pyflakes
pyflakes copied to clipboard
PyCQA
Report unassigned expressions
I am -0.5. There are legitimate use-cases for calling == etc. without assigning it. Worse, if someone tries to assign it to work around this, they will hit an unused variable warning.
I really think there needs to be a separate project, perhaps depending on pyflakes, for these "probably wrong, but not technically wrong" errors. They don't really fit pep8 (they aren't in the PEP, and aren't really style issues), but they don't follow pyflakes' "no false positives" rule.
I tend to agree with @asmeurer here. This can be developed as a flake8 plugin easily. Otherwise, I don't think it makes sense for here or pep8.
This is a flake8 plugin I would use! But it should be on an opt-in basis because it presents a design problem, not a clear-cut error. As such, it doesn't belong to pyflakes (as discussed at length in #59).
As I understand the problem, it trips up people coming from Ruby, expecting the last expression in a function to be returned from the block. Other usages are harder to judge. There's many examples of code using list comprehensions as single-line for-loops. It's questionable style but it's not technically incorrect.
Does this have any tests asserting that the new functionality works?
What's the reason for the special doctest cases in the tests?
I also wonder if in some instances an apparently useless expression is being used to invoke a side effect of some object with overridden operators. Some people think < and > look like arrows, for example, and so override these operators to mean something like "put into". I'd like to see this change run against some existing code to see if this will be a problem.
