Module omnipy.modules.raw.serializers

Overview

View Source

from typing import Any, IO, Type

from omnipy.data.dataset import Dataset

from omnipy.data.model import Model

from omnipy.data.serializer import TarFileSerializer

# from typing_inspect import get_generic_bases, get_generic_type, get_origin, get_parameters

class RawDatasetToTarFileSerializer(TarFileSerializer):

    """"""

    @classmethod

    def is_dataset_directly_supported(cls, dataset: Dataset) -> bool:

        return type(dataset) is Dataset[Model[str]]

    @classmethod

    def get_dataset_cls_for_new(cls) -> Type[Dataset]:

        return Dataset[Model[str]]

    @classmethod

    def get_output_file_suffix(cls) -> str:

        return 'raw'

    @classmethod

    def serialize(cls, dataset: Dataset[Model[str]]) -> bytes:

        def raw_encode_func(contents: str) -> bytes:

            return contents.encode('utf8')

        return cls.create_tarfile_from_dataset(dataset, data_encode_func=raw_encode_func)

    @classmethod

    def deserialize(cls, tarfile_bytes: bytes) -> Dataset[Model[str]]:

        dataset = Dataset[Model[str]]()

        def raw_decode_func(file_stream: IO[bytes]) -> str:

            return file_stream.read().decode('utf8')

        def python_dictify_object(obj_type: str, obj_val: Any) -> dict:

            return {obj_type: obj_val}

        cls.create_dataset_from_tarfile(

            dataset,

            tarfile_bytes,

            data_decode_func=raw_decode_func,

            dictify_object_func=python_dictify_object,

            import_method='from_data')  # noqa

        return dataset

Classes

RawDatasetToTarFileSerializer

class RawDatasetToTarFileSerializer(
    /,
    *args,
    **kwargs
)

View Source

class RawDatasetToTarFileSerializer(TarFileSerializer):

    """"""

    @classmethod

    def is_dataset_directly_supported(cls, dataset: Dataset) -> bool:

        return type(dataset) is Dataset[Model[str]]

    @classmethod

    def get_dataset_cls_for_new(cls) -> Type[Dataset]:

        return Dataset[Model[str]]

    @classmethod

    def get_output_file_suffix(cls) -> str:

        return 'raw'

    @classmethod

    def serialize(cls, dataset: Dataset[Model[str]]) -> bytes:

        def raw_encode_func(contents: str) -> bytes:

            return contents.encode('utf8')

        return cls.create_tarfile_from_dataset(dataset, data_encode_func=raw_encode_func)

    @classmethod

    def deserialize(cls, tarfile_bytes: bytes) -> Dataset[Model[str]]:

        dataset = Dataset[Model[str]]()

        def raw_decode_func(file_stream: IO[bytes]) -> str:

            return file_stream.read().decode('utf8')

        def python_dictify_object(obj_type: str, obj_val: Any) -> dict:

            return {obj_type: obj_val}

        cls.create_dataset_from_tarfile(

            dataset,

            tarfile_bytes,

            data_decode_func=raw_decode_func,

            dictify_object_func=python_dictify_object,

            import_method='from_data')  # noqa

        return dataset

---

Static methods

create_dataset_from_tarfile

def create_dataset_from_tarfile(
    dataset: omnipy.data.dataset.Dataset,
    tarfile_bytes: bytes,
    data_decode_func: Callable[[IO[bytes]], Any],
    dictify_object_func: Callable[[str, Any], dict | str],
    import_method='from_data'
)

Parameters:

Name	Type	Description	Default
dataset	Dataset
tarfile_bytes	bytes
data_decode_func	Callable[[IO[bytes]], Any]
dictify_object_func	Callable[[str, Any], dict	str]
import_method			from_data

View Source

    @classmethod

    def create_dataset_from_tarfile(cls,

                                    dataset: Dataset,

                                    tarfile_bytes: bytes,

                                    data_decode_func: Callable[[IO[bytes]], Any],

                                    dictify_object_func: Callable[[str, Any], dict | str],

                                    import_method='from_data'):

        with tarfile.open(fileobj=BytesIO(tarfile_bytes), mode='r:gz') as tarfile_stream:

            for filename in tarfile_stream.getnames():

                obj_type_file = tarfile_stream.extractfile(filename)

                assert filename.endswith(f'.{cls.get_output_file_suffix()}')

                obj_type = '.'.join(filename.split('.')[:-1])

                getattr(dataset, import_method)(

                    dictify_object_func(obj_type, data_decode_func(obj_type_file)))

create_tarfile_from_dataset

def create_tarfile_from_dataset(
    dataset: omnipy.data.dataset.Dataset,
    data_encode_func: Callable[[Any], bytes | memoryview]
)

Parameters:

Name	Type	Description	Default
dataset	Dataset
data_encode_func	Callable[[Any], bytes	memoryview]

View Source

    @classmethod

    def create_tarfile_from_dataset(cls,

                                    dataset: Dataset,

                                    data_encode_func: Callable[[Any], bytes | memoryview]):

        bytes_io = BytesIO()

        with tarfile.open(fileobj=bytes_io, mode='w:gz') as tarfile_stream:

            for obj_type, data_obj in dataset.items():

                json_data_bytestream = BytesIO(data_encode_func(data_obj))

                json_data_bytestream.seek(0)

                tarinfo = TarInfo(name=f'{obj_type}.{cls.get_output_file_suffix()}')

                tarinfo.size = len(json_data_bytestream.getbuffer())

                tarfile_stream.addfile(tarinfo, json_data_bytestream)

        return bytes_io.getbuffer().tobytes()

deserialize

def deserialize(
    tarfile_bytes: bytes
) -> omnipy.data.dataset.Dataset[Annotated[Optional[omnipy.data.model.Model[Annotated[Optional[str], 'Fake Optional from Model']]], 'Fake Optional from Dataset']]

Parameters:

Name	Type	Description	Default
tarfile_bytes	bytes

Returns:

Type	Description
Dataset[Annotated[Optional[Model[Annotated[Optional[str], 'Fake Optional from Model']]], 'Fake Optional from Dataset']]

View Source

    @classmethod

    def deserialize(cls, tarfile_bytes: bytes) -> Dataset[Model[str]]:

        dataset = Dataset[Model[str]]()

        def raw_decode_func(file_stream: IO[bytes]) -> str:

            return file_stream.read().decode('utf8')

        def python_dictify_object(obj_type: str, obj_val: Any) -> dict:

            return {obj_type: obj_val}

        cls.create_dataset_from_tarfile(

            dataset,

            tarfile_bytes,

            data_decode_func=raw_decode_func,

            dictify_object_func=python_dictify_object,

            import_method='from_data')  # noqa

        return dataset

get_dataset_cls_for_new

def get_dataset_cls_for_new(

) -> Type[omnipy.data.dataset.Dataset]

Returns:

Type	Description
Type[Dataset]

View Source

    @classmethod

    def get_dataset_cls_for_new(cls) -> Type[Dataset]:

        return Dataset[Model[str]]

get_output_file_suffix

def get_output_file_suffix(

) -> str

Returns:

Type	Description
str

View Source

    @classmethod

    def get_output_file_suffix(cls) -> str:

        return 'raw'

is_dataset_directly_supported

def is_dataset_directly_supported(
    dataset: omnipy.data.dataset.Dataset
) -> bool

Parameters:

Name	Type	Description	Default
dataset	Dataset

Returns:

Type	Description
bool

View Source

    @classmethod

    def is_dataset_directly_supported(cls, dataset: Dataset) -> bool:

        return type(dataset) is Dataset[Model[str]]

serialize

def serialize(
    dataset: omnipy.data.dataset.Dataset[Annotated[Optional[omnipy.data.model.Model[Annotated[Optional[str], 'Fake Optional from Model']]], 'Fake Optional from Dataset']]
) -> bytes

Parameters:

Name	Type	Description	Default
dataset	Dataset[Annotated[Optional[Model[Annotated[Optional[str], 'Fake Optional from Model']]], 'Fake Optional from Dataset']]

Returns:

Type	Description
bytes

View Source

    @classmethod

    def serialize(cls, dataset: Dataset[Model[str]]) -> bytes:

        def raw_encode_func(contents: str) -> bytes:

            return contents.encode('utf8')

        return cls.create_tarfile_from_dataset(dataset, data_encode_func=raw_encode_func)