Merge branch 'update'

docs/blog/index.md

@@ -1 +1 @@
-# Blog
+# Note

docs/blog/posts/2022-01-30-numpy_reshape.md

@@ -8,7 +8,7 @@ description: >
 categories:
     - Data Analysis
 tags:
-    - numpy
+    - NumPy
     - array
 ---
 

docs/blog/posts/2022-07-17-sympy_tutorial.md

@@ -8,7 +8,7 @@ description: >
 categories:
     - Mathematics
 tags:
-    - sympy
+    - SymPy
 ---
 
 symbolic computation을 다루는 SymPy 튜토리얼  

docs/blog/posts/2023-06-10-fastapi_orm.md

@@ -10,8 +10,8 @@ categories:
 tags:
     - FastAPI
     - ORM
-    - SQLAlchemy
-    - Alembic
+    - aqlalchemy
+    - alembic
 ---
 
 SQLAlchemy와 Alembic을 활용한 데이터베이스 FastAPI 서버 ORM 활용  

docs/blog/posts/2023-09-19-tensorflow_install_note.md

@@ -9,7 +9,7 @@ categories:
     - AI
 tags:
     - AI
-    - tensorflow
+    - TensorFlow
 ---
 
 TensorFlow 설치 관련 주의할 점들에 대한 기록들  

docs/blog/posts/2023-09-21-python_logging.md

@@ -135,14 +135,14 @@ Python이 기본 제공하는 다양한 Log Handler 중에 [TimedRotatingFileHan
 로그 필터를 직접 만들어 주입해주면 **특정 정보의 로그**만 필터링 해줄 수 있다.  
 
 ```python title="src/log/filter.py"
-import logging
+from logging import Filter, LogRecord
 
 
-class MyFilter(logging.Filter):  # (1)!
+class MyFilter(Filter):  # (1)!
     def __init__(self, levels: list[int]):
         self.__level = levels
 
-    def filter(self, logRecord):
+    def filter(self, logRecord: LogRecord):
         return logRecord.levelno in self.__level
 ```
 
@@ -200,12 +200,12 @@ class MyFilter(logging.Filter):  # (1)!
     ```python title="src/log/formatter.py"
     import datetime as dt
     import json
-    import logging
+    from logging import Formatter, LogRecord
 
-    KST = dt.timezone(dt.timedelta(hours=9))
+    KST = dt.timezone(offset=dt.timedelta(hours=9))
 
 
-    class JsonFormatter(logging.Formatter):
+    class JsonFormatter(Formatter):
         def __init__(
             self,
             *,
@@ -215,11 +215,11 @@ class MyFilter(logging.Filter):  # (1)!
             self.fmt_keys = fmt_keys if fmt_keys is not None else {}
 
         # override
-        def format(self, record: logging.LogRecord) -> str:
+        def format(self, record: LogRecord) -> str:
             message = self._prepare_log(record=record)
-            return json.dumps(message, default=str)
+            return json.dumps(obj=message, default=str)
 
-        def _prepare_log(self, record: logging.LogRecord):
+        def _prepare_log(self, record: LogRecord):
             always_fields = {
                 "message": record.getMessage(),
                 "timestamp": (
@@ -234,9 +234,11 @@ class MyFilter(logging.Filter):  # (1)!
                 always_fields["stack_info"] = self.formatStack(record.stack_info)
 
             message = {
-                key: msg_val
-                if (msg_val := always_fields.pop(val, None)) is not None
-                else getattr(record, val)
+                key: (
+                    msg_val
+                    if (msg_val := always_fields.pop(val, None)) is not None
+                    else getattr(record, val)
+                )
                 for key, val in self.fmt_keys.items()
             }
             message.update(always_fields)
@@ -251,13 +253,13 @@ class MyFilter(logging.Filter):  # (1)!
     ```python title="src/log/formatter.py"
     import datetime as dt
     import json
-    import logging
+    from logging import Formatter, LogRecord
     from typing import override
 
     KST = dt.timezone(dt.timedelta(hours=9))
 
 
-    class JsonFormatter(logging.Formatter):
+    class JsonFormatter(Formatter):
         def __init__(
             self,
             *,
@@ -267,15 +269,15 @@ class MyFilter(logging.Filter):  # (1)!
             self.fmt_keys = fmt_keys if fmt_keys is not None else {}
 
         @override
-        def format(self, record: logging.LogRecord) -> str:
+        def format(self, record: LogRecord) -> str:
             message = self._prepare_log(record=record)
             return json.dumps(message, default=str)
 
-        def _prepare_log(self, record: logging.LogRecord):
+        def _prepare_log(self, record: LogRecord):
             always_fields = {
                 "message": record.getMessage(),
                 "timestamp": (
-                    dt.datetime.fromtimestamp(record.created, tz=KST)  # (1)!
+                    dt.datetime.fromtimestamp(timestamp=record.created, tz=KST)  # (1)!
                 ).isoformat(),
             }
 
@@ -286,17 +288,19 @@ class MyFilter(logging.Filter):  # (1)!
                 always_fields["stack_info"] = self.formatStack(record.stack_info)
 
             message = {
-                key: msg_val
-                if (msg_val := always_fields.pop(val, None)) is not None
-                else getattr(record, val)
+                key: (
+                    msg_val
+                    if (msg_val := always_fields.pop(val, None)) is not None
+                    else getattr(record, val)
+                )
                 for key, val in self.fmt_keys.items()
             }
             message.update(always_fields)
 
             return message
     ```
 
-    1. UTC 기준으로 로그를 생성하고 싶다면 `dt.datetime.fromtimestamp(record.created, tz=dt.timezone.utc)`으로 만들면 된다.
+    1. UTC 기준으로 로그를 생성하고 싶다면 `dt.datetime.fromtimestamp(timestamp=record.created, tz=dt.timezone.utc)`으로 만들면 된다.
 
 ## Best Practice
 
@@ -305,7 +309,7 @@ class MyFilter(logging.Filter):  # (1)!
 ```python title="src/log/config.py"
 import logging
 import queue
-from logging import StreamHandler
+from logging import Formatter, StreamHandler
 from logging.handlers import QueueHandler, QueueListener, TimedRotatingFileHandler
 from pathlib import Path
 
@@ -319,22 +323,22 @@ except FileExistsError:
     ...
 
 # create Logger instance
-logger = logging.getLogger("logger")
-logger.setLevel(logging.DEBUG)
+logger = logging.getLogger(name="logger")
+logger.setLevel(level=logging.DEBUG)
 
 # set log format
-simple_formatter = logging.Formatter(
+simple_formatter = Formatter(
     fmt="%(asctime)s - %(levelname)s - %(message)s",
     datefmt="%Y-%m-%dT%H:%M:%S%z",
 )
-detailed_formatter = logging.Formatter(
+detailed_formatter = Formatter(
     fmt="%(asctime)s - %(levelname)s - [%(module)s:%(lineno)d] %(message)s",
     datefmt="%Y-%m-%dT%H:%M:%S%z",
 )
 
 # StreamHandler
 stream_handler = StreamHandler()
-stream_handler.setFormatter(simple_formatter)
+stream_handler.setFormatter(fmt=simple_formatter)
 
 # TimedRotatingFileHandler
 file_handler = TimedRotatingFileHandler(
@@ -343,7 +347,7 @@ file_handler = TimedRotatingFileHandler(
     backupCount=3,  # define number of log files, 0 to save all log files
     encoding="utf-8",
 )
-file_handler.setFormatter(detailed_formatter)
+file_handler.setFormatter(fmt=detailed_formatter)
 
 # JsonlHandler
 json_handler = TimedRotatingFileHandler(
@@ -360,7 +364,7 @@ fmt_keys = {
     "function": "funcName",
     "line": "lineno",
 }
-json_handler.setFormatter(formatter.JsonFormatter(fmt_keys=fmt_keys))
+json_handler.setFormatter(fmt=formatter.JsonFormatter(fmt_keys=fmt_keys))
 
 # DebugHandler
 debug_handler = TimedRotatingFileHandler(
@@ -369,15 +373,21 @@ debug_handler = TimedRotatingFileHandler(
     backupCount=3,
     encoding="utf-8",
 )
-debug_handler.setFormatter(detailed_formatter)
+debug_handler.setFormatter(fmt=detailed_formatter)
 debug_handler.addFilter(
-    filter.MyFilter([logging.DEBUG, logging.ERROR, logging.CRITICAL])
+    filter=filter.MyFilter(
+        levels=[
+            logging.DEBUG,
+            logging.ERROR,
+            logging.CRITICAL,
+        ]
+    )
 )
 
 # QueueHandler
 log_queue = queue.Queue()  # (1)!
-queue_handler = QueueHandler(log_queue)
-logger.addHandler(queue_handler)
+queue_handler = QueueHandler(queue=log_queue)
+logger.addHandler(hdlr=queue_handler)
 
 # QueueListener
 log_listener = QueueListener(
@@ -409,7 +419,7 @@ def main():
     try:
         raise Exception
     except Exception as e:
-        logger.exception(e)  # log for error catch
+        logger.exception(e)  # logging error traceback
 
     log_listener.stop()
 
@@ -520,24 +530,30 @@ from pathlib import Path
 
 from src.log import filter
 
-logger = logging.getLogger("logger")
+logger = logging.getLogger(name="logger")
 
 
 def set_logger():
     log_config_file = Path(r"config\log_config.json")
-    with open(log_config_file, encoding="utf-8") as f:
-        log_config = json.load(f)
-    logging.config.dictConfig(log_config)
+    with open(file=log_config_file, encoding="utf-8") as f:
+        log_config = json.load(fp=f)
+    logging.config.dictConfig(config=log_config)
 
-    queue_handler = logging.getHandlerByName("queue_handler")
+    queue_handler = logging.getHandlerByName(name="queue_handler")
     if queue_handler is not None:
         queue_handler.listener.start()
-        atexit.register(queue_handler.listener.stop)
+        atexit.register(func=queue_handler.listener.stop)
 
-    debug_handler = logging.getHandlerByName("debug_handler")
+    debug_handler = logging.getHandlerByName(name="debug_handler")
     if debug_handler is not None:
         debug_handler.addFilter(
-            filter.MyFilter([logging.DEBUG, logging.ERROR, logging.CRITICAL])
+            filter=filter.MyFilter(
+                levels=[
+                    logging.DEBUG,
+                    logging.ERROR,
+                    logging.CRITICAL,
+                ]
+            )
         )
 ```
 

docs/blog/posts/2023-12-16-redis_basic.md

@@ -9,7 +9,7 @@ categories:
     - Data Engineering
 tags:
     - in-memory database
-    - redis
+    - Redis
 ---
 
 Redis의 입문을 위한 간단 정리  

docs/blog/posts/2024-05-05-sqlalchemy_alembic.md

@@ -48,10 +48,10 @@ class User(Base):
 매핑 Entity에 속성을 부여하는 방법은 여러 가지가 있는데, 아래와 같이 추상 테이블 객체에 칼럼을 만들어두면, 해당 추상 테이블을 상속 받는 테이블들에는 해당 칼럼이 기본적으로 생성되게 된다.  
 
 ```python title="src/db/entity.py"
+from datetime import datetime
 from enum import IntEnum
 
-from sqlalchemy.orm import DeclarativeBase, mapped_column
-from sqlalchemy.schema import Column
+from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column
 from sqlalchemy.types import BigInteger, DateTime, String
 
 
@@ -65,7 +65,7 @@ class BaseEntity(DeclarativeBase): ...
 class BigintIdEntity(BaseEntity):
     __abstract__ = True  # (1)!
 
-    id = mapped_column(
+    id: Mapped[int] = mapped_column(
         type_=BigInteger,
         primary_key=True,
         autoincrement=True,
@@ -76,11 +76,8 @@ class BigintIdEntity(BaseEntity):
 class UserEntity(BigintIdEntity):
     __tablename__ = "user"
 
-    name = Column(
-        String(length=UserEntityEnum.NAME.value),  # (3)!
-        nullable=False,
-    )
-    created_at = Column(DateTime, nullable=False)
+    name: Mapped[str] = mapped_column(String(length=UserEntityEnum.NAME.value))  # (3)!
+    created_at: Mapped[datetime] = mapped_column(DateTime)
 ```
 
 1. 추상 테이블 표시를 통해 Alembic으로 테이블을 관리할 때 테이블 생성에서 제외
@@ -89,11 +86,6 @@ class UserEntity(BigintIdEntity):
 
 `BaseEntity`에 속성을 부여하면 모든 테이블이 해당 속성을 사용하게 된다. 특정 테이블들만 속성을 공유하도록 하려면 중간에 추상 테이블을 별도로 생성해야 한다.  
 
-!!! tip
-    개인적으로는 더 직관적으로 보이는 `Column` 객체 사용을 더 선호하는데, `Column` 객체는 `sort_order` 속성이 없어서 추상 테이블에서 선언한 칼럼의 순서를 설정해줄 수 없다.  
-
-    기존 테이블에 ORM을 통해 연결할 경우 문제가 없지만, Alembic을 통해 테이블을 생성하고 관리하게 된다면 상속 받은 칼럼이 나중에 생성되어 칼럼 순서가 직관적이지 않은 문제가 발생한다.  
-
 !!! warning
     위 예시처럼 칼럼 길이를 Enum으로 별도 관리할 때는 SQLAlchemy Entity에 입력할 때는 반드시 `.value` 까지 입력해서 값만 불러오도록 해야한다. Alembic이 Enum 객체를 제대로 인식하지 못해 revision 생성 시 칼럼 길이에 값 대신 Enum 객체를 입력해버리는 문제가 있다.