Davide Scaini
35 lines
1.0 KiB
Python
35 lines
1.0 KiB
Python
"""Abstract base class for all activity parsers."""
|
|
|
|
import gzip
|
|
import hashlib
|
|
from abc import ABC, abstractmethod
|
|
from pathlib import Path
|
|
|
|
from bincio.extract.models import ParsedActivity
|
|
|
|
|
|
class BaseParser(ABC):
|
|
@abstractmethod
|
|
def parse(self, path: Path, raw_bytes: bytes) -> ParsedActivity:
|
|
"""Parse activity from raw file bytes.
|
|
|
|
Receives pre-read bytes so the factory can compute the hash once and
|
|
handle decompression transparently before dispatching.
|
|
"""
|
|
|
|
@staticmethod
|
|
def _sha256(data: bytes) -> str:
|
|
return "sha256:" + hashlib.sha256(data).hexdigest()
|
|
|
|
@staticmethod
|
|
def _read_file(path: Path) -> tuple[bytes, bytes]:
|
|
"""Return (raw_bytes, decompressed_bytes).
|
|
|
|
raw_bytes is the original file content (used for hashing).
|
|
decompressed_bytes is what parsers should actually parse.
|
|
"""
|
|
raw = path.read_bytes()
|
|
if path.suffix == ".gz":
|
|
return raw, gzip.decompress(raw)
|
|
return raw, raw
|