pocketutils.core.dot_dict.NestedDotDict.write_toml() - Code Metrics - Inspection of "fix: misc" - dmyersturnbull/pocketutils - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Push — main ( 393c8e...520e83 )

by Douglas

created 2021-10-28 22:38 UTC

NestedDotDict.write_toml() A

↳ Parent: pocketutils.core.dot_dict

Complexity

Conditions

Size

Total Lines	2
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	1
eloc	2
nop	2
dl	0
loc	2
rs	10
c	0
b	0
f	0

from __future__ import annotations


import gzip

import json
import pickle
from copy import copy
from datetime import date, datetime
from pathlib import Path, PurePath
from typing import Any, ByteString, Callable, Mapping, Optional, Sequence
from typing import Tuple as Tup
from typing import Type, TypeVar, Union

import orjson

import toml


from pocketutils.core import PathLike

from pocketutils.core._internal import read_txt_or_gz, write_txt_or_gz

from pocketutils.core.exceptions import XKeyError, XTypeError, XValueError


PICKLE_PROTOCOL = 5
T = TypeVar("T")



def _json_encode_default(obj: Any) -> Any:

    if isinstance(obj, NestedDotDict):
        # noinspection PyProtectedMember
        return dict(obj._x)
class MyParent:
    def __init__(self):
        self._x = 1;
        self.y = 2;

class MyChild(MyParent):
    def some_method(self):
        return self._x    # Ok, since accessed from a child class

class AnotherClass:
    def some_method(self, instance_of_my_child):
        return instance_of_my_child._x   # Would be flagged as AnotherClass is not
                                         # a child class of MyParent


class NestedDotDict(Mapping):

    """
    A thin wrapper around a nested dict to make getting values easier.
    This was designed as a wrapper for TOML, but it works more generally too.

    Keys must be strings that do not contain a dot (.).
    A dot is reserved for splitting values to traverse the tree.
    For example, ``dotdict["pet.species.name"]``.

    """

    @classmethod
    def read_toml(cls, path: PathLike) -> NestedDotDict:

        return NestedDotDict(toml.loads(read_txt_or_gz(path)))

    @classmethod
    def read_json(cls, path: Union[PurePath, str]) -> NestedDotDict:
        """
        Reads JSON from a file, into a NestedDotDict.
        If the JSON data is a list type, converts into a dict with the key ``data``.
        Can read .json or .json.gz.
        """
        data = orjson.loads(read_txt_or_gz(path))
        if isinstance(data, list):
            data = {"data": data}
        return cls(data)

    @classmethod
    def read_pickle(cls, path: Union[PurePath, str]) -> NestedDotDict:
        """

        Note that this function has potential security concerns.
        This is because it relies on the pickle module.
        """
        data = Path(path).read_bytes()
        data = pickle.loads(data)  # nosec
        return cls(data)

    @classmethod
    def parse_toml(cls, data: str) -> NestedDotDict:

        return cls(toml.loads(data))

    @classmethod
    def parse_json(cls, data: str) -> NestedDotDict:
        """
        Parses JSON from a string, into a NestedDotDict.
        If the JSON data is a list type, converts into a dict with the key ``data``.
        """
        data = json.loads(data)
        if isinstance(data, list):
            data = {"data": data}
        return cls(data)

    @classmethod
    def parse_pickle(cls, data: ByteString) -> NestedDotDict:

        if not isinstance(data, bytes):
            data = bytes(data)
        return NestedDotDict(pickle.loads(data))

    def __init__(self, x: Mapping[str, Any]) -> None:
class SomeParent:
    def __init__(self):
        self.x = 1

class SomeChild(SomeParent):
    def __init__(self):
        # Initialize the super class
        SomeParent.__init__(self)
        """
        Constructor.

        Raises:
            XValueError: If a key (in this dict or a sub-dict) is not a str or contains a dot
        """
        if not (hasattr(x, "items") and hasattr(x, "keys") and hasattr(x, "values")):
            raise XTypeError(
                f"Type {type(x)} for value {x} appears not to be dict-like", actual=str(type(x))
            )
        bad = [k for k in x if not isinstance(k, str)]
        if len(bad) > 0:
            raise XValueError(f"Keys were not strings for these values: {bad}", value=bad)
        bad = [k for k in x if "." in k]
        if len(bad) > 0:
            raise XValueError(f"Keys contained dots (.) for these values: {bad}", value=bad)
        self._x = x
        # Let's make sure this constructor gets called on subdicts:
        self.leaves()

    def write_json(self, path: PathLike, *, indent: bool = False) -> None:

        write_txt_or_gz(self.to_json(indent=indent), path)

    def write_toml(self, path: PathLike) -> None:

        write_txt_or_gz(self.to_toml(), path)

    def write_pickle(self, path: PathLike) -> None:

        Path(path).write_bytes(pickle.dumps(self._x, protocol=PICKLE_PROTOCOL))

    def to_json(self, *, indent: bool = False) -> str:

        kwargs = dict(option=orjson.OPT_INDENT_2) if indent else {}
        encoded = orjson.dumps(self._x, default=_json_encode_default, **kwargs)
        return encoded.decode(encoding="utf8")

    def to_toml(self) -> str:

        return toml.dumps(self._x)

    def leaves(self) -> Mapping[str, Any]:
        """
        Gets the leaves in this tree.

        Returns:
            A dict mapping dot-joined keys to their values
        """
        mp = {}

        for key, value in self._x.items():
            if len(key) == 0:
                raise AssertionError(f"Key is empty (value={value})")
            if isinstance(value, dict):
                mp.update({key + "." + k: v for k, v in NestedDotDict(value).leaves().items()})
            else:
                mp[key] = value
        return mp

    def sub(self, items: str) -> NestedDotDict:

        return NestedDotDict(self[items])

    def exactly(self, items: str, astype: Type[T]) -> T:
        """
        Gets the key ``items`` from the dict if it has type ``astype``.
        Calling ``dotdict.exactly(k, t) is equivalent to calling ``t(dotdict[k])``,
        but a raised ``TypeError`` will note the key, making this a useful shorthand for the above within a try-except.


        Args:
            items: The key hierarchy, with a dot (.) as a separator
            astype: The type, which will be checked using ``isinstance``

        Returns:
            The value in the required type

        Raises:
            TypeError: If not ``isinstance(value, astype)``
        """
        z = self[items]

        if not isinstance(z, astype):
            raise XTypeError(
                f"Value {z} from {items} is a {type(z)}, not {astype}",
                actual=str(type(z)),
                expected=str(astype),
            )
        return z

    def get_as(
        self, items: str, astype: Callable[[Any], T], default: Optional[T] = None

    ) -> Optional[T]:
        """
        Gets the value of an *optional* key, or ``default`` if it doesn't exist.
        Also see ``req_as``.

        Args:
            items: The key hierarchy, with a dot (.) as a separator.
                   Ex: ``animal.species.name``.
            astype: Any function that converts the found value to type ``T``.
                    Can be a ``Type``, such as ``int``.
                    Despite the annotated type, this function only needs to accept the actual value of the key

                    as input, not ``Any``.
            default: Return this value if the key is not found (at any level)

        Returns:
            The value of found key in this dot-dict, or ``default``.

        Raises:
            XValueError: Likely exception raised if calling ``astype`` fails
        """
        x = self.get(items)

        if x is None:
            return default
        if astype is date:
            return self._to_date(x)
        if astype is datetime:
            return self._to_datetime(x)
        return astype(x)

    def req_as(self, items: str, astype: Optional[Callable[[Any], T]]) -> T:
        """
        Gets the value of a *required* key.
        Also see ``get_as`` and ``exactly``.

        Args:
            items: The key hierarchy, with a dot (.) as a separator.
                   Ex: ``animal.species.name``.
            astype: Any function that converts the found value to type ``T``.
                    Can be a ``Type``, such as ``int``.
                    Despite the annotated type, this function only needs to accept the actual value of the key

                    as input, not ``Any``.

        Returns:
            The value of found key in this dot-dict.

        Raises:
            XKeyError: If the key is not found (at any level).
            XValueError: Likely exception raised if calling ``astype`` fails
        """
        x = self[items]

        return astype(x)

    def get_list_as(
        self, items: str, astype: Callable[[Any], T], default: Optional[Sequence[T]] = None

    ) -> Optional[Sequence[T]]:
        """
        Gets list values from an *optional* key.
        Note that ``astype`` here converts elements *within* the list, not the whole list.
        Also see ``req_list_as``.

        Args:
            items: The key hierarchy, with a dot (.) as a separator. Ex: ``animal.species.name``.
            astype: Any function that converts the found value to type ``T``. Ex: ``int``.
            default: Return this value if the key wasn't found

        Returns:
            ``[astype(v) for v in self[items]]``, or ``default`` if ``items`` was not found.

        Raises:
            XValueError: Likely exception raised if calling ``astype`` fails
            XTypeError: If the found value is not a (non-``str``) ``Sequence``
        """
        x = self.get(items)

        if x is None:
            return default
        if not isinstance(x, Sequence) or isinstance(x, str):

            raise XTypeError(f"Value {x} is not a list for lookup {items}", actual=str(type(x)))
        return [astype(y) for y in x]

    def req_list_as(self, items: str, astype: Optional[Callable[[Any], T]]) -> Sequence[T]:
        """
        Gets list values from a *required* key.
        Note that ``astype`` here converts elements *within* the list, not the whole list.
        Also see ``get_list_as``.

        Args:
            items: The key hierarchy, with a dot (.) as a separator. Ex: ``animal.species.name``.
            astype: Any function that converts the found value to type ``T``. Ex: ``int``.

        Returns:
            ``[astype(v) for v in self[items]]``

        Raises:
            XValueError: Likely exception raised if calling ``astype`` fails
            XTypeError: If the found value is not a (non-``str``) ``Sequence``
            XKeyError: If the key was not found (at any level)
        """
        x = self[items]

        if not isinstance(x, Sequence) or isinstance(x, str):

            raise XTypeError(f"Value {x} is not a list for lookup {items}", actual=str(type(x)))
        return [astype(y) for y in x]

    def get(self, items: str, default: Any = None) -> Any:
        """
        Gets a value from an optional key.
        Also see ``__getitem__``.
        """
        try:
            return self[items]
        except KeyError:
            return default

    def __getitem__(self, items: str) -> Any:
        """
        Gets a value from a required key.
        Analogous to ``dict.__getitem__``, but this can operate on dot-joined strings.

        **NOTE:** The number of keys for which this returns a value can be different from ``len(self)``.


        Example:
            >>> d = NestedDotDict(dict(a=dict(b=1)))
            >>> assert d["a.b"] == 1
        """
        at = self._x

        for item in items.split("."):
            if item not in at:
                raise XKeyError(f"{items} not found: {item} does not exist")
            at = at[item]

        return NestedDotDict(at) if isinstance(at, dict) else copy(at)

    def items(self) -> Sequence[Tup[str, Any]]:

        return list(self._x.items())

    def keys(self) -> Sequence[str]:

        return list(self._x.keys())

    def values(self) -> Sequence[Any]:

        return list(self._x.values())

    def pretty_str(self) -> str:
        """
        Pretty-prints the leaves of this dict using ``json.dumps``.

        Returns:
            A multi-line string
        """
        return json.dumps(
            self.leaves(),
            sort_keys=True,
            indent=4,
            ensure_ascii=False,
        )

    def __len__(self) -> int:
        """
        Returns the number of values in this dict.
        Does **NOT** include nested values.
        """
        return len(self._x)

    def __iter__(self):
        """
        Iterates over values in this dict.
        Does **NOT** include nested items.
        """
        return iter(self._x)

    def __repr__(self):
        return repr(self._x)

    def __str__(self):
        return str(self._x)

    def __eq__(self, other):
        return str(self) == str(other)

    def _to_date(self, s) -> date:
class Foo:
    def some_method(self, x, y):
        return x + y;
        if isinstance(s, date):

            return s
        elif isinstance(s, str):
            # This is MUCH faster than tomlkit's
            return date.fromisoformat(s)
        else:
            raise XTypeError(f"Invalid type {type(s)} for {s}", actual=str(type(s)))

    def _to_datetime(self, s) -> datetime:
class Foo:
    def some_method(self, x, y):
        return x + y;
        if isinstance(s, datetime):

            return s
        elif isinstance(s, str):
            # This is MUCH faster than tomlkit's
            if s.count(":") < 2:
                raise XValueError(
                    f"Datetime {s} does not contain hours, minutes, and seconds", value=s
                )
            return datetime.fromisoformat(s.upper().replace("Z", "+00:00"))
        else:
            raise XTypeError(f"Invalid type {type(s)} for {s}", actual=str(type(s)))


__all__ = ["NestedDotDict"]


1			from __future__ import annotations
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing module docstring Loading history...
2
3			import gzip
			0 ignored issues – show Unused Code introduced 2021-10-28 22:40 UTC by Report Bug Copy Issue Report The import `gzip` seems to be unused. Loading history...
4			import json
5			import pickle
6			from copy import copy
7			from datetime import date, datetime
8			from pathlib import Path, PurePath
9			from typing import Any, ByteString, Callable, Mapping, Optional, Sequence
10			from typing import Tuple as Tup
11			from typing import Type, TypeVar, Union
12
13			import orjson
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Unable to import 'orjson' Loading history...
14			import toml
			0 ignored issues – show introduced 2021-10-05 00:06 UTC by Report Bug Copy Issue Report third party import "import toml" should be placed before "import orjson" Loading history...
15
16			from pocketutils.core import PathLike
			0 ignored issues – show introduced 2021-10-28 22:40 UTC by Report Bug Copy Issue Report Cannot import 'pocketutils.core' due to syntax error 'invalid syntax (<unknown>, line 134)' Loading history... Bug introduced 2021-10-28 22:40 UTC by Report Bug Copy Issue Report The name `core` does not seem to exist in module `pocketutils`. Loading history...
17			from pocketutils.core._internal import read_txt_or_gz, write_txt_or_gz
			0 ignored issues – show Bug introduced 2021-10-28 22:40 UTC by Report Bug Copy Issue Report The name `core` does not seem to exist in module `pocketutils`. Loading history...
18			from pocketutils.core.exceptions import XKeyError, XTypeError, XValueError
			0 ignored issues – show Bug introduced 2021-10-28 22:40 UTC by Report Bug Copy Issue Report The name `core` does not seem to exist in module `pocketutils`. Loading history...
19
20			PICKLE_PROTOCOL = 5
21			T = TypeVar("T")
			0 ignored issues – show Coding Style Naming introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Class name "T" doesn't conform to PascalCase naming style ('[^\\W\\da-z][^\\W_]+$' pattern) This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
22
23
24			def _json_encode_default(obj: Any) -> Any:
			0 ignored issues – show Unused Code introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Either all return statements in a function should return an expression, or none of them should. Loading history...
25			if isinstance(obj, NestedDotDict):
26			# noinspection PyProtectedMember
27			return dict(obj._x)
			0 ignored issues – show Coding Style Best Practice introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report It seems like `_x` was declared protected and should not be accessed from this context. Prefixing a member variable `_` is usually regarded as the equivalent of declaring it with protected visibility that exists in other languages. Consequentially, such a member should only be accessed from the same class or a child class: class MyParent: def __init__(self): self._x = 1; self.y = 2; class MyChild(MyParent): def some_method(self): return self._x # Ok, since accessed from a child class class AnotherClass: def some_method(self, instance_of_my_child): return instance_of_my_child._x # Would be flagged as AnotherClass is not # a child class of MyParent Loading history...
28
29
30			class NestedDotDict(Mapping):

dmyersturnbull / pocketutils

Push — main ( 393c8e...520e83 )

NestedDotDict.write_toml() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like