Decorator for writing and reading the dataclass as binary buffer using struct.
Project description
dataclass_struct 0.9.3
Decorator providing capability to emit and read the decorated dataclass as a binary buffer.
A special new metadata field STRUCT_TYPE
contains the struct format for the dataclass field.
See struct documentation for the formatting options.
FUNCTIONS
The decorator @dataclass_struct
adds following methods
to the decorated dataclass.
from_buffer(self, buffer: bytes, offset=0)
Read the wrapped dataclass from a binary buffer.
:param self: wrapped instance
:param buffer: buffer tp read
:param offset: (optional) offset o start reading
:return: offset after last consumed byte
instance_from_buffer(buffer: bytes, offset=0)
Construct a wrapped class instance from a buffer.
:param buffer: buffer with source binary data
:return: class instance
to_buffer(self, buffer=b'')
Store the wrapped dataclass to a binary buffer.
:param self: wrapped instance
:return: resulting buffer
Limitations
Currently, some features are not supported:
- no inheritance hierarchy
- no tuples
If a dataclass member provides methods from_buffer
and to_buffer
,
it will be included into the resulting buffer and loaded from it
without having STRUCT_TYPE
metadata on its own.
Lists of primitive types can be used, if the number of list members matches the number of format elements. If not, the error message from underlying struct call will be emitted.
Lists of user defined objects will be written/loaded
when the object class provides corresponding methods.
These also don't need STRUCT_TYPE
metadata.
Usage
Simple buffer for a float and an integer
Fields having metadata field STRUCT_TYPE
will be written/read out from the buffer.
from dataclass_struct import STRUCT_TYPE, dataclass_struct
@dataclass_struct
class MyDataClass:
my_flt: float = field(default=0, metadata={STRUCT_TYPE: '<f'})
my_num: int = field(default=0, metadata={STRUCT_TYPE: '<i'})
my_name: str = ''
test_obj = TestModel(3.14, 42)
buff = test_obj.to_buffer()
# buff equals to b'\xc3\xf5H@*\x00\x00\x00'
Class containing another instance of dataclass_struct
Fields providing methods from_buffer
and to_buffer
from dataclasses import field
from dataclass_struct import STRUCT_TYPE, dataclass_struct
@dataclass_struct
class DataClassA:
my_flt: float = field(default=0, metadata={STRUCT_TYPE: '<f'})
my_num: int = field(default=0, metadata={STRUCT_TYPE: '<i'})
@dataclass_struct
class DataClassB:
data_part_a: DataClassA = field(default=DataClassA())
just_a_num: int = field(default=0, metadata={STRUCT_TYPE: '<i'})
test_obj = DataClassB(DataClassA(3.14, 42), 42)
buff = test_obj.to_buffer()
# buff equals to b'\xc3\xf5H@*\x00\x00\x00*\x00\x00\x00'
String with custom encoding
Default encoding for strings is 'utf-8'
.
This can be changed for the whole dataclass by
the decorator parameter use_encoding
,
or for the particular field using metadata ENCODING
.
@dataclass_struct(use_encoding='ascii')
class DefaultEncodingTest:
byte_name: bytes = field(default=b'', metadata={STRUCT_TYPE: '16s'})
str_name: str = field(default='', metadata={STRUCT_TYPE: '16s'})
str_with_enc: str = field(default='', metadata={STRUCT_TYPE: '32s', ENCODING: 'utf-16'})
test_obj = StringTest(b'Hello World', 'Bye bye', 'another one')
buff = test_obj.to_buffer())
# buff equals to b'Hello World\x00\x00\x00\x00\x00Bye bye\x00\x00\x00'\
# b'\x00\x00\x00\x00\x00\x00\xff\xfea\x00n\x00o\x00t\x00h'\
# b'\x00e\x00r\x00 \x00o\x00n\x00e\x00\x00\x00\x00\x00\x00'\
# b'\x00\x00\x00'
Use custom parameters for the underlying dataclass
@dataclass_struct
@dataclass(order=True)
class ExplicitDataclass:
my_flt: float = field(default=0, metadata={STRUCT_TYPE: '<f'})
my_num: int = field(default=0, metadata={STRUCT_TYPE: '<i'})
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file dataclass_struct-0.9.4.tar.gz
.
File metadata
- Download URL: dataclass_struct-0.9.4.tar.gz
- Upload date:
- Size: 4.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.0 CPython/3.9.1
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | af93bbfcf757ac9e1fb9382eee0fdaf189429c9ef531affd8086fc78555381c4 |
|
MD5 | 7ea1c37648643701cdef78fc2288373d |
|
BLAKE2b-256 | eb13139db3d98983845ded5367a572d7aa4c2e5a9135446a94fa11a9b8105bf8 |