DLMS-SPODES 0.87.17__py3-none-any.whl → 0.88.1__py3-none-any.whl

DLMS_SPODES/types/useful_types.py

@@ -1,677 +1,627 @@
-from __future__ import annotations
-from functools import lru_cache
-from abc import ABC, abstractmethod
-from typing import Type, Any, Callable
-from dataclasses import dataclass
-from ..types import common_data_types as cdt
-from ..exceptions import DLMSException
-from ..settings import settings
-
-
-class UserfulTypesException(DLMSException):
-    """override DLMSException"""
-
-
-class _String(ABC):
-    LENGTH: int | None
-
-    def __init__(self, value: bytes | bytearray | str | int | tuple | UsefulType = None):
-        match value:
-            case None:                                                        self.__dict__["contents"] = bytes(self.LENGTH)
-            case bytes() if self.LENGTH is None or self.LENGTH <= len(value): self.__dict__["contents"] = value[:self.LENGTH]
-            # case bytes() if self.LENGTH <= len(value):                       self.contents = value[:self.LENGTH]
-            case bytes():            raise ValueError(F'Length of contents for {self.__class__.__name__} must be at least {self.LENGTH}, but got {len(value)}')
-            case tuple():
-                if len(value) == self.LENGTH:
-                    self.__dict__["contents"] = bytes(value)
-                else:
-                    raise ValueError(F"in {self.__class__.__name__} with {value=} got length: {len(value)}, expect {self.LENGTH}")
-            case bytearray():                                                 self.__dict__["contents"] = bytes(value)  # Attention!!! changed method content getting from bytearray
-            case str():                                                       self.__dict__["contents"] = self.from_str(value)
-            case int():                                                       self.__dict__["contents"] = self.from_int(value)
-            case UsefulType():                                                self.__dict__["contents"] = value.contents  # TODO: make right type
-            case _:                                                           raise ValueError(F'Error create {self.__class__.__name__} with value {value}')
-
-    @abstractmethod
-    def __len__(self):
-        """ define in subclasses """
-
-
-class OCTET_STRING(_String):
-    """ An ordered sequence of octets (8 bit bytes) """
-
-    def from_str(self, value: str) -> bytes:
-        """ input as hex code """
-        return bytes.fromhex(value)
-
-    def from_int(self, value: int) -> bytes:
-        """ Convert with recursion. Maximum convert length is 32 """
-        def to_bytes_with(length_):
-            try:
-                return int.to_bytes(value, length_, 'big')
-            except OverflowError:
-                if length_ > 31:
-                    raise ValueError(F'Value {value} is big to convert to bytes')
-                return to_bytes_with(length_+1)
-        length = 1
-        return to_bytes_with(length)
-
-    def __str__(self):
-        return self.contents.hex(' ')
-
-    def __len__(self):
-        return len(self.contents)
-
-    def __getitem__(self, item):
-        return self.contents[item]
-
-    # TODO: maybe remove as redundante?
-    def decode(self) -> bytes:
-        """ decode to build in bytes type """
-        return self.contents
-
-    # TODO: maybe remove as redundante?
-    def to_str(self, encoding: str = 'cp1251') -> str:
-        """ decode to cp1251 by default, replace to '?' if unsupported """
-        temp = list()
-        for i in self.contents:
-            temp.append(i if i > 32 else 63)
-        return bytes(temp).decode(encoding)
-
-
-class CHOICE(ABC):
-    """ TODO: with cdt.CHOICE """
-    ELEMENTS: dict[int, SequenceElement | dict[int, SequenceElement]]
-
-    @property
-    @abstractmethod
-    def TYPE(self) -> Any:
-        """ return valid types """
-
-    def __init_subclass__(cls, **kwargs):
-        if hasattr(cls, 'ELEMENTS'):
-            for el in cls.ELEMENTS.values():
-                if isinstance(el, dict):
-                    """pass, maybe it is for cst.AnyTime"""
-                elif issubclass(el.TYPE, cls.TYPE):
-                    """ type in order """
-                else:
-                    raise ValueError(F'For {cls.__name__} got type {el.TYPE.__name__} with {el.NAME=}, expected {cls.TYPE.__name__}')
-        else:
-            """ subclass with type carry initiate """
-
-    def __getitem__(self, item: int) -> SequenceElement:
-        return self.ELEMENTS[item]
-
-    @property
-    def NAME(self) -> str:
-        return F'CHOICE[{len(self)}]'
-
-    def __len__(self):
-        return len(self.ELEMENTS)
-
-    def is_key(self, value: int) -> bool:
-        return value in self.ELEMENTS.keys()
-
-    @classmethod
-    def parse(cls, value: cdt.Transcript) -> cdt.CommonDataType:
-        """get instance by pattern: <tag>:<value>"""
-        tag, value = value.split(":", maxsplit=1)
-        if (d_t := cls.ELEMENTS.get(int(tag))) is None:
-            raise UserfulTypesException(F"for {cls.__name__} got {cdt.CommonDataType.__name__}: {cdt.TAG(int(tag).to_bytes(1))}; "
-                                        F"expected: {', '.join(map(lambda el: el.NAME, cls.ELEMENTS.values()))}")
-        else:
-            return d_t.TYPE.parse(value)
-
-    def __call__(self,
-                 value: bytes | int = None,
-                 force: bool = False) -> cdt.CommonDataType:
-        """ get instance from encoding or tag(with default value). For CommonDataType only """
-        try:
-            match value:
-                case bytes() as encoding:
-                    match self.ELEMENTS[encoding[0]]:
-                        case SequenceElement() as el: return el.TYPE(encoding)
-                        case dict() as ch:
-                            if encoding[1] in ch.keys():
-                                return ch[encoding[1]].TYPE(encoding)  # use for choice cst.Time | DateTime | Date as OctetString
-                            else:
-                                raise ValueError(F"got type with tag: {encoding[0]} and length: {encoding[1]}, expected length {tuple(ch.keys())}")
-                        case err:                     raise ValueError(F"got {err.__name__}, expected {SequenceElement.__name__} or {dict.__name__}")
-                case int() if force:                  return cdt.get_common_data_type_from(value.to_bytes(1, "big"))()
-                case int() as tag:                    return self.ELEMENTS[tag].TYPE()
-                case None:                            return tuple(self.ELEMENTS.values())[0].TYPE()
-                case error:                           raise ValueError(F'Unknown value type {error}')
-        except KeyError as e:
-            raise UserfulTypesException(F"for {self.__class__.__name__} got {cdt.CommonDataType.__name__}: {cdt.TAG(e.args[0].to_bytes(1))}; expected: {', '.join(map(lambda el: el.NAME, self.ELEMENTS.values()))}")
-
-    def __get_elements(self) -> list[SequenceElement]:
-        """all elements with nested values"""
-        elements = list()
-        for el in self.ELEMENTS.values():
-            match el:
-                case SequenceElement():
-                    elements.append(el)
-                case dict() as dict_el:
-                    elements.extend(dict_el.values())
-                case err:
-                    raise ValueError(F"unknown CHOICE element type {err}")
-        return elements
-
-    def get_types(self) -> tuple[Type[cdt.CommonDataType]]:
-        """ Use in setter attribute.value for validate """
-        return tuple((seq_el.TYPE for seq_el in self.__get_elements()))
-
-    def __str__(self):
-        return F'{CHOICE}: {", ".join((el.NAME for el in self.__get_elements()))}'
-
-
-def get_instance_and_context(meta: Type[UsefulType], value: bytes) -> tuple[UsefulType, bytes]:
-    instance = meta(value)
-    return instance, value[len(instance.contents):]
-
-
-@dataclass(frozen=True)
-class SequenceElement:
-    NAME: str
-    TYPE: Type[UsefulType | SEQUENCE | CHOICE | cdt.CommonDataType]
-
-    def __str__(self):
-        return F'{self.NAME}: {self.TYPE.__name__}'
-
-
-class SEQUENCE(ABC):
-    """ TODO: """
-    ELEMENTS: tuple[SequenceElement | SEQUENCE, ...]
-    values: list[UsefulType]
-
-    def __init__(self, value: bytes | tuple | list | None | SEQUENCE = None):
-        self.__dict__['values'] = [None] * len(self.ELEMENTS)
-        match value:
-            case tuple() | list():
-                if len(value) != len(self):
-                    raise ValueError(F'Struct {self.__class__.__name__} got length:{len(value)}, expected length:{len(self)}')
-                self.from_tuple(value)
-            case bytes():           self.from_bytes(value)
-            case None:              self.from_default()
-            case self.__class__():  self.from_bytes(value.contents)
-            case _:                 raise TypeError(F'Value: "{value}" not supported')
-
-    def from_default(self):
-        for i in range(len(self)):
-            self.values[i] = self.ELEMENTS[i].TYPE()
-
-    def from_bytes(self, value: bytes | bytearray):
-        for i, element in enumerate(self.ELEMENTS):
-            self.values[i], value = get_instance_and_context(element.TYPE, value)
-
-    def from_tuple(self, value: tuple | list):
-        for i, val in enumerate(value):
-            self.values[i] = self.ELEMENTS[i].TYPE(val)
-
-    def __str__(self):
-        return F'{{{", ".join(map(lambda val: F"{val[0].NAME}: {val[1]}", zip(self.ELEMENTS, self.values)))}}}'
-
-    def __repr__(self):
-        return F'{self.__class__.__name__}(({(", ".join(map(str, self.values)))}))'
-
-    def __get_index(self, name: str) -> int | None:
-        """ get index by name. Return None if not found """
-        for i, element in enumerate(self.ELEMENTS):
-            if element.NAME == name:
-                return i
-        else:
-            return None
-
-    def __setattr__(self, key, value: UsefulType):
-        match self.__get_index(key):
-            case int() as i if isinstance(value, self.ELEMENTS[i]):                self.values[i] = value
-            case int() as i:                raise ValueError(F'Try assign {key} Type got {value.__class__.__name__}, expected {self.ELEMENTS[i].NAME}')
-            case _:                raise ValueError(F'Unsupported change: {key}')
-
-    def __getattr__(self, item: str) -> UsefulType:
-        match self.__get_index(item):
-            case int() as i: return self.values[i]
-            case _:          return self.__dict__[item]
-
-    def __getitem__(self, item: str | int) -> UsefulType:
-        """ get element by index or name """
-        match item:
-            case str() as name:  return self.values[self.__get_index(name)]
-            case int() as index: return self.values[index]
-            case _:              raise ValueError(F'Unsupported type {item.__class__}')
-
-    def __setitem__(self, key: int, value: UsefulType):
-        """ set data to element by index """
-        if isinstance(value, self.ELEMENTS[key].TYPE):
-            self.values[key] = value
-        else:
-            raise ValueError(F'Type got {value.__class__.__name__}, expected {self.ELEMENTS[key].TYPE}')
-
-    def __len__(self):
-        return len(self.ELEMENTS)
-
-    @property
-    def contents(self) -> bytes:
-        return b''.join(el.contents for el in self.values)
-
-    @property
-    def NAME(self) -> str:
-        return F'{self.__class__.__name__}[{len(self)}]'
-
-
-class UsefulType(ABC):
-    """"""
-    contents: bytes
-    __match_args__ = ('contents',)
-    cb_post_set: Callable
-    cb_preset: Callable
-
-    @abstractmethod
-    def __init__(self, value):
-        """ constructor """
-
-    def __eq__(self, other: UsefulType):
-        match other:
-            case self.__class__(self.contents): return True
-            case _:                             return False
-
-    def set_contents_from(self, value: UsefulType | bytes | bytearray | str | int | bool | None):
-        new_value = self.__class__(value)
-        if hasattr(self, 'cb_preset'):
-            self.cb_preset(new_value)
-        self.__dict__['contents'] = new_value.contents
-        if hasattr(self, 'cb_post_set'):
-            self.cb_post_set()
-
-    def __setattr__(self, key, value):
-        match key:
-            case 'contents' as prop if hasattr(self, 'contents'): raise ValueError(F"Don't support set {prop}")
-            case _: super().__setattr__(key, value)
-
-
-class DigitalMixin(ABC):
-    """ Default value is 0 """
-    SIGNED: bool
-    LENGTH: int
-    contents: bytes
-
-    def __init__(self, value: bytes | bytearray | str | int | DigitalMixin = None):
-        match value:
-            case bytes() if self.LENGTH <= len(value): self.__dict__["contents"] = value[:self.LENGTH]
-            case bytes():            raise ValueError(F'Length of contents for {self.__class__.__name__} must be at least {self.LENGTH}, but got {len(value)}')
-            case bytearray():                          self.__dict__["contents"] = bytes(value)  # Attention!!! changed method content getting from bytearray
-            case str('-') if self.SIGNED:              self.__dict__["contents"] = bytes(self.LENGTH)
-            case int():                                self.__dict__["contents"] = self.from_int(value)
-            case str():                                self.__dict__["contents"] = self.from_str(value)
-            case None:                                 self.__dict__["contents"] = bytes(self.LENGTH)
-            case self.__class__():                     self.__dict__["contents"] = value.contents
-            case _:                                    raise ValueError(F'Error create {self.__class__.__name__} with value: {value}')
-
-    def from_int(self, value: int | float) -> bytes:
-        try:
-            return int(value).to_bytes(self.LENGTH, 'big', signed=self.SIGNED)
-        except OverflowError:
-            raise ValueError(F'value {value} out of range')
-
-    def from_str(self, value: str) -> bytes:
-        return self.from_int(float(value))
-
-    def __int__(self):
-        """ return the build in integer type """
-        return int.from_bytes(self.contents, 'big', signed=self.SIGNED)
-
-    def __str__(self):
-        return str(int(self))
-
-    def __repr__(self):
-        return F'{self.__class__.__name__}({self})'
-
-    def __gt__(self, other: DigitalMixin):
-        match other:
-            case DigitalMixin(): return int(self) > int(other)
-            case _:          raise TypeError(F'Compare type is {other.__class__}, expected Digital')
-
-    def __len__(self) -> int:
-        return self.LENGTH
-
-    def __hash__(self):
-        return int(self)
-
-    def validate_from(self, value: str, cursor_position: int) -> tuple[str, int]:
-        """ return validated value and cursor position. Use in Entry. TODO: remove it, make better """
-        type(self)(value=value)
-        return value, cursor_position
-
-
-class Integer8(DigitalMixin, UsefulType):
-    """ INTEGER(-127…128) """
-    LENGTH = 1
-    SIGNED = True
-
-
-class Integer16(DigitalMixin, UsefulType):
-    """ INTEGER(-32 768...32 767) """
-    LENGTH = 2
-    SIGNED = True
-
-
-class Integer32(DigitalMixin, UsefulType):
-    """ INTEGER(-2 147 483 648...2 147 483 647) """
-    LENGTH = 4
-    SIGNED = True
-
-
-class Integer64(DigitalMixin, UsefulType):
-    """ INTEGER(-2^63...2^63-1) """
-    LENGTH = 8
-    SIGNED = True
-
-
-class Unsigned8(DigitalMixin, UsefulType):
-    """ INTEGER(0...255) """
-    LENGTH = 1
-    SIGNED = False
-    # __match_args__ = ('contents',)
-
-
-class Unsigned16(DigitalMixin, UsefulType):
-    """ INTEGER(0...65 535) """
-    LENGTH = 2
-    SIGNED = False
-
-
-class Unsigned32(DigitalMixin, UsefulType):
-    """ INTEGER(0...4 294 967 295) """
-    LENGTH = 4
-    SIGNED = False
-
-
-class Unsigned64(DigitalMixin, UsefulType):
-    """ INTEGER(0...264-1) """
-    LENGTH = 8
-    SIGNED = False
-
-
-class CosemClassId(Unsigned16):
-    """ Identification code of the IC (range 0 to 65 535). The class_id of each object is retrieved together with the logical name by reading the object_list attribute of an
-    “Association LN” / ”Association SN” object.
-        - class_id-s from 0 to 8 191 are reserved to be specified by the DLMS UA.
-        - class_id-s from 8 192 to 32 767 are reserved for manufacturer specific ICs.
-        - class_id-s from 32 768 to 65 535 are reserved for user group specific ICs.
-    The DLMS UA reserves the right to assign ranges to individual manufacturers or user groups. """
-
-    def __str__(self) -> str:
-        if res := settings.class_name.get(int(self)):
-            return res
-        else:
-            return repr(self)
-
-    def __repr__(self):
-        return F"{self.__class__.__name__}({int(self)})"
-
-
-class CosemObjectInstanceId(OCTET_STRING, UsefulType):
-    LENGTH = 6
-
-    def __str__(self):
-        return F"\"{'.'.join(map(str, self.contents))}\""
-
-    def from_str(self, value: str) -> bytes:
-        """ create logical_name: octet_string from string type ddd.ddd.ddd.ddd.ddd.ddd, ex.: 0.0.1.0.0.255 """
-        raw_value = bytes()
-        for typecast, separator in zip((self.__from_group_A, )*5+(self.__from_group_F, ), ('.', '.', '.', '.', '.', ' ')):
-            try:
-                element, value = value.split(separator, 1)
-            except ValueError:
-                element, value = value, ''
-            raw_value += typecast(element)
-        return raw_value
-
-    def __from_group_A(self, value: str) -> bytes:
-        if isinstance(value, str):
-            if value == '':
-                return b'\x00'
-            try:
-                return int(value).to_bytes(1, 'big')
-            except OverflowError:
-                raise ValueError(F'Int too big to convert {value}')
-        else:
-            raise TypeError(F'Unsupported type validation from string, got {value.__class__}')
-
-    def __from_group_F(self, value: str) -> bytes:
-        if isinstance(value, str):
-            if value == '':
-                return b'\xff'
-            try:
-                return int(value).to_bytes(1, 'big')
-            except OverflowError:
-                raise ValueError(F'Int too big to convert {value}')
-        else:
-            raise TypeError(F'Unsupported type validation from string, got {value.__class__}')
-
-
-class CosemObjectAttributeId(Integer8):
-    """ TODO """
-
-    @lru_cache(14)  # for test
-    def __new__(cls, *args, **kwargs):
-        return super().__new__(cls)
-
-
-class CosemObjectMethodId(Integer8):
-    """ TODO """
-
-
-class AccessSelectionParameters(Unsigned8):
-    """ Unsigned8(0..1) """
-
-    def __init__(self, value: int | str | Unsigned8 = 1):
-        super(AccessSelectionParameters, self).__init__(value)
-        if int(self) > 1 or int(self) < 0:
-            raise ValueError(F'The {self.__class__.__name__} got {self}, expected 0..1')
-
-
-class CosemAttributeDescriptor(SEQUENCE):
-    class_id: CosemClassId
-    instance_id: CosemObjectInstanceId
-    attribute_id: CosemObjectAttributeId
-    access_selection_parameters: AccessSelectionParameters
-    ELEMENTS = (SequenceElement('class_id', CosemClassId),
-                SequenceElement('instance_id', CosemObjectInstanceId),
-                SequenceElement('attribute_id', CosemObjectAttributeId))
-
-    def __init__(self, value: bytes | tuple | list = None):
-        super(CosemAttributeDescriptor, self).__init__(value)
-        self.__dict__['access_selection_parameters'] = AccessSelectionParameters(0)
-
-    @property
-    def contents(self) -> bytes:
-        """ Always contain Access_Selection_Parameters. DLMS UA 1000-2 Ed.9 Excerpt 9.3.9.1.3 """
-        return super(CosemAttributeDescriptor, self).contents + self.access_selection_parameters.contents
-
-
-class CosemMethodDescriptor(SEQUENCE):
-    class_id: CosemClassId
-    instance_id: CosemObjectInstanceId
-    method_id: CosemObjectMethodId
-    ELEMENTS = (SequenceElement('class_id', CosemClassId),
-                SequenceElement('instance_id', CosemObjectInstanceId),
-                SequenceElement('method_id', CosemObjectMethodId))
-
-    def __init__(self, value: tuple[CosemClassId, CosemObjectInstanceId, CosemObjectMethodId]):
-        super(CosemMethodDescriptor, self).__init__(value)
-
-
-class Data(CHOICE, ABC):
-    TYPE = cdt.CommonDataType
-
-
-class SelectiveAccessDescriptor(SEQUENCE, ABC):
-    """ Selective access specification always starts with an access selector, followed by an access-specific access parameter list.
-    Specified IS/IEC 62056-53 : 2006, 7.4.1.6 Selective access """
-    access_selector: Unsigned8
-    access_parameters: cdt.CommonDataType
-    ELEMENTS: tuple[SequenceElement, SequenceElement]
-
-    def __init__(self, value: tuple | bytes | None = None):
-        super(SelectiveAccessDescriptor, self).__init__(value)
-        self.access_selector.cb_post_set = self.__validate_selector
-
-    @property
-    @abstractmethod
-    def ELEMENTS(self) -> tuple[SequenceElement, SequenceElement]:
-        """ return elements """
-
-    def from_default(self):
-        self.values[0] = self.ELEMENTS[0].TYPE()
-        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE()
-
-    def from_bytes(self, value: bytes | bytearray):
-        self.values[0], value = get_instance_and_context(self.ELEMENTS[0].TYPE, value)
-        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE(value)
-
-    def from_tuple(self, value: tuple | list):
-        self.values[0] = self.ELEMENTS[0].TYPE(value[0])
-        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE(value[1])
-
-    @property
-    def contents(self) -> bytes:
-        return self.access_selector.contents+self.access_parameters.encoding
-
-    def __validate_selector(self):
-        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE()
-        print('change sel')
-
-    def set_selector(self, index: int, value=None):
-        """ set selector value from int type """
-        self.access_selector.set_contents_from(index)
-        if value:
-            self.access_parameters.set(value)
-
-
-class CosemAttributeDescriptorWithSelection(SEQUENCE):
-    cosem_attribute_descriptor: CosemAttributeDescriptor
-    access_selection: SelectiveAccessDescriptor
-    ELEMENTS: tuple[CosemAttributeDescriptor, SequenceElement]
-
-    def __init__(self, value: bytes | tuple | list = None):
-        super(CosemAttributeDescriptorWithSelection, self).__init__(value)
-        self.cosem_attribute_descriptor.access_selection_parameters.set_contents_from(1)
-
-    @property
-    @abstractmethod
-    def ELEMENTS(self) -> tuple[SequenceElement, SequenceElement]:
-        """ return elements. Need initiate in subclass """
-
-
-class InvokeIdAndPriority(Unsigned8):
-
-    def __init__(self, value: bytes | bytearray | str | int | DigitalMixin = 0b1100_0000):
-        super(InvokeIdAndPriority, self).__init__(value)
-        if self.contents[0] & 0b00110000:
-            raise ValueError(F'For {self.__class__.__name__} set reserved bits')
-
-    @classmethod
-    def from_parameters(cls, invoke_id: int = 0,
-                        service_class: int = 0,
-                        priority: int = 0):
-        instance = cls()
-        instance.invoke_id = invoke_id
-        instance.service_class = service_class
-        instance.priority = priority
-        return instance
-
-    @property
-    def invoke_id(self) -> int:
-        return self.contents[0] & 0b0000_1111
-
-    @invoke_id.setter
-    def invoke_id(self, value: int):
-        if value & 0b00001111:
-            self.__dict__['contents'] = int((self.contents[0] & 0b1111_0000) | value).to_bytes(1, 'big')
-        else:
-            raise ValueError(F'Got {value} for invoke-id, expected 0..15')
-
-    @property
-    def service_class(self) -> int:
-        """ 0: Unconfirmed or 1: Confirmed bit return """
-        return (self.contents[0] >> 6) & 0b1
-
-    @service_class.setter
-    def service_class(self, value: int):
-        match value:
-            case 0 | 1: self.__dict__['contents'] = int((self.contents[0] & 0b1011_1111) | (value << 6)).to_bytes(1, 'big')
-            case _:     raise ValueError(F'Got {value} for service_class, expected 0..1')
-
-    @property
-    def priority(self) -> int:
-        """ 0: Normal or 1: High bit return """
-        return (self.contents[0] >> 7) & 0b1
-
-    @priority.setter
-    def priority(self, value: int):
-        match value:
-            case 0 | 1: self.__dict__['contents'] = int((self.contents[0] & 0b0111_1111) | (value << 7)).to_bytes(1, 'big')
-            case _:     raise ValueError(F'Got {value} for service_class, expected 0..1')
-
-    def __str__(self):
-        return F'priority: {"High" if self.contents[0] & 0b1000_0000 else "Normal"}, ' \
-               F'service-class: {"Confirmed" if self.contents[0] & 0b0100_0000 else "Unconfirmed"}, ' \
-               F'invoke-id: {self.invoke_id},'
-
-
-if __name__ == '__main__':
-    a = CosemObjectAttributeId(1)
-    b = CosemObjectAttributeId(2)
-    print(a > b)
-
-    a = CosemClassId(cdt.LongUnsigned(1).contents)
-    a = InvokeIdAndPriority()
-    c = a.invoke_id
-    d = a.service_class
-    f = a.priority
-    a.service_class = 1
-    a.invoke_id = 14
-    e = InvokeIdAndPriority.from_parameters(1, 1, 1)
-    class AccessSelector(Unsigned8):
-        """ Unsigned8 1..4 """
-        def __init__(self, value: int | str | Unsigned8 = 1):
-            super(AccessSelector, self).__init__(value)
-            if int(self) > 4 or int(self) < 1:
-                raise ValueError(F'The {self.__class__.__name__} got {self}, expected 1..4')
-
-
-    class MyData(Data):
-        ELEMENTS = {1: SequenceElement('0 a', cdt.NullData),
-                    2: SequenceElement('1 d', cdt.Integer),
-                    3: SequenceElement('second', cdt.ScalUnitType),
-                    4: SequenceElement('3', cdt.Integer)}
-
-    class My(SelectiveAccessDescriptor):
-        access_selector: AccessSelector
-        access_parameters: MyData
-        ELEMENTS = (SequenceElement('access_selector', AccessSelector),
-                    SequenceElement('access_parameters', MyData))
-
-    ba = My()
-    b = My((3, (10, 10)))
-    b2 = b.access_selector
-    b3 = b.access_parameters
-    b4 = b.access_parameters.unit
-    b_repr = My(b'\x03\x0f"')
-    b.set_selector(3, 34)
-    a = CosemAttributeDescriptor((1, '1.1.1.1.1.1', 1))
-    a_repr = CosemAttributeDescriptor(b'\x00\x01\x01\x01\x01\x01\x01\x01\x01\x00')
-
-    class MyWith(CosemAttributeDescriptorWithSelection):
-        access_selection: My
-        ELEMENTS = (SequenceElement('cosem_attribute_descriptor', CosemAttributeDescriptor),
-                    SequenceElement('access_selection', My))
-
-    c = MyWith()
-    c_from = MyWith((a, b))
-    c_repr = MyWith(b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x01\x00')
-    print(a)
-    c = CosemAttributeDescriptor(b'\x00\x01\x01\x01\x01\x01\x01\x01\x01\x00')
-    print(c)
+from typing_extensions import deprecated
+from functools import lru_cache
+from abc import ABC, abstractmethod
+from typing import Type, Any, Callable, Protocol, runtime_checkable
+from dataclasses import dataclass
+from StructResult import result
+from ..types import common_data_types as cdt
+from ..types.type_alias import Encoding
+from ..exceptions import DLMSException
+from ..settings import settings
+
+
+class UserfulTypesException(DLMSException):
+    """override DLMSException"""
+
+
+@runtime_checkable
+class UsefulType(Protocol):
+    """"""
+    contents: bytes
+    __match_args__ = ('contents',)
+    cb_post_set: Callable
+    cb_preset: Callable
+
+    def __init__(self, value):
+        """ constructor """
+
+    def __eq__(self, other: "UsefulType"):
+        match other:
+            case self.__class__(self.contents): return True
+            case _:                             return False
+
+    def set_contents_from(self, value: bytes | bytearray | str | int | bool | None):
+        new_value = self.__class__(value)
+        if hasattr(self, 'cb_preset'):
+            self.cb_preset(new_value)
+        self.__dict__['contents'] = new_value.contents
+        if hasattr(self, 'cb_post_set'):
+            self.cb_post_set()
+
+    def __setattr__(self, key, value):
+        match key:
+            case 'contents' as prop if hasattr(self, 'contents'): raise ValueError(F"Don't support set {prop}")
+            case _: super().__setattr__(key, value)
+
+
+class _String(Protocol):
+    LENGTH: int | None
+
+    def __init__(self, value: bytes | bytearray | str | int | tuple | UsefulType = None):
+        match value:
+            case None:                                                        self.__dict__["contents"] = bytes(self.LENGTH)
+            case bytes() if self.LENGTH is None or self.LENGTH <= len(value): self.__dict__["contents"] = value[:self.LENGTH]
+            # case bytes() if self.LENGTH <= len(value):                       self.contents = value[:self.LENGTH]
+            case bytes():            raise ValueError(F'Length of contents for {self.__class__.__name__} must be at least {self.LENGTH}, but got {len(value)}')
+            case tuple():
+                if len(value) == self.LENGTH:
+                    self.__dict__["contents"] = bytes(value)
+                else:
+                    raise ValueError(F"in {self.__class__.__name__} with {value=} got length: {len(value)}, expect {self.LENGTH}")
+            case bytearray():                                                 self.__dict__["contents"] = bytes(value)  # Attention!!! changed method content getting from bytearray
+            case str():                                                       self.__dict__["contents"] = self.from_str(value)
+            case int():                                                       self.__dict__["contents"] = self.from_int(value)
+            case UsefulType():                                                self.__dict__["contents"] = value.contents  # TODO: make right type
+            case _:                                                           raise ValueError(F'Error create {self.__class__.__name__} with value {value}')
+
+    def __len__(self):
+        """ define in subclasses """
+
+
+class OCTET_STRING(_String):
+    """ An ordered sequence of octets (8 bit bytes) """
+
+    def from_str(self, value: str) -> bytes:
+        """ input as hex code """
+        return bytes.fromhex(value)
+
+    def from_int(self, value: int) -> bytes:
+        """ Convert with recursion. Maximum convert length is 32 """
+        def to_bytes_with(length_):
+            try:
+                return int.to_bytes(value, length_, 'big')
+            except OverflowError:
+                if length_ > 31:
+                    raise ValueError(F'Value {value} is big to convert to bytes')
+                return to_bytes_with(length_+1)
+        length = 1
+        return to_bytes_with(length)
+
+    def __str__(self):
+        return self.contents.hex(' ')
+
+    def __len__(self):
+        return len(self.contents)
+
+    def __getitem__(self, item):
+        return self.contents[item]
+
+    # TODO: maybe remove as redundante?
+    def decode(self) -> bytes:
+        """ decode to build in bytes type """
+        return self.contents
+
+    # TODO: maybe remove as redundante?
+    def to_str(self, encoding: str = 'cp1251') -> str:
+        """ decode to cp1251 by default, replace to '?' if unsupported """
+        temp = list()
+        for i in self.contents:
+            temp.append(i if i > 32 else 63)
+        return bytes(temp).decode(encoding)
+
+
+@runtime_checkable
+class CHOICE(Protocol):
+    """ TODO: with cdt.CHOICE """
+    ELEMENTS: "dict[int, SequenceElement | dict[int, SequenceElement]]"
+
+    @property
+    def TYPE(self) -> Any:
+        """ return valid types """
+
+    # def __init_subclass__(cls, **kwargs):
+    #     if hasattr(cls, 'ELEMENTS'):
+    #         for el in cls.ELEMENTS.values():
+    #             if isinstance(el, dict):
+    #                 """pass, maybe it is for cst.AnyTime"""
+    #             elif issubclass(el.TYPE, cls.TYPE):
+    #                 """ type in order """
+    #             else:
+    #                 raise ValueError(F'For {cls.__name__} got type {el.TYPE.__name__} with {el.NAME=}, expected {cls.TYPE.__name__}')
+    #     else:
+    #         """ subclass with type carry initiate """
+
+    def __getitem__(self, item: int) -> "SequenceElement":
+        return self.ELEMENTS[item]
+
+    @property
+    def NAME(self) -> str:
+        return F'CHOICE[{len(self)}]'
+
+    def __len__(self):
+        return len(self.ELEMENTS)
+
+    def is_key(self, value: int) -> bool:
+        return value in self.ELEMENTS.keys()
+
+    @classmethod
+    def parse(cls, value: cdt.Transcript) -> cdt.CommonDataType:
+        """get instance by pattern: <tag>:<value>"""
+        tag, value = value.split(":", maxsplit=1)
+        if (d_t := cls.ELEMENTS.get(int(tag))) is None:
+            raise UserfulTypesException(F"for {cls.__name__} got {cdt.CommonDataType.__name__}: {cdt.TAG(int(tag).to_bytes(1))}; "
+                                        F"expected: {', '.join(map(lambda el: el.NAME, cls.ELEMENTS.values()))}")
+        else:
+            return d_t.TYPE.parse(value)
+
+    @classmethod
+    def from_encoding(cls, encoding: Encoding) -> result.SimpleOrError[cdt.CommonDataType]:
+        match cls.ELEMENTS[encoding[0]]:
+            case SequenceElement() as el:
+                return el.TYPE.from_encoding(encoding)
+            case dict() as ch:
+                if encoding[1] in ch.keys():
+                    return ch[encoding[1]].TYPE.from_encoding(encoding)  # use for choice cst.Time | DateTime | Date as OctetString
+                else:
+                    return result.Error.from_e(ValueError(F"got type with tag: {encoding[0]} and length: {encoding[1]}, expected length {tuple(ch.keys())}"))
+            case err:
+                raise RuntimeError(F"got {err.__name__}, expected {SequenceElement.__name__} or {dict.__name__}")
+
+    @deprecated("use <from_encoding> and more")
+    def __call__(self,
+                 value: bytes | int = None,
+                 force: bool = False) -> cdt.CommonDataType:
+        """ get instance from encoding or tag(with default value). For CommonDataType only """
+        try:
+            match value:
+                case bytes() as encoding:             return self.__class__.from_encoding(encoding).unwrap()
+                case int() if force:                  return cdt.get_common_data_type_from(value.to_bytes(1, "big"))()
+                case int() as tag:                    return self.ELEMENTS[tag].TYPE()
+                case None:                            return tuple(self.ELEMENTS.values())[0].TYPE()
+                case error:                           raise ValueError(F'Unknown value type {error}')
+        except KeyError as e:
+            raise UserfulTypesException(F"for {self.__class__.__name__} got {cdt.CommonDataType.__name__}: {cdt.TAG(e.args[0].to_bytes(1))}; expected: {', '.join(map(lambda el: el.NAME, self.ELEMENTS.values()))}")
+
+    def __get_elements(self) -> "list[SequenceElement]":
+        """all elements with nested values"""
+        elements = []
+        for el in self.ELEMENTS.values():
+            match el:
+                case SequenceElement():
+                    elements.append(el)
+                case dict() as dict_el:
+                    elements.extend(dict_el.values())
+                case err:
+                    raise ValueError(F"unknown CHOICE element type {err}")
+        return elements
+
+    def get_types(self) -> tuple[Type[cdt.CommonDataType]]:
+        """ Use in setter attribute.value for validate """
+        return tuple((seq_el.TYPE for seq_el in self.__get_elements()))
+
+    def __str__(self):
+        return F'{CHOICE}: {", ".join((el.NAME for el in self.__get_elements()))}'
+
+
+def get_instance_and_context(meta: type[UsefulType], value: bytes) -> tuple[UsefulType, bytes]:
+    instance = meta(value)
+    return instance, value[len(instance.contents):]
+
+
+class SEQUENCE(Protocol):
+    """ TODO: """
+    ELEMENTS: "tuple[SequenceElement | SEQUENCE, ...]"
+    values: list[UsefulType]
+
+    def __init__(self, value: "bytes | tuple | list | None | SEQUENCE" = None):
+        self.__dict__['values'] = [None] * len(self.ELEMENTS)
+        match value:
+            case tuple() | list():
+                if len(value) != len(self):
+                    raise ValueError(F'Struct {self.__class__.__name__} got length:{len(value)}, expected length:{len(self)}')
+                self.from_tuple(value)
+            case bytes():           self.from_bytes(value)
+            case None:              self.from_default()
+            case self.__class__():  self.from_bytes(value.contents)
+            case _:                 raise TypeError(F'Value: "{value}" not supported')
+
+    def from_default(self):
+        for i in range(len(self)):
+            self.values[i] = self.ELEMENTS[i].TYPE()
+
+    def from_bytes(self, value: bytes | bytearray):
+        for i, element in enumerate(self.ELEMENTS):
+            self.values[i], value = get_instance_and_context(element.TYPE, value)
+
+    def from_tuple(self, value: tuple | list):
+        for i, val in enumerate(value):
+            self.values[i] = self.ELEMENTS[i].TYPE(val)
+
+    def __str__(self):
+        return F'{{{", ".join(map(lambda val: F"{val[0].NAME}: {val[1]}", zip(self.ELEMENTS, self.values)))}}}'
+
+    def __repr__(self):
+        return F'{self.__class__.__name__}(({(", ".join(map(str, self.values)))}))'
+
+    def __get_index(self, name: str) -> int | None:
+        """ get index by name. Return None if not found """
+        for i, element in enumerate(self.ELEMENTS):
+            if element.NAME == name:
+                return i
+        else:
+            return None
+
+    def __setattr__(self, key, value: UsefulType):
+        match self.__get_index(key):
+            case int() as i if isinstance(value, self.ELEMENTS[i]):                self.values[i] = value
+            case int() as i:                raise ValueError(F'Try assign {key} Type got {value.__class__.__name__}, expected {self.ELEMENTS[i].NAME}')
+            case _:                raise ValueError(F'Unsupported change: {key}')
+
+    def __getattr__(self, item: str) -> UsefulType:
+        match self.__get_index(item):
+            case int() as i: return self.values[i]
+            case _:          return self.__dict__[item]
+
+    def __getitem__(self, item: str | int) -> UsefulType:
+        """ get element by index or name """
+        match item:
+            case str() as name:  return self.values[self.__get_index(name)]
+            case int() as index: return self.values[index]
+            case _:              raise ValueError(F'Unsupported type {item.__class__}')
+
+    def __setitem__(self, key: int, value: UsefulType):
+        """ set data to element by index """
+        if isinstance(value, self.ELEMENTS[key].TYPE):
+            self.values[key] = value
+        else:
+            raise ValueError(F'Type got {value.__class__.__name__}, expected {self.ELEMENTS[key].TYPE}')
+
+    def __len__(self):
+        return len(self.ELEMENTS)
+
+    @property
+    def contents(self) -> bytes:
+        return b''.join(el.contents for el in self.values)
+
+    @property
+    def NAME(self) -> str:
+        return F'{self.__class__.__name__}[{len(self)}]'
+
+
+@dataclass(frozen=True)
+class SequenceElement:
+    NAME: str
+    TYPE: Type[UsefulType | SEQUENCE | CHOICE | cdt.CommonDataType]
+
+    def __str__(self):
+        return F'{self.NAME}: {self.TYPE.__name__}'
+
+
+
+class DigitalMixin:
+    """ Default value is 0 """
+    SIGNED: bool
+    LENGTH: int
+    contents: bytes
+
+    def __init__(self, value: "bytes | bytearray | str | int | DigitalMixin" = None):
+        match value:
+            case bytes() if self.LENGTH <= len(value): self.__dict__["contents"] = value[:self.LENGTH]
+            case bytes():            raise ValueError(F'Length of contents for {self.__class__.__name__} must be at least {self.LENGTH}, but got {len(value)}')
+            case bytearray():                          self.__dict__["contents"] = bytes(value)  # Attention!!! changed method content getting from bytearray
+            case str('-') if self.SIGNED:              self.__dict__["contents"] = bytes(self.LENGTH)
+            case int():                                self.__dict__["contents"] = self.from_int(value)
+            case str():                                self.__dict__["contents"] = self.from_str(value)
+            case None:                                 self.__dict__["contents"] = bytes(self.LENGTH)
+            case self.__class__():                     self.__dict__["contents"] = value.contents
+            case _:                                    raise ValueError(F'Error create {self.__class__.__name__} with value: {value}')
+
+    def from_int(self, value: int | float) -> bytes:
+        try:
+            return int(value).to_bytes(self.LENGTH, 'big', signed=self.SIGNED)
+        except OverflowError:
+            raise ValueError(F'value {value} out of range')
+
+    def from_str(self, value: str) -> bytes:
+        return self.from_int(float(value))
+
+    def __int__(self):
+        """ return the build in integer type """
+        return int.from_bytes(self.contents, 'big', signed=self.SIGNED)
+
+    def __str__(self):
+        return str(int(self))
+
+    def __repr__(self):
+        return F'{self.__class__.__name__}({self})'
+
+    def __gt__(self, other: "DigitalMixin"):
+        match other:
+            case DigitalMixin(): return int(self) > int(other)
+            case _:          raise TypeError(F'Compare type is {other.__class__}, expected Digital')
+
+    def __len__(self) -> int:
+        return self.LENGTH
+
+    def __hash__(self):
+        return int(self)
+
+    def validate_from(self, value: str, cursor_position: int) -> tuple[str, int]:
+        """ return validated value and cursor position. Use in Entry. TODO: remove it, make better """
+        type(self)(value=value)
+        return value, cursor_position
+
+
+class Integer8(DigitalMixin, UsefulType):
+    """ INTEGER(-127…128) """
+    LENGTH = 1
+    SIGNED = True
+
+
+class Integer16(DigitalMixin, UsefulType):
+    """ INTEGER(-32 768...32 767) """
+    LENGTH = 2
+    SIGNED = True
+
+
+class Integer32(DigitalMixin, UsefulType):
+    """ INTEGER(-2 147 483 648...2 147 483 647) """
+    LENGTH = 4
+    SIGNED = True
+
+
+class Integer64(DigitalMixin, UsefulType):
+    """ INTEGER(-2^63...2^63-1) """
+    LENGTH = 8
+    SIGNED = True
+
+
+class Unsigned8(DigitalMixin, UsefulType):
+    """ INTEGER(0...255) """
+    LENGTH = 1
+    SIGNED = False
+    # __match_args__ = ('contents',)
+
+
+class Unsigned16(DigitalMixin, UsefulType):
+    """ INTEGER(0...65 535) """
+    LENGTH = 2
+    SIGNED = False
+
+
+class Unsigned32(DigitalMixin, UsefulType):
+    """ INTEGER(0...4 294 967 295) """
+    LENGTH = 4
+    SIGNED = False
+
+
+class Unsigned64(DigitalMixin, UsefulType):
+    """ INTEGER(0...264-1) """
+    LENGTH = 8
+    SIGNED = False
+
+
+class CosemClassId(Unsigned16):
+    """ Identification code of the IC (range 0 to 65 535). The class_id of each object is retrieved together with the logical name by reading the object_list attribute of an
+    “Association LN” / ”Association SN” object.
+        - class_id-s from 0 to 8 191 are reserved to be specified by the DLMS UA.
+        - class_id-s from 8 192 to 32 767 are reserved for manufacturer specific ICs.
+        - class_id-s from 32 768 to 65 535 are reserved for user group specific ICs.
+    The DLMS UA reserves the right to assign ranges to individual manufacturers or user groups. """
+
+    def __str__(self) -> str:
+        if res := settings.class_name.get(int(self)):
+            return res
+        else:
+            return repr(self)
+
+    def __repr__(self):
+        return F"{self.__class__.__name__}({int(self)})"
+
+
+class CosemObjectInstanceId(OCTET_STRING, UsefulType):
+    LENGTH = 6
+
+    def __str__(self):
+        return F"\"{'.'.join(map(str, self.contents))}\""
+
+    def from_str(self, value: str) -> bytes:
+        """ create logical_name: octet_string from string type ddd.ddd.ddd.ddd.ddd.ddd, ex.: 0.0.1.0.0.255 """
+        raw_value = bytes()
+        for typecast, separator in zip((self.__from_group_A, )*5+(self.__from_group_F, ), ('.', '.', '.', '.', '.', ' ')):
+            try:
+                element, value = value.split(separator, 1)
+            except ValueError:
+                element, value = value, ''
+            raw_value += typecast(element)
+        return raw_value
+
+    def __from_group_A(self, value: str) -> bytes:
+        if isinstance(value, str):
+            if value == '':
+                return b'\x00'
+            try:
+                return int(value).to_bytes(1, 'big')
+            except OverflowError:
+                raise ValueError(F'Int too big to convert {value}')
+        else:
+            raise TypeError(F'Unsupported type validation from string, got {value.__class__}')
+
+    def __from_group_F(self, value: str) -> bytes:
+        if isinstance(value, str):
+            if value == '':
+                return b'\xff'
+            try:
+                return int(value).to_bytes(1, 'big')
+            except OverflowError:
+                raise ValueError(F'Int too big to convert {value}')
+        else:
+            raise TypeError(F'Unsupported type validation from string, got {value.__class__}')
+
+
+class CosemObjectAttributeId(Integer8):
+    """ TODO """
+
+    @lru_cache(14)  # for test
+    def __new__(cls, *args, **kwargs):
+        return super().__new__(cls)
+
+
+class CosemObjectMethodId(Integer8):
+    """ TODO """
+
+
+class AccessSelectionParameters(Unsigned8):
+    """ Unsigned8(0..1) """
+
+    def __init__(self, value: int | str | Unsigned8 = 1):
+        super(AccessSelectionParameters, self).__init__(value)
+        if int(self) > 1 or int(self) < 0:
+            raise ValueError(F'The {self.__class__.__name__} got {self}, expected 0..1')
+
+
+class CosemAttributeDescriptor(SEQUENCE):
+    class_id: CosemClassId
+    instance_id: CosemObjectInstanceId
+    attribute_id: CosemObjectAttributeId
+    access_selection_parameters: AccessSelectionParameters
+    ELEMENTS = (SequenceElement('class_id', CosemClassId),
+                SequenceElement('instance_id', CosemObjectInstanceId),
+                SequenceElement('attribute_id', CosemObjectAttributeId))
+
+    def __init__(self, value: bytes | tuple | list = None):
+        super(CosemAttributeDescriptor, self).__init__(value)
+        self.__dict__['access_selection_parameters'] = AccessSelectionParameters(0)
+
+    @property
+    def contents(self) -> bytes:
+        """ Always contain Access_Selection_Parameters. DLMS UA 1000-2 Ed.9 Excerpt 9.3.9.1.3 """
+        return super(CosemAttributeDescriptor, self).contents + self.access_selection_parameters.contents
+
+
+class CosemMethodDescriptor(SEQUENCE):
+    class_id: CosemClassId
+    instance_id: CosemObjectInstanceId
+    method_id: CosemObjectMethodId
+    ELEMENTS = (SequenceElement('class_id', CosemClassId),
+                SequenceElement('instance_id', CosemObjectInstanceId),
+                SequenceElement('method_id', CosemObjectMethodId))
+
+    def __init__(self, value: tuple[CosemClassId, CosemObjectInstanceId, CosemObjectMethodId]):
+        super(CosemMethodDescriptor, self).__init__(value)
+
+
+class Data(CHOICE, Protocol):
+    TYPE = cdt.CommonDataType
+
+
+class SelectiveAccessDescriptor(SEQUENCE):
+    """ Selective access specification always starts with an access selector, followed by an access-specific access parameter list.
+    Specified IS/IEC 62056-53 : 2006, 7.4.1.6 Selective access """
+    access_selector: Unsigned8
+    access_parameters: cdt.CommonDataType
+    ELEMENTS: tuple[SequenceElement, SequenceElement]
+
+    def __init__(self, value: tuple | bytes | None = None):
+        super(SelectiveAccessDescriptor, self).__init__(value)
+        self.access_selector.cb_post_set = self.__validate_selector
+
+    @property
+    def ELEMENTS(self) -> tuple[SequenceElement, SequenceElement]:
+        """ return elements """
+
+    def from_default(self):
+        self.values[0] = self.ELEMENTS[0].TYPE()
+        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE()
+
+    def from_bytes(self, value: bytes | bytearray):
+        self.values[0], value = get_instance_and_context(self.ELEMENTS[0].TYPE, value)
+        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE(value)
+
+    def from_tuple(self, value: tuple | list):
+        self.values[0] = self.ELEMENTS[0].TYPE(value[0])
+        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE(value[1])
+
+    @property
+    def contents(self) -> bytes:
+        return self.access_selector.contents+self.access_parameters.encoding
+
+    def __validate_selector(self):
+        self.values[1] = self.ELEMENTS[1].TYPE.ELEMENTS[int(self.access_selector)].TYPE()
+        print('change sel')
+
+    def set_selector(self, index: int, value=None):
+        """ set selector value from int type """
+        self.access_selector.set_contents_from(index)
+        if value:
+            self.access_parameters.set(value)
+
+
+class CosemAttributeDescriptorWithSelection(SEQUENCE):
+    cosem_attribute_descriptor: CosemAttributeDescriptor
+    access_selection: SelectiveAccessDescriptor
+    ELEMENTS: tuple[CosemAttributeDescriptor, SequenceElement]
+
+    def __init__(self, value: bytes | tuple | list = None):
+        super(CosemAttributeDescriptorWithSelection, self).__init__(value)
+        self.cosem_attribute_descriptor.access_selection_parameters.set_contents_from(1)
+
+    @property
+    def ELEMENTS(self) -> tuple[SequenceElement, SequenceElement]:
+        """ return elements. Need initiate in subclass """
+
+
+class InvokeIdAndPriority(Unsigned8):
+
+    def __init__(self, value: bytes | bytearray | str | int | DigitalMixin = 0b1100_0000):
+        super(InvokeIdAndPriority, self).__init__(value)
+        if self.contents[0] & 0b00110000:
+            raise ValueError(F'For {self.__class__.__name__} set reserved bits')
+
+    @classmethod
+    def from_parameters(cls, invoke_id: int = 0,
+                        service_class: int = 0,
+                        priority: int = 0):
+        instance = cls()
+        instance.invoke_id = invoke_id
+        instance.service_class = service_class
+        instance.priority = priority
+        return instance
+
+    @property
+    def invoke_id(self) -> int:
+        return self.contents[0] & 0b0000_1111
+
+    @invoke_id.setter
+    def invoke_id(self, value: int):
+        if value & 0b00001111:
+            self.__dict__['contents'] = int((self.contents[0] & 0b1111_0000) | value).to_bytes(1, 'big')
+        else:
+            raise ValueError(F'Got {value} for invoke-id, expected 0..15')
+
+    @property
+    def service_class(self) -> int:
+        """ 0: Unconfirmed or 1: Confirmed bit return """
+        return (self.contents[0] >> 6) & 0b1
+
+    @service_class.setter
+    def service_class(self, value: int):
+        match value:
+            case 0 | 1: self.__dict__['contents'] = int((self.contents[0] & 0b1011_1111) | (value << 6)).to_bytes(1, 'big')
+            case _:     raise ValueError(F'Got {value} for service_class, expected 0..1')
+
+    @property
+    def priority(self) -> int:
+        """ 0: Normal or 1: High bit return """
+        return (self.contents[0] >> 7) & 0b1
+
+    @priority.setter
+    def priority(self, value: int):
+        match value:
+            case 0 | 1: self.__dict__['contents'] = int((self.contents[0] & 0b0111_1111) | (value << 7)).to_bytes(1, 'big')
+            case _:     raise ValueError(F'Got {value} for service_class, expected 0..1')
+
+    def __str__(self):
+        return F'priority: {"High" if self.contents[0] & 0b1000_0000 else "Normal"}, ' \
+               F'service-class: {"Confirmed" if self.contents[0] & 0b0100_0000 else "Unconfirmed"}, ' \
+               F'invoke-id: {self.invoke_id},'