Source code for wemake_python_styleguide.violations.refactoring

These checks ensure that you don't have patterns that can be refactored.

There are so many ways of doing the same thing in Python.
Here we collect know patterns that can be rewritten into
much easier or just more pythonic version.

.. currentmodule:: wemake_python_styleguide.violations.refactoring


.. autosummary::


Refactoring opportunities

.. autoclass:: UselessLoopElseViolation
.. autoclass:: UselessFinallyViolation
.. autoclass:: SimplifiableIfViolation
.. autoclass:: UselessReturningElseViolation
.. autoclass:: NegatedConditionsViolation
.. autoclass:: NestedTryViolation
.. autoclass:: UselessLambdaViolation
.. autoclass:: UselessLenCompareViolation
.. autoclass:: NotOperatorWithCompareViolation
.. autoclass:: NestedTernaryViolation
.. autoclass:: WrongInCompareTypeViolation
.. autoclass:: UnmergedIsinstanceCallsViolation
.. autoclass:: WrongIsinstanceWithTupleViolation
.. autoclass:: ImplicitElifViolation
.. autoclass:: ImplicitInConditionViolation
.. autoclass:: OpenWithoutContextManagerViolation
.. autoclass:: TypeCompareViolation
.. autoclass:: PointlessStarredViolation
.. autoclass:: ImplicitEnumerateViolation
.. autoclass:: ImplicitSumViolation
.. autoclass:: FalsyConstantCompareViolation
.. autoclass:: WrongIsCompareViolation
.. autoclass:: ImplicitPrimitiveViolation
.. autoclass:: AlmostSwappedViolation
.. autoclass:: MisrefactoredAssignmentViolation
.. autoclass:: InCompareWithSingleItemContainerViolation
.. autoclass:: ImplicitYieldFromViolation
.. autoclass:: NotATupleArgumentViolation
.. autoclass:: ImplicitItemsIteratorViolation
.. autoclass:: ImplicitDictGetViolation
.. autoclass:: ImplicitNegativeIndexViolation
.. autoclass:: SimplifiableReturningIfViolation


from typing_extensions import final

from wemake_python_styleguide.violations.base import (

[docs]@final class UselessLoopElseViolation(ASTViolation): """ Forbid ``else`` without ``break`` in a loop. We use the same logic for ``for`` and ``while`` loops. Reasoning: When there's no ``break`` keyword in loop's body it means that ``else`` will always be called. This rule will reduce complexity, improve readability, and protect from possible errors. Solution: Refactor your ``else`` case logic to be inside the loop's body. Or right after it. Example:: # Correct: for letter in 'abc': if letter == 'b': break else: print('"b" is not found') for letter in 'abc': print(letter) print('always called') # Wrong: for letter in 'abc': print(letter) else: print('always called') .. versionadded:: 0.3.0 .. versionchanged:: 0.11.0 """ error_template = 'Found `else` in a loop without `break`' code = 500 previous_codes = {436}
[docs]@final class UselessFinallyViolation(ASTViolation): """ Forbid ``finally`` in ``try`` block without ``except`` block. However, we allow to use ``try`` with just ``finally`` block when function or method is decorated. Because we cannot control what is going on in this decorator. It might be ``@contextmanager`` or similar thing that requires this API. Reasoning: This rule will reduce complexity and improve readability. Solution: Refactor your ``try`` logic. Replace the ``try-finally`` statement with a ``with`` statement. Example:: # Correct: with open("filename") as f: f.write(...) # Wrong: try: f = open("filename") f.write(...) finally: f.close() .. versionadded:: 0.3.0 .. versionchanged:: 0.11.0 .. versionchanged:: 0.14.0 """ error_template = 'Found `finally` in `try` block without `except`' code = 501 previous_codes = {437}
[docs]@final class SimplifiableIfViolation(ASTViolation): """ Forbid simplifiable ``if`` conditions. Reasoning: These complex constructions can cause frustration among other developers. They are longer, more verbose, and more complex. Solution: Either use ``bool()`` to convert test values to boolean values, or just leave it as it is in case your test already returns a boolean value. Use can also use ``not`` keyword to switch boolean values. Example:: # Correct: my_bool = bool(some_call()) other_value = 8 if some_call() else None # Wrong: my_bool = True if some_call() else False We only check ``if`` nodes where ``True`` and ``False`` values are used. We check both ``if`` nodes and ``if`` expressions. .. versionadded:: 0.7.0 .. versionchanged:: 0.11.0 """ error_template = 'Found simplifiable `if` condition' code = 502 previous_codes = {451}
[docs]@final class UselessReturningElseViolation(ASTViolation): """ Forbid useless ``else`` cases in returning functions. We check single ``if``, ``for``, ``while``, and ``try`` statements that all contain ``return``, ``raise``, ``continue``, or ``break`` statements with this rule. Reasoning: Using extra ``else`` creates a situation when the whole node could and should be dropped without any changes in logic. So, we prefer to have less code than more code. Solution: Remove useless ``else`` case. Example:: # Correct: def some_function(): if some_call(): return 'yeap' return 'nope' # Wrong: def some_function(): if some_call(): raise ValueError('yeap') else: raise ValueError('nope') .. versionadded:: 0.7.0 .. versionchanged:: 0.11.0 .. versionchanged:: 0.15.1 """ error_template = 'Found useless returning `else` statement' code = 503 previous_codes = {457}
[docs]@final class NegatedConditionsViolation(ASTViolation): """ Forbid negated conditions together with ``else`` clause. Reasoning: It easier to read and name regular conditions. Not negated ones. Solution: Move actions from the negated ``if`` condition to the ``else`` condition. Example:: # Correct: if some == 1: ... else: ... if not some: ... if not some: ... elif other: ... # Wrong: if not some: ... else: ... .. versionadded:: 0.8.0 .. versionchanged:: 0.11.0 """ error_template = 'Found negated condition' code = 504 previous_codes = {463}
[docs]@final class NestedTryViolation(ASTViolation): """ Forbid nested ``try`` blocks. Notice, we check all possible slots for ``try`` block: 1. the ``try`` block itself 2. all ``except`` cases 3. ``else`` case 4. and ``finally`` case Reasoning: Nesting ``try`` blocks indicates that something really bad happens to your logic. Why does it require two separate exception handlers? It is a perfect case to refactor your code. Solution: Collapse two exception handlers together. Or create a separate function that will handle this second nested case. Example:: # Wrong: try: try: ... except SomeException: ... except SomeOtherException: ... try: ... except SomeOtherException: try: ... except SomeException: ... .. versionadded:: 0.8.0 .. versionchanged:: 0.11.0 """ error_template = 'Found nested `try` block' code = 505 previous_codes = {464}
[docs]@final class UselessLambdaViolation(ASTViolation): """ Forbid useless proxy ``lambda`` expressions. Reasoning: Sometimes developers tend to overuse ``lambda`` expressions and they wrap code that can be passed as is, without extra wrapping. The code without extra ``lambda`` is easier to read and is more performant. Solution: Remove wrapping ``lambda`` declaration, use just the internal function. Example:: # Correct: numbers = map(int, ['1', '2']) # Wrong: numbers = map(lambda string: int(string), ['1', '2']) .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = 'Found useless lambda declaration' code = 506 previous_codes = {467}
[docs]@final class UselessLenCompareViolation(ASTViolation): """ Forbid unpythonic zero-length compare. Note, that we allow to check arbitrary length, like ``len(arr) == 3``. Reasoning: Python's structures like dicts, lists, sets, and tuples all have ``__bool__`` method to checks their length. So, there's no point in wrapping them into ``len(...)`` and checking that it is bigger that ``0`` or less then ``1``, etc. Solution: Remove extra ``len()`` call. Example:: # Correct: if some_array or not other_array or len(third_array) == 1: ... # Wrong: if len(some_array) > 0 or len(other_array) < 1: ... .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = 'Found useless `len()` compare' code = 507 previous_codes = {468}
[docs]@final class NotOperatorWithCompareViolation(ASTViolation): """ Forbid ``not`` with compare expressions. Reasoning: This version of ``not`` operator is unreadable. Solution: Refactor the expression without ``not`` operator. Change the compare signs. Example:: # Correct: if x <= 5: ... # Wrong: if not x > 5: ... .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = 'Found incorrect `not` with compare usage' code = 508 previous_codes = {470}
[docs]@final class NestedTernaryViolation(ASTViolation): """ Forbid nesting ternary expressions in certain places. Note, that we restrict to nest ternary expressions inside: - ``if`` conditions - boolean and binary operations like ``and`` or ``+`` - unary operators Reasoning: Nesting ternary in random places can lead to very hard debug and testing problems. Solution: Refactor the ternary expression to be either a new variable, or nested ``if`` statement, or a new function. Example:: # Correct: some = x if cond() else y # Wrong: if x if cond() else y: ... .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = 'Found incorrectly nested ternary' code = 509 previous_codes = {472}
[docs]@final class WrongInCompareTypeViolation(ASTViolation): """ Forbid ``in`` with static containers except ``set`` nodes. We enforce people to use sets as a static containers. You can also use variables, calls, methods, etc. Dynamic values are not checked. Reasoning: Using static ``list``, ``tuple``, or ``dict`` elements to check that some element is inside the container is a bad practice. Because we need to iterate all over the container to find the element. Sets are the best suit for this task. Moreover, it makes your code consistent. Solution: Use ``set`` elements or comprehensions to check that something is contained in a container. Example:: # Correct: print(needle in {'one', 'two'}) # Wrong: print(needle in ['one', 'two']) .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 .. versionchanged:: 0.14.0 """ error_template = 'Found `in` used with a non-set container' code = 510 previous_codes = {473}
[docs]@final class UnmergedIsinstanceCallsViolation(ASTViolation): """ Forbid multiple ``isinstance`` calls on the same variable. Reasoning: The best practice is to use ``isinstance`` with tuple as the second argument, instead of multiple conditions joined with ``or``. Solution: Use tuple of types as the second argument. Example:: # Correct: isinstance(some, (int, float)) # Wrong: isinstance(some, int) or isinstance(some, float) See also: .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = ( 'Found separate `isinstance` calls that can be merged for: {0}' ) code = 511 previous_codes = {474}
[docs]@final class WrongIsinstanceWithTupleViolation(ASTViolation): """ Forbid multiple ``isinstance`` calls with single-item tuples. Reasoning: There's no need to use tuples with single elements. You can use single variables or tuples with multiple elements. Solution: Use tuples with multiple elements or a single variable. Example:: # Correct: isinstance(some, (int, float)) isinstance(some, int) # Wrong: isinstance(some, (int, )) See: .. versionadded:: 0.10.0 .. versionchanged:: 0.11.0 """ error_template = 'Found `isinstance` call with a single element tuple' code = 512 previous_codes = {475}
[docs]@final class ImplicitElifViolation(TokenizeViolation): """ Forbid implicit ``elif`` conditions. Reasoning: Nested ``if`` in ``else`` cases are bad for readability because of the nesting level. Solution: Use ``elif`` on the same level. Example:: # Correct: if some: ... elif other: ... # Wrong: if some: ... else: if other: ... .. versionadded:: 0.12.0 """ error_template = 'Found implicit `elif` condition' code = 513
[docs]@final class ImplicitInConditionViolation(ASTViolation): """ Forbid multiple equality comparisons with the same variable. Reasoning: Using double+ equality compare with ``or`` or double+ non-equality compare with ``and`` indicates that you have implicit ``in`` or ``not in`` condition. It is just hidden from you. Solution: Refactor compares to use ``in`` or ``not in`` clauses. Example:: # Correct: print(some in {'first', 'second'}) print(some not in {'first', 'second'}) # Wrong: print(some == 'first' or some == 'second') print(some != 'first' and some != 'second') .. versionadded:: 0.10.0 .. versionchanged:: 0.12.0 """ code = 514 error_template = 'Found implicit `in` condition' previous_codes = {336}
[docs]@final class OpenWithoutContextManagerViolation(ASTViolation): """ Forbid ``open()`` without a context manager. Reasoning: When you ``open()`` something, you need to close it. When using a context manager - it is automatically done for you. When not using it - you might find yourself in a situation when file is not closed and is not accessible anymore. Solution: Refactor ``open()`` call to use ``with``. Example:: # Correct: with open(filename) as file_obj: ... # Wrong: file_obj = open(filename) .. versionadded:: 0.12.0 """ code = 515 error_template = 'Found `open()` used without a context manager'
[docs]@final class TypeCompareViolation(ASTViolation): """ Forbid comparing types with ``type()`` function. Reasoning: When you compare types with ``type()`` function call it means that you break polymorphism and disallow child classes of a node to work here. That's incorrect. Solution: Use ``isinstance`` to compare types. Example:: # Correct: print(something, type(something)) if isinstance(something, int): ... # Wrong: if type(something) == int: ... .. versionadded:: 0.12.0 """ code = 516 error_template = 'Found `type()` used to compare types'
[docs]@final class PointlessStarredViolation(ASTViolation): """ Forbid useless starred expressions. Reasoning: Using starred expression with constants is useless. This piece of code can be rewritten to be flat. Eg.: ``print(*[1, 2, 3])`` is ``print(1, 2, 3)``. Solution: Refactor your code not to use starred expressions with ``list``, ``dict``, ``tuple``, and ``set`` constants. Use regular argument passing instead. Example:: # Correct: my_list = [1, 2, 3, *other_iterable] # Wrong: print(*[1, 2, 3], **{{}}) .. versionadded:: 0.12.0 """ code = 517 error_template = 'Found pointless starred expression'
[docs]@final class ImplicitEnumerateViolation(ASTViolation): """ Forbid implicit ``enumerate()`` calls. Reasoning: Using ``range(len(...))`` is not pythonic. Python uses collection iterators, not index-based loops. Solution: Use ``enumerate(...)`` instead of ``range(len(...))``. Example:: # Correct: for index, person in enumerate(people): ... # Wrong: for index in range(len(people)): ... See also: .. versionadded:: 0.12.0 """ code = 518 error_template = 'Found implicit `enumerate()` call'
[docs]@final class ImplicitSumViolation(ASTViolation): """ Forbid implicit ``sum()`` calls. When summing types different from numbers, you might need to provide the second argument to the ``sum`` function: ``sum([[1], [2], [3]], [])`` You might also use ``str.join`` to join iterable of strings. Reasoning: Using ``for`` loops with ``+=`` assign inside indicates that you iteratively sum things inside your collection. That's what ``sum()`` builtin function does. Solution: Use ``sum(...)`` instead of a loop with ``+=`` operation. Example:: # Correct: sum_result = sum(get_elements()) # Wrong: sum_result = 0 for to_sum in get_elements(): sum_result += to_sum See also: .. versionadded:: 0.12.0 """ code = 519 error_template = 'Found implicit `sum()` call'
[docs]@final class FalsyConstantCompareViolation(ASTViolation): """ Forbid comparing with explicit falsy constants. We allow to compare with falsy numbers, strings, booleans, ``None``. We disallow complex constants like tuple, dicts, and lists. Reasoning: When comparing ``something`` with explicit falsy constants what we really mean is ``not something``. Solution: Use ``not`` with your variable. Fix your data types. Example:: # Correct: if not my_check: ... if some_other is None: ... if some_num == 0: ... # Wrong: if my_check == []: ... .. versionadded:: 0.12.0 """ code = 520 error_template = 'Found compare with falsy constant'
[docs]@final class WrongIsCompareViolation(ASTViolation): """ Forbid comparing values with constants using ``is`` or ``is not``. However, we allow to compare with ``None`` and booleans. Reasoning: ``is`` compares might not do what you want them to do. Firstly, they check for the same object, not equality. Secondly, they behave unexpectedly even with the simple values like ``257``. Solution: Use ``==`` to compare with constants. Example:: # Correct: if my_check == [1, 2, 3]: ... # Wrong: if my_check is [1, 2, 3]: ... See also: .. versionadded:: 0.12.0 """ code = 521 error_template = 'Found wrong `is` compare'
[docs]@final class ImplicitPrimitiveViolation(ASTViolation): """ Forbid implicit primitives in the form of ``lambda`` functions. Reasoning: When you use ``lambda`` that returns a primitive value and takes no arguments, it means that you should use a primitive type instead. Solution: Replace ``lambda`` with ``int``, ``float``, ``list``, or any other primitive. Example:: # Correct: defaultdict(int) # Wrong: defaultdict(lambda: 0) .. versionadded:: 0.13.0 """ code = 522 error_template = 'Found implicit primitive in a form of `lambda`'
[docs]@final class AlmostSwappedViolation(ASTViolation): """ Forbid unpythonic variable swaps. We check for ``a = b; b = a`` sequences. Reasoning: This looks like a failed attempt to swap. Solution: Use standard way to swap two variables. Example:: # Correct: a, b = b, a # Wrong: a = b b = a temp = a a = b b = temp .. versionadded:: 0.13.0 """ error_template = 'Found incorrectly swapped variables' code = 523
[docs]@final class MisrefactoredAssignmentViolation(ASTViolation): """ Forbid misrefactored self assignment. Reasoning: Self assignment does not need to have the same operand on the left hand side and on the right hand side. Solution: Refactor you code to use multiple self assignments or fix your code. Example:: # Correct: test += 1 test *= 2 # Wrong: test += test + 1 See :py:data:`~wemake_python_styleguide.constants.MATH_APPROXIMATE_CONSTANTS` for full list of math constants that we check for. .. versionadded:: 0.13.0 """ error_template = 'Found self assignment with refactored assignment' code = 524
[docs]@final class InCompareWithSingleItemContainerViolation(ASTViolation): """ Forbid comparisons where ``in`` is compared with single item container. Reasoning: ``in`` comparison with a container which contains only one item looks like overhead and unneeded complexity. Solution: Refactor your code to use ``==`` instead ``in``. Example:: # Correct: a == 's' # Wrong: a in {'s'} .. versionadded:: 0.13.0 """ error_template = 'Found wrong `in` compare with single item container' code = 525
[docs]@final class ImplicitYieldFromViolation(ASTViolation): """ Forbid ``yield`` inside ``for`` loop instead of ``yield from``. Reasoning: It is known that ``yield from`` is a semantically identical to a ``for`` loop with a ``yield`` inside. But, it is way more readable. Solution: Use ``yield from`` some iterable directly instead iterating over it inside a loop and ``yield`` it one by one. Example:: # Correct: yield from some() yield from ( value[index:index + chunk_size] for index in range(0, len(value), chunk_size) ) # Wrong: for index in chunk: yield index .. versionadded:: 0.13.0 """ error_template = 'Found implicit `yield from` usage' code = 526
[docs]@final class NotATupleArgumentViolation(ASTViolation): """ Require tuples as arguments for certain functions. Reasoning: For some functions, it is better to use tuples instead of another iterable types (list, sets,...) as arguments. Solution: Use tuples as arguments. Example:: # Correct: a = frozenset((2,)) # Wrong: a = frozenset([2]) See :py:data:`~wemake_python_styleguide.constants.TUPLE_ARGUMENTS_METHODS` for full list of methods that we check for. .. versionadded:: 0.13.0 """ error_template = 'Found not a tuple used as an argument' code = 527
[docs]@final class ImplicitItemsIteratorViolation(ASTViolation): """ Forbid implicit ``.items()`` iterator. Reasoning: When iterating over collection it is easy to forget to use ``.items()`` when you need to access both keys and values. So, when you access the iterable with the key inside a ``for`` loop, that's a sign to refactor your code. Solution: Use ``.items()`` with direct keys and values when you need them. Example:: # Correct: for some_key, some_value in collection.items(): print(some_key, some_value) # Wrong: for some_key in collection: print(some_key, collection[some_key]) .. versionadded:: 0.13.0 """ error_template = 'Found implicit `.items()` usage' code = 528
[docs]@final class ImplicitDictGetViolation(ASTViolation): """ Forbid implicit ``.get()`` dict method. Reasoning: When using ``in`` with a dict key it is hard to keep the code clean. It is more convenient to use ``.get()`` and check for ``None`` later. Solution: Use ``.get()`` with the key you need. Check for ``None`` in case you need it, or just act with the default value of the same type. Example:: # Correct: value = collection.get(key) if value is not None: print(value) # Wrong: if key in collection: print(collection[key]) .. versionadded:: 0.13.0 """ error_template = 'Found implicit `.get()` dict usage' code = 529
[docs]@final class ImplicitNegativeIndexViolation(ASTViolation): """ Forbid implicit negative indexes. Reasoning: There's no need in getting the length of an iterable and then having a negative offset, when you can specify negative indexes in the first place. Solution: Use negative indexes. Example:: # Correct: some_list[-1] # Wrong: some_list[len(some_list) - 1] .. versionadded:: 0.13.0 """ error_template = 'Found implicit negative index' code = 530
[docs]@final class SimplifiableReturningIfViolation(ASTViolation): """ Forbid if statements that simply return booleans in functions or methods. Reasoning: There is no need to test a condition and simply return a boolean depending on its outcome if there is not going to be any additional code. Solution: Instead of testing the condition and returning a boolean, return the condition itself. This applies to early returning ifs too. Example:: # Correct: def some_function(): return some_condition # Wrong: def some_function(): if some_condition: return True else: return False .. versionadded:: 0.15.0 """ error_template = 'Found simplifiable returning `if` condition in a function' code = 531