit check for r.match(x) that means .group() will never be called for None.

How can I work around this?

[srittau]

r.match() is called twice. mypy has no way of knowing that it returns the same value both times. The solution is to assign the result of r.match() to a variable and check that. Of course, this does not work in a lambda.

[msullivan]

In Python 3.8 you'll be able to use assignment expressions which should work with mypy and be more efficient, but until then try a cast or a type ignore.

[Bidek56]

mypy with Python 3.8 still has the same issue with the example code:

import re

advertisement = '10% discount'
discount = 0.0
if (mo := re.search(r'(\d+)% discount', advertisement)):
    discount = float(mo.group(1)) / 100.0

results in: error: Item "None" of "Optional[Match[str]]" has no attribute "group"
Any ideas would be appreciated. Thanks

[JelleZijlstra]

This is #7316.