fix: Remove msgspec dependency

The builtin json module works just fine. It is a bit slower for large
files, but this will be fixed with the upcoming rust based logfile
parser.

pyproject.toml

@@ -33,7 +33,6 @@ dependencies = [
     "boltons>=24.1.0",
     "construct >=2.10,<3.0",
     "more-itertools >=10.3.0,<11.0.0",
-    "msgspec >=0.11,<0.19",
     "platformdirs >=2.6,<5.0",
     "pydantic >=2.0,<3.0",
     "tabulate >=0.9",

src/gallia/cli/hr.py

@@ -3,15 +3,14 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import argparse
+import json
 import os
 import signal
 import sys
 from itertools import islice
 from pathlib import Path
 from typing import cast
 
-import msgspec
-
 from gallia import exitcodes
 from gallia.log import ColorMode, PenlogPriority, PenlogReader, resolve_color_mode
 
@@ -89,7 +88,7 @@ def _main() -> int:
 def main() -> None:
     try:
         sys.exit(_main())
-    except (msgspec.DecodeError, msgspec.ValidationError) as e:
+    except json.JSONDecodeError as e:
         print(f"invalid file format: {e}", file=sys.stderr)
         sys.exit(exitcodes.DATAERR)
     # BrokenPipeError appears when stuff is piped to | head.

src/gallia/command/base.py

@@ -3,6 +3,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import asyncio
+import dataclasses
 import json
 import os
 import os.path
@@ -19,7 +20,6 @@
 from tempfile import gettempdir
 from typing import Any, Protocol, Self, cast
 
-import msgspec
 from pydantic import ConfigDict, field_serializer, model_validator
 
 from gallia import exitcodes
@@ -49,15 +49,16 @@ class HookVariant(Enum):
     POST = "post"
 
 
-class RunMeta(msgspec.Struct):
+@dataclasses.dataclass
+class RunMeta:
     command: str
     start_time: str
     end_time: str
     exit_code: int
     config: MutableMapping[str, Any]
 
     def json(self) -> str:
-        return msgspec.json.encode(self).decode()
+        return json.dumps(dataclasses.asdict(self))
 
 
 logger = get_logger(__name__)

src/gallia/log.py

@@ -5,9 +5,11 @@
 from __future__ import annotations
 
 import atexit
+import dataclasses
 import datetime
 import gzip
 import io
+import json
 import logging
 import mmap
 import os
@@ -18,15 +20,13 @@
 import time
 import traceback
 from collections.abc import Iterator
-from dataclasses import dataclass
 from enum import Enum, IntEnum, unique
 from logging.handlers import QueueHandler, QueueListener
 from pathlib import Path
 from queue import Queue
 from types import TracebackType
 from typing import TYPE_CHECKING, Any, BinaryIO, Self, TextIO, TypeAlias, cast
 
-import msgspec
 import zstandard
 
 if TYPE_CHECKING:
@@ -327,12 +327,14 @@ def add_zst_log_handler(
     return zstd_handler
 
 
-class _PenlogRecordV2(msgspec.Struct, omit_defaults=True, tag=2, tag_field="version"):
+@dataclasses.dataclass
+class _PenlogRecordV2:
     module: str
     host: str
     data: str
     datetime: str
     priority: int
+    version: int
     tags: list[str] | None = None
     line: str | None = None
     stacktrace: str | None = None
@@ -417,7 +419,7 @@ def _format_record(  # noqa: PLR0913
     return msg
 
 
-@dataclass
+@dataclasses.dataclass
 class PenlogRecord:
     module: str
     host: str
@@ -459,20 +461,26 @@ def parse_json(cls, data: bytes) -> Self:
         if data.startswith(b"<"):
             data = data[data.index(b">") + 1 :]
 
-        record = msgspec.json.decode(data, type=_PenlogRecordV2)
+        record = json.loads(data.decode())
+        if (v := record["version"]) != 2:
+            raise json.JSONDecodeError(f"invalid log record version {v}", data.decode(), 0)
 
         return cls(
-            module=record.module,
-            host=record.host,
-            data=record.data,
-            datetime=datetime.datetime.fromisoformat(record.datetime),
-            priority=PenlogPriority(record.priority),
-            tags=record.tags,
-            line=record.line,
-            stacktrace=record.stacktrace,
-            _python_level_no=record._python_level_no,
-            _python_level_name=record._python_level_name,
-            _python_func_name=record._python_func_name,
+            module=record["module"],
+            host=record["host"],
+            data=record["data"],
+            datetime=datetime.datetime.fromisoformat(record["datetime"]),
+            priority=PenlogPriority(record["priority"]),
+            tags=record["tags"] if "tags" in record else None,
+            line=record["line"] if "line" in record else None,
+            stacktrace=record["stacktrace"] if "stacktrace" in record else None,
+            _python_level_no=record["_python_level_no"] if "_python_level_no" in record else None,
+            _python_level_name=record["_python_level_name"]
+            if "_python_level_name" in record
+            else None,
+            _python_func_name=record["_python_func_name"]
+            if "_python_func_name" in record
+            else None,
         )
 
     def to_log_record(self) -> logging.LogRecord:
@@ -692,9 +700,9 @@ def format(self, record: logging.LogRecord) -> str:
             _python_level_no=record.levelno,
             _python_level_name=record.levelname,
             _python_func_name=record.funcName,
+            version=2,
         )
-
-        return msgspec.json.encode(penlog_record).decode()
+        return json.dumps(dataclasses.asdict(penlog_record))
 
 
 class _ConsoleFormatter(logging.Formatter):