path: root/pylint/checkers/refactoring/recommendation_checker.py

# Licensed under the GPL: https://www.gnu.org/licenses/old-licenses/gpl-2.0.html
# For details: https://github.com/pylint-dev/pylint/blob/main/LICENSE
# Copyright (c) https://github.com/pylint-dev/pylint/blob/main/CONTRIBUTORS.txt

from __future__ import annotations

import astroid
from astroid import nodes

from pylint import checkers
from pylint.checkers import utils
from pylint.interfaces import HIGH, INFERENCE


class RecommendationChecker(checkers.BaseChecker):
    name = "refactoring"
    msgs = {
        "C0200": (
            "Consider using enumerate instead of iterating with range and len",
            "consider-using-enumerate",
            "Emitted when code that iterates with range and len is "
            "encountered. Such code can be simplified by using the "
            "enumerate builtin.",
        ),
        "C0201": (
            "Consider iterating the dictionary directly instead of calling .keys()",
            "consider-iterating-dictionary",
            "Emitted when the keys of a dictionary are iterated through the ``.keys()`` "
            "method or when ``.keys()`` is used for a membership check. "
            "It is enough to iterate through the dictionary itself, "
            "``for key in dictionary``. For membership checks, "
            "``if key in dictionary`` is faster.",
        ),
        "C0206": (
            "Consider iterating with .items()",
            "consider-using-dict-items",
            "Emitted when iterating over the keys of a dictionary and accessing the "
            "value by index lookup. "
            "Both the key and value can be accessed by iterating using the .items() "
            "method of the dictionary instead.",
        ),
        "C0207": (
            "Use %s instead",
            "use-maxsplit-arg",
            "Emitted when accessing only the first or last element of str.split(). "
            "The first and last element can be accessed by using "
            "str.split(sep, maxsplit=1)[0] or str.rsplit(sep, maxsplit=1)[-1] "
            "instead.",
        ),
        "C0208": (
            "Use a sequence type when iterating over values",
            "use-sequence-for-iteration",
            "When iterating over values, sequence types (e.g., ``lists``, ``tuples``, ``ranges``) "
            "are more efficient than ``sets``.",
        ),
        "C0209": (
            "Formatting a regular string which could be a f-string",
            "consider-using-f-string",
            "Used when we detect a string that is being formatted with format() or % "
            "which could potentially be a f-string. The use of f-strings is preferred. "
            "Requires Python 3.6 and ``py-version >= 3.6``.",
        ),
    }

    def open(self) -> None:
        py_version = self.linter.config.py_version
        self._py36_plus = py_version >= (3, 6)

    @staticmethod
    def _is_builtin(node: nodes.NodeNG, function: str) -> bool:
        inferred = utils.safe_infer(node)
        if not inferred:
            return False
        return utils.is_builtin_object(inferred) and inferred.name == function

    @utils.only_required_for_messages(
        "consider-iterating-dictionary", "use-maxsplit-arg"
    )
    def visit_call(self, node: nodes.Call) -> None:
        self._check_consider_iterating_dictionary(node)
        self._check_use_maxsplit_arg(node)

    def _check_consider_iterating_dictionary(self, node: nodes.Call) -> None:
        if not isinstance(node.func, nodes.Attribute):
            return
        if node.func.attrname != "keys":
            return

        if isinstance(node.parent, nodes.BinOp) and node.parent.op in {"&", "|", "^"}:
            return

        comp_ancestor = utils.get_node_first_ancestor_of_type(node, nodes.Compare)
        if (
            isinstance(node.parent, (nodes.For, nodes.Comprehension))
            or comp_ancestor
            and any(
                op
                for op, comparator in comp_ancestor.ops
                if op in {"in", "not in"}
                and (comparator in node.node_ancestors() or comparator is node)
            )
        ):
            inferred = utils.safe_infer(node.func)
            if not isinstance(inferred, astroid.BoundMethod) or not isinstance(
                inferred.bound, nodes.Dict
            ):
                return
            self.add_message(
                "consider-iterating-dictionary", node=node, confidence=INFERENCE
            )

    def _check_use_maxsplit_arg(self, node: nodes.Call) -> None:
        """Add message when accessing first or last elements of a str.split() or
        str.rsplit().
        """

        # Check if call is split() or rsplit()
        if not (
            isinstance(node.func, nodes.Attribute)
            and node.func.attrname in {"split", "rsplit"}
            and isinstance(utils.safe_infer(node.func), astroid.BoundMethod)
        ):
            return
        inferred_expr = utils.safe_infer(node.func.expr)
        if isinstance(inferred_expr, astroid.Instance) and any(
            inferred_expr.nodes_of_class(nodes.ClassDef)
        ):
            return

        try:
            sep = utils.get_argument_from_call(node, 0, "sep")
        except utils.NoSuchArgumentError:
            return

        try:
            # Ignore if maxsplit arg has been set
            utils.get_argument_from_call(node, 1, "maxsplit")
            return
        except utils.NoSuchArgumentError:
            pass

        if isinstance(node.parent, nodes.Subscript):
            try:
                subscript_value = utils.get_subscript_const_value(node.parent).value
            except utils.InferredTypeError:
                return

            # Check for cases where variable (Name) subscripts may be mutated within a loop
            if isinstance(node.parent.slice, nodes.Name):
                # Check if loop present within the scope of the node
                scope = node.scope()
                for loop_node in scope.nodes_of_class((nodes.For, nodes.While)):
                    if not loop_node.parent_of(node):
                        continue

                    # Check if var is mutated within loop (Assign/AugAssign)
                    for assignment_node in loop_node.nodes_of_class(nodes.AugAssign):
                        if node.parent.slice.name == assignment_node.target.name:
                            return
                    for assignment_node in loop_node.nodes_of_class(nodes.Assign):
                        if node.parent.slice.name in [
                            n.name for n in assignment_node.targets
                        ]:
                            return

            if subscript_value in (-1, 0):
                fn_name = node.func.attrname
                new_fn = "rsplit" if subscript_value == -1 else "split"
                new_name = (
                    node.func.as_string().rsplit(fn_name, maxsplit=1)[0]
                    + new_fn
                    + f"({sep.as_string()}, maxsplit=1)[{subscript_value}]"
                )
                self.add_message("use-maxsplit-arg", node=node, args=(new_name,))

    @utils.only_required_for_messages(
        "consider-using-enumerate",
        "consider-using-dict-items",
        "use-sequence-for-iteration",
    )
    def visit_for(self, node: nodes.For) -> None:
        self._check_consider_using_enumerate(node)
        self._check_consider_using_dict_items(node)
        self._check_use_sequence_for_iteration(node)

    def _check_consider_using_enumerate(self, node: nodes.For) -> None:
        """Emit a convention whenever range and len are used for indexing."""
        # Verify that we have a `range([start], len(...), [stop])` call and
        # that the object which is iterated is used as a subscript in the
        # body of the for.

        # Is it a proper range call?
        if not isinstance(node.iter, nodes.Call):
            return
        if not self._is_builtin(node.iter.func, "range"):
            return
        if not node.iter.args:
            return
        is_constant_zero = (
            isinstance(node.iter.args[0], nodes.Const) and node.iter.args[0].value == 0
        )
        if len(node.iter.args) == 2 and not is_constant_zero:
            return
        if len(node.iter.args) > 2:
            return

        # Is it a proper len call?
        if not isinstance(node.iter.args[-1], nodes.Call):
            return
        second_func = node.iter.args[-1].func
        if not self._is_builtin(second_func, "len"):
            return
        len_args = node.iter.args[-1].args
        if not len_args or len(len_args) != 1:
            return
        iterating_object = len_args[0]
        if isinstance(iterating_object, nodes.Name):
            expected_subscript_val_type = nodes.Name
        elif isinstance(iterating_object, nodes.Attribute):
            expected_subscript_val_type = nodes.Attribute
        else:
            return
        # If we're defining __iter__ on self, enumerate won't work
        scope = node.scope()
        if (
            isinstance(iterating_object, nodes.Name)
            and iterating_object.name == "self"
            and scope.name == "__iter__"
        ):
            return

        # Verify that the body of the for loop uses a subscript
        # with the object that was iterated. This uses some heuristics
        # in order to make sure that the same object is used in the
        # for body.
        for child in node.body:
            for subscript in child.nodes_of_class(nodes.Subscript):
                if not isinstance(subscript.value, expected_subscript_val_type):
                    continue

                value = subscript.slice
                if not isinstance(value, nodes.Name):
                    continue
                if subscript.value.scope() != node.scope():
                    # Ignore this subscript if it's not in the same
                    # scope. This means that in the body of the for
                    # loop, another scope was created, where the same
                    # name for the iterating object was used.
                    continue
                if value.name == node.target.name and (
                    isinstance(subscript.value, nodes.Name)
                    and iterating_object.name == subscript.value.name
                    or isinstance(subscript.value, nodes.Attribute)
                    and iterating_object.attrname == subscript.value.attrname
                ):
                    self.add_message("consider-using-enumerate", node=node)
                    return

    def _check_consider_using_dict_items(self, node: nodes.For) -> None:
        """Add message when accessing dict values by index lookup."""
        # Verify that we have a .keys() call and
        # that the object which is iterated is used as a subscript in the
        # body of the for.

        iterating_object_name = utils.get_iterating_dictionary_name(node)
        if iterating_object_name is None:
            return

        # Verify that the body of the for loop uses a subscript
        # with the object that was iterated. This uses some heuristics
        # in order to make sure that the same object is used in the
        # for body.
        for child in node.body:
            for subscript in child.nodes_of_class(nodes.Subscript):
                if not isinstance(subscript.value, (nodes.Name, nodes.Attribute)):
                    continue

                value = subscript.slice
                if (
                    not isinstance(value, nodes.Name)
                    or value.name != node.target.name
                    or iterating_object_name != subscript.value.as_string()
                ):
                    continue
                last_definition_lineno = value.lookup(value.name)[1][-1].lineno
                if last_definition_lineno > node.lineno:
                    # Ignore this subscript if it has been redefined after
                    # the for loop. This checks for the line number using .lookup()
                    # to get the line number where the iterating object was last
                    # defined and compare that to the for loop's line number
                    continue
                if (
                    isinstance(subscript.parent, nodes.Assign)
                    and subscript in subscript.parent.targets
                    or isinstance(subscript.parent, nodes.AugAssign)
                    and subscript == subscript.parent.target
                ):
                    # Ignore this subscript if it is the target of an assignment
                    # Early termination as dict index lookup is necessary
                    return

                self.add_message("consider-using-dict-items", node=node)
                return

    @utils.only_required_for_messages(
        "consider-using-dict-items",
        "use-sequence-for-iteration",
    )
    def visit_comprehension(self, node: nodes.Comprehension) -> None:
        self._check_consider_using_dict_items_comprehension(node)
        self._check_use_sequence_for_iteration(node)

    def _check_consider_using_dict_items_comprehension(
        self, node: nodes.Comprehension
    ) -> None:
        """Add message when accessing dict values by index lookup."""
        iterating_object_name = utils.get_iterating_dictionary_name(node)
        if iterating_object_name is None:
            return

        for child in node.parent.get_children():
            for subscript in child.nodes_of_class(nodes.Subscript):
                if not isinstance(subscript.value, (nodes.Name, nodes.Attribute)):
                    continue

                value = subscript.slice
                if (
                    not isinstance(value, nodes.Name)
                    or value.name != node.target.name
                    or iterating_object_name != subscript.value.as_string()
                ):
                    continue

                self.add_message("consider-using-dict-items", node=node)
                return

    def _check_use_sequence_for_iteration(
        self, node: nodes.For | nodes.Comprehension
    ) -> None:
        """Check if code iterates over an in-place defined set.

        Sets using `*` are not considered in-place.
        """
        if isinstance(node.iter, nodes.Set) and not any(
            utils.has_starred_node_recursive(node)
        ):
            self.add_message(
                "use-sequence-for-iteration", node=node.iter, confidence=HIGH
            )

    @utils.only_required_for_messages("consider-using-f-string")
    def visit_const(self, node: nodes.Const) -> None:
        if self._py36_plus:
            # f-strings require Python 3.6
            if node.pytype() == "builtins.str" and not isinstance(
                node.parent, nodes.JoinedStr
            ):
                self._detect_replacable_format_call(node)

    def _detect_replacable_format_call(self, node: nodes.Const) -> None:
        """Check whether a string is used in a call to format() or '%' and whether it
        can be replaced by an f-string.
        """
        if (
            isinstance(node.parent, nodes.Attribute)
            and node.parent.attrname == "format"
        ):
            # Don't warn on referencing / assigning .format without calling it
            if not isinstance(node.parent.parent, nodes.Call):
                return

            if node.parent.parent.args:
                for arg in node.parent.parent.args:
                    # If star expressions with more than 1 element are being used
                    if isinstance(arg, nodes.Starred):
                        inferred = utils.safe_infer(arg.value)
                        if (
                            isinstance(inferred, astroid.List)
                            and len(inferred.elts) > 1
                        ):
                            return
                    # Backslashes can't be in f-string expressions
                    if "\\" in arg.as_string():
                        return

            elif node.parent.parent.keywords:
                keyword_args = [
                    i[0] for i in utils.parse_format_method_string(node.value)[0]
                ]
                for keyword in node.parent.parent.keywords:
                    # If keyword is used multiple times
                    if keyword_args.count(keyword.arg) > 1:
                        return

                    keyword = utils.safe_infer(keyword.value)

                    # If lists of more than one element are being unpacked
                    if isinstance(keyword, nodes.Dict):
                        if len(keyword.items) > 1 and len(keyword_args) > 1:
                            return

            # If all tests pass, then raise message
            self.add_message(
                "consider-using-f-string",
                node=node,
                line=node.lineno,
                col_offset=node.col_offset,
            )

        elif isinstance(node.parent, nodes.BinOp) and node.parent.op == "%":
            # Backslashes can't be in f-string expressions
            if "\\" in node.parent.right.as_string():
                return

            # If % applied to another type than str, it's modulo and can't be replaced by formatting
            if not hasattr(node.parent.left, "value") or not isinstance(
                node.parent.left.value, str
            ):
                return

            inferred_right = utils.safe_infer(node.parent.right)

            # If dicts or lists of length > 1 are used
            if isinstance(inferred_right, nodes.Dict):
                if len(inferred_right.items) > 1:
                    return
            elif isinstance(inferred_right, nodes.List):
                if len(inferred_right.elts) > 1:
                    return

            # If all tests pass, then raise message
            self.add_message(
                "consider-using-f-string",
                node=node,
                line=node.lineno,
                col_offset=node.col_offset,
            )