db2_hj3415 0.1.7__tar.gz → 0.1.9__tar.gz

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: db2_hj3415
-Version: 0.1.7
+Version: 0.1.9
 Summary: Gathering the stock data by playwright
 Author-email: Hyungjin Kim <hj3415@gmail.com>
 Description-Content-Type: text/markdown
@@ -10,7 +10,6 @@ Requires-Dist: pandas
 Requires-Dist: pandas-stubs
 Requires-Dist: deepdiff
 Requires-Dist: utils_hj3415>=3.2.3
-Requires-Dist: redis
 Requires-Dist: pydantic
 Project-URL: Home, https://www.hyungjin.kr
 

db2_hj3415-0.1.9/db2_hj3415/common/connection.py

@@ -0,0 +1,36 @@
+# MongoDB 연결
+import os
+from motor.motor_asyncio import AsyncIOMotorClient
+
+# 싱글톤 몽고 클라이언트 정의
+MONGO_URI = os.getenv("MONGO_ADDR", "mongodb://localhost:27017")
+client: AsyncIOMotorClient = None
+
+def get_mongo_client() -> AsyncIOMotorClient:
+    global client
+    if client is None:
+        client = AsyncIOMotorClient(MONGO_URI)
+    return client
+
+def close_mongo_client():
+    if client:
+        client.close()
+
+
+from pymongo import MongoClient
+
+client_sync: MongoClient = None  # 동기 클라이언트 타입으로 변경
+
+def get_mongo_client_sync() -> MongoClient:
+    """
+    MongoDB 동기 클라이언트를 반환합니다.
+    전역 client가 None일 경우 새 클라이언트를 생성합니다.
+    """
+    global client_sync
+    if client_sync is None:
+        client_sync = MongoClient(MONGO_URI)
+    return client_sync
+
+def close_mongo_client_sync():
+    if client_sync:
+        client_sync.close()

db2_hj3415-0.1.9/db2_hj3415/common/utils.py

@@ -0,0 +1,25 @@
+# 자주 쓰는 간단한 유틸 함수
+import pandas as pd
+import numpy as np
+
+def df_to_dict_replace_nan(df: pd.DataFrame) -> list[dict]:
+    # NaN → None으로 변환
+    return df.replace({np.nan: None}).to_dict(orient="records")
+
+import json
+from bson import json_util
+from pydantic import BaseModel
+
+def pretty_print(obj):
+    def convert(o):
+        if isinstance(o, BaseModel):
+            return o.model_dump(by_alias=True)
+        if isinstance(o, dict):
+            return {k: convert(v) for k, v in o.items()}
+        if isinstance(o, list):
+            return [convert(v) for v in o]
+        return o  # 기본값 (예: str, int, float 등)
+
+    data = convert(obj)
+
+    print(json.dumps(data, indent=2, ensure_ascii=False, default=json_util.default))

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/__init__.py

@@ -4,5 +4,6 @@ DATE_FORMAT = "%Y.%m.%d"
 from db2_hj3415.common import connection
 from db2_hj3415.common.utils import *
 from db2_hj3415.common.db_ops import *
+
 from db2_hj3415.nfs._ops import *
 from db2_hj3415.nfs.models import *

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/_c10346.py

@@ -62,7 +62,7 @@ async def _compare_and_log_diff(code: str, new_doc: dict, latest_doc: dict | Non
     return True
 
 
-def _prepare_c10346_document(code: str, data: dict[str, pd.DataFrame]) -> dict:
+def _prepare_c10346_document_for_save(code: str, data: dict[str, pd.DataFrame]) -> dict:
     now = datetime.now(timezone.utc)
     document = {"코드": code, "날짜": now}
 
@@ -80,7 +80,7 @@ async def save(col: str, code: str, data: dict[str, pd.DataFrame], client: Async
 
     await collection.create_index([("코드", ASCENDING), ("날짜", ASCENDING)], unique=True)
 
-    document = _prepare_c10346_document(code, data)
+    document = _prepare_c10346_document_for_save(code, data)
     latest_doc = await collection.find_one({"코드": code}, sort=[("날짜", DESCENDING)])
 
     need_save = await _compare_and_log_diff(code, document, latest_doc, client)
@@ -106,7 +106,7 @@ async def save_many(col: str, many_data: dict[str, dict[str, pd.DataFrame]], cli
     results = []
 
     for code, data in many_data.items():
-        document = _prepare_c10346_document(code, data)
+        document = _prepare_c10346_document_for_save(code, data)
         latest_doc = await collection.find_one({"코드": code}, sort=[("날짜", DESCENDING)])
         need_save = await _compare_and_log_diff(code, document, latest_doc, client)
 
@@ -127,7 +127,7 @@ async def save_many(col: str, many_data: dict[str, dict[str, pd.DataFrame]], cli
     return results
 
 
-async def get_latest(col: str, code: str, client: AsyncIOMotorClient) -> C103 | C104 | C106 | None:
+async def fetch_latest_doc(col: str, code: str, client: AsyncIOMotorClient) -> dict | None:
     collection = get_collection(client, DB_NAME, col)
 
     # 최신 날짜 기준으로 정렬하여 1건만 조회
@@ -142,19 +142,89 @@ async def get_latest(col: str, code: str, client: AsyncIOMotorClient) -> C103 |
 
     latest_doc["_id"] = str(latest_doc["_id"])
 
-    try:
-        match col:
-            case 'c103':
-                return C103(**latest_doc)
-            case 'c104':
-                return C104(**latest_doc)
-            case 'c106':
-                return C106(**latest_doc)
-
-    except Exception as e:
-        print(f"[{code}] C103 파싱 실패: {e}")
+    return latest_doc
+
+
+async def get_latest_as_model(col: str, code: str, client: AsyncIOMotorClient) -> C103 | C104 | C106 | None:
+    latest_doc = await fetch_latest_doc(col, code, client)
+
+    if not latest_doc:
         return None
 
+    match col:
+        case 'c103':
+            return C103(**latest_doc)
+        case 'c104':
+            return C104(**latest_doc)
+        case 'c106':
+            return C106(**latest_doc)
+        case _:
+            raise ValueError(f"지원하지 않는 컬렉션 이름: {col}")
+
+
+async def get_latest_doc_as_df_dict(col: str, code: str, client: AsyncIOMotorClient) -> dict[str, pd.DataFrame] | None:
+    latest_doc = await fetch_latest_doc(col, code, client)
+
+    if not latest_doc:
+        return None
+
+    keys_map = {
+        'c103': [
+            "손익계산서y", "재무상태표y", "현금흐름표y",
+            "손익계산서q", "재무상태표q", "현금흐름표q"
+        ],
+        'c104': [
+            '수익성y', '성장성y', '안정성y', '활동성y', '가치분석y',
+            '수익성q', '성장성q', '안정성q', '활동성q', '가치분석q'
+        ],
+        'c106': ['q', 'y']
+    }
+
+    def make_df_dict(keys: list[str]) -> dict[str, pd.DataFrame]:
+        """latest_doc에서 주어진 키들을 기반으로 DataFrame 생성"""
+        return {
+            key: pd.DataFrame(latest_doc.get(key) or []) for key in keys
+        }
+
+    def concat_by_suffix(dfs: dict[str, pd.DataFrame]) -> dict[str, pd.DataFrame]:
+        """'q', 'y' 접미사 기준으로 그룹 후 하나의 DataFrame으로 합침"""
+        grouped: dict[str, list[pd.DataFrame]] = {'q': [], 'y': []}
+
+        for key, df in dfs.items():
+            if key.endswith('q'):
+                grouped['q'].append(df.assign(분류=key))
+            elif key.endswith('y'):
+                grouped['y'].append(df.assign(분류=key))
+
+        def clean_df(df: pd.DataFrame) -> pd.DataFrame:
+            # 모든 값이 NaN인 열 제거
+            return df.dropna(axis=1, how='all')
+
+        return {
+            period: (
+                pd.concat(
+                    [clean_df(f) for f in frames if not f.empty],
+                    ignore_index=True
+                ) if any(not f.empty for f in frames) else pd.DataFrame()
+            )
+            for period, frames in grouped.items()
+        }
+
+    # 2. 실제 처리 로직
+    match col:
+        case 'c103':
+            return make_df_dict(keys_map['c103'])
+
+        case 'c104':
+            raw_result = make_df_dict(keys_map['c104'])
+            return concat_by_suffix(raw_result)
+
+        case 'c106':
+            return make_df_dict(keys_map['c106'])
+
+        case _:
+            raise ValueError(f"지원하지 않는 컬렉션 이름: {col}")
+
 
 async def has_doc_changed(col: str, code: str, client: AsyncIOMotorClient) -> bool:
     """

db2_hj3415-0.1.9/db2_hj3415/nfs/_ops.py

@@ -0,0 +1,82 @@
+from motor.motor_asyncio import AsyncIOMotorClient
+from db2_hj3415.nfs import DB_NAME, connection
+from db2_hj3415.nfs.models import CodeName
+
+
+async def get_all_codes(client: AsyncIOMotorClient) -> list[str]:
+    """
+    c103, c104, c106 컬렉션에 모두 존재하는 코드의 리스트를 반환함.
+
+    Args:
+        client (AsyncIOMotorClient): MongoDB 비동기 클라이언트 객체
+
+    Returns:
+        list[str]: c103, c104, c106 컬렉션에 공통으로 존재하는 종목 코드 리스트
+    """
+    db = client[DB_NAME]
+
+    collections = ['c103', 'c104', 'c106']
+
+    # 첫 컬렉션으로 초기화
+    s = set(await db[collections[0]].distinct("코드"))
+
+    for col in collections[1:]:
+        codes = await db[col].distinct("코드")
+        s &= set(codes)
+
+    return list(s)
+
+
+def get_all_codes_sync() -> list[str]:
+    """
+    c103, c104, c106 컬렉션에 모두 존재하는 코드의 리스트를 반환함.
+    """
+    client = connection.get_mongo_client_sync()
+    try:
+        db = client[DB_NAME]
+        collections = ['c103', 'c104', 'c106']
+
+        # 첫 컬렉션 코드 셋팅
+        common_codes = set(db[collections[0]].distinct("코드"))
+
+        for col in collections[1:]:
+            codes = db[col].distinct("코드")
+            common_codes &= set(codes)
+
+        return sorted(common_codes)  # 필요에 따라 정렬
+    finally:
+        connection.close_mongo_client_sync()
+
+
+async def get_all_codes_names(client: AsyncIOMotorClient) -> list[CodeName]:
+    collection = client[DB_NAME]['c101']
+    cursor = collection.find({}, {"코드": 1, "종목명": 1, "_id": 0})
+    result = []
+    async for doc in cursor:
+        result.append(CodeName(**doc))
+    return result
+
+
+def get_all_codes_names_sync() -> list[CodeName] | None:
+    client = connection.get_mongo_client_sync()
+    try:
+        collection = client[DB_NAME]['c101']
+        cursor = collection.find({}, {"코드": 1, "종목명": 1, "_id": 0})
+        return [CodeName(**doc) for doc in cursor]
+    finally:
+        connection.close_mongo_client_sync()
+
+async def delete_code_from_all_collections(code: str, client: AsyncIOMotorClient) -> dict[str, int]:
+    db = client[DB_NAME]
+
+    collections = ['c101', 'c103', 'c104', 'c106', 'c108']
+
+    deleted_counts = {}
+
+    for col in collections:
+        result = await db[col].delete_many({"코드": code})
+        deleted_counts[col] = result.deleted_count
+
+    print(f"삭제된 도큐먼트 갯수: {deleted_counts}")
+    return deleted_counts
+

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/c101.py

@@ -104,10 +104,18 @@ async def get_latest(code: str, client: AsyncIOMotorClient) -> C101 | None:
         mylogger.debug(doc)
         return C101(**doc)
     else:
-        print(f"데이터 없음: {code}")
+        mylogger.warning(f"데이터 없음: {code}")
         return None
 
 
+async def get_name(code: str, client: AsyncIOMotorClient) -> str | None:
+    c101_data = await get_latest(code, client)
+    if c101_data is None:
+        return None
+    else:
+        return c101_data.종목명
+
+
 SortOrder = Literal["asc", "desc"]
 
 async def get_all_data(code: str, client: AsyncIOMotorClient, sort: SortOrder = 'asc') -> list[C101]:

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/c103.py

@@ -1,3 +1,5 @@
+from typing import Literal
+
 from motor.motor_asyncio import AsyncIOMotorClient
 import pandas as pd
 
@@ -17,8 +19,18 @@ async def save_many(many_data: dict[str, dict[str, pd.DataFrame]], client: Async
     return await _c10346.save_many(COL_NAME, many_data, client)
 
 
-async def get_latest(code: str, client: AsyncIOMotorClient) -> C103 | None:
-    return await _c10346.get_latest(COL_NAME, code, client)
+ReturnType = C103 | dict[str, pd.DataFrame] | None
+
+
+async def get_latest(code: str, client: AsyncIOMotorClient, as_type: Literal["model", "dataframe"] = "model") -> ReturnType:
+    if as_type == "model":
+        return await _c10346.get_latest_as_model(COL_NAME, code, client)
+
+    elif as_type == "dataframe":
+        return await _c10346.get_latest_doc_as_df_dict(COL_NAME, code, client)
+
+    else:
+        raise ValueError(f"지원하지 않는 반환 타입: '{as_type}' (허용값: 'model', 'dataframe')")
 
 
 async def has_doc_changed(code: str, client: AsyncIOMotorClient) -> bool:

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/c104.py

@@ -1,3 +1,5 @@
+from typing import Literal
+
 from motor.motor_asyncio import AsyncIOMotorClient
 import pandas as pd
 
@@ -17,8 +19,18 @@ async def save_many(many_data: dict[str, dict[str, pd.DataFrame]], client: Async
     return await _c10346.save_many(COL_NAME, many_data, client)
 
 
-async def get_latest(code: str, client: AsyncIOMotorClient) -> C104 | None:
-    return await _c10346.get_latest(COL_NAME, code, client)
+ReturnType = C104 | dict[str, pd.DataFrame] | None
+
+
+async def get_latest(code: str, client: AsyncIOMotorClient, as_type: Literal["model", "dataframe"] = "model") -> ReturnType:
+    if as_type == "model":
+        return await _c10346.get_latest_as_model(COL_NAME, code, client)
+
+    elif as_type == "dataframe":
+        return await _c10346.get_latest_doc_as_df_dict(COL_NAME, code, client)
+
+    else:
+        raise ValueError(f"지원하지 않는 반환 타입: '{as_type}' (허용값: 'model', 'dataframe')")
 
 
 async def has_doc_changed(code: str, client: AsyncIOMotorClient) -> bool:

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/c106.py

@@ -1,3 +1,5 @@
+from typing import Literal
+
 from motor.motor_asyncio import AsyncIOMotorClient
 import pandas as pd
 
@@ -17,8 +19,18 @@ async def save_many(many_data: dict[str, dict[str, pd.DataFrame]], client: Async
     return await _c10346.save_many(COL_NAME, many_data, client)
 
 
-async def get_latest(code: str, client: AsyncIOMotorClient) -> C106 | None:
-    return await _c10346.get_latest(COL_NAME, code, client)
+ReturnType = C106 | dict[str, pd.DataFrame] | None
+
+
+async def get_latest(code: str, client: AsyncIOMotorClient, as_type: Literal["model", "dataframe"] = "model") -> ReturnType:
+    if as_type == "model":
+        return await _c10346.get_latest_as_model(COL_NAME, code, client)
+
+    elif as_type == "dataframe":
+        return await _c10346.get_latest_doc_as_df_dict(COL_NAME, code, client)
+
+    else:
+        raise ValueError(f"지원하지 않는 반환 타입: '{as_type}' (허용값: 'model', 'dataframe')")
 
 
 async def has_doc_changed(code: str, client: AsyncIOMotorClient) -> bool:

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/db2_hj3415/nfs/models.py

@@ -1,10 +1,15 @@
 from pydantic import BaseModel, Field, field_serializer, ConfigDict, field_validator
 from datetime import datetime
 
+class CodeName(BaseModel):
+    코드: str
+    종목명: str | None
+
 class C101(BaseModel):
     id: str | None = Field(default=None, alias="_id")
-    날짜: datetime
     코드: str
+    날짜: datetime
+    종목명: str | None
     bps: int | None
     eps: int | None
     pbr: float | None
@@ -26,7 +31,7 @@ class C101(BaseModel):
     외국인지분율: float | None
     유동비율: float | None
     전일대비: int | None
-    종목명: str | None
+
     주가: int | None
     최고52: int | None
     최저52: int | None

db2_hj3415-0.1.9/db2_hj3415/valuation/__init__.py

@@ -0,0 +1,9 @@
+DB_NAME = "valuation"
+DATE_FORMAT = "%Y.%m.%d"
+
+from db2_hj3415.common import connection
+from db2_hj3415.common.utils import *
+from db2_hj3415.common.db_ops import *
+from db2_hj3415.nfs._ops import get_all_codes_sync, get_all_codes
+
+from db2_hj3415.valuation.models import *

db2_hj3415-0.1.9/db2_hj3415/valuation/_ops.py

@@ -0,0 +1,65 @@
+from pymongo import ASCENDING
+from motor.motor_asyncio import AsyncIOMotorClient
+from datetime import datetime, timezone, time
+
+from db2_hj3415.valuation import DB_NAME, MilData, RedData
+from db2_hj3415.common.db_ops import get_collection
+from utils_hj3415 import setup_logger
+from bson import ObjectId
+
+mylogger = setup_logger(__name__, 'WARNING')
+
+T = RedData | MilData
+
+async def save(col: str, data: T, client: AsyncIOMotorClient) -> dict:
+    collection = get_collection(client, DB_NAME, col)
+    await collection.create_index([("날짜", ASCENDING), ("코드", ASCENDING)], unique=True)
+
+    data.날짜 = datetime.now(timezone.utc)
+
+    # 날짜 기준 중복 확인
+    today = data.날짜.date()
+
+    existing = await collection.find_one({
+        "코드": data.코드,
+        "날짜": {
+            "$gte": datetime.combine(today, time.min).replace(tzinfo=timezone.utc),
+            "$lt": datetime.combine(today, time.max).replace(tzinfo=timezone.utc)
+        }
+    })
+    mylogger.debug(f"이미 저장된 오늘 날짜 데이터가 있나?: {existing}")
+
+    if existing:
+        return {"status": "skipped", "reason": "already_saved_today"}
+
+    # datetime 그대로 유지하기 위해 mode='python' 사용
+    doc = data.model_dump(by_alias=True, mode='python', exclude_none=False)
+
+    # ObjectId가 존재하면 업데이트, 아니면 삽입
+    if '_id' in doc:
+        if doc['_id'] is None:
+            doc.pop('_id')  # None이면 제거하여 MongoDB에서 자동 생성되게 함
+        else:
+            doc['_id'] = ObjectId(doc['_id']) if isinstance(doc['_id'], str) else doc['_id']
+            await collection.replace_one({'_id': doc['_id']}, doc, upsert=True)
+            return {"status": "updated", "_id": str(doc['_id'])}
+
+    result = await collection.insert_one(doc)
+    data.id = str(result.inserted_id)
+    return {"status": "inserted", "_id": data.id}
+
+
+async def save_many(col: str, many_data: dict[str, T], client: AsyncIOMotorClient) -> dict:
+    results = {}
+
+    # 이 방식이 속도는 느리지만 제일 간단하고 안정적인 방식임.
+    for code, data in many_data.items():
+        try:
+            result = await save(col, data, client)
+            results[code] = result
+        except Exception as e:
+            # 에러 발생 시 로깅 또는 실패 처리
+            results[code] = {"status": "error", "error": str(e)}
+            mylogger.error(f"[{code}] 저장 중 오류 발생: {e}")
+
+    return results

db2_hj3415-0.1.9/db2_hj3415/valuation/blue.py

@@ -0,0 +1,17 @@
+from motor.motor_asyncio import AsyncIOMotorClient
+
+from db2_hj3415.valuation import BlueData, _ops
+from utils_hj3415 import setup_logger
+
+
+mylogger = setup_logger(__name__, 'WARNING')
+
+COL_NAME = "blue"
+
+
+async def save(blue_data: BlueData, client: AsyncIOMotorClient) -> dict:
+    return await _ops.save(COL_NAME, blue_data, client)
+
+
+async def save_many(many_data: dict[str, BlueData], client: AsyncIOMotorClient) -> dict:
+    return await _ops.save_many(COL_NAME, many_data, client)

db2_hj3415-0.1.9/db2_hj3415/valuation/growth.py

@@ -0,0 +1,17 @@
+from motor.motor_asyncio import AsyncIOMotorClient
+
+from db2_hj3415.valuation import GrowthData, _ops
+from utils_hj3415 import setup_logger
+
+
+mylogger = setup_logger(__name__, 'WARNING')
+
+COL_NAME = "growth"
+
+
+async def save(growth_data: GrowthData, client: AsyncIOMotorClient) -> dict:
+    return await _ops.save(COL_NAME, growth_data, client)
+
+
+async def save_many(many_data: dict[str, GrowthData], client: AsyncIOMotorClient) -> dict:
+    return await _ops.save_many(COL_NAME, many_data, client)

db2_hj3415-0.1.9/db2_hj3415/valuation/mil.py

@@ -0,0 +1,17 @@
+from motor.motor_asyncio import AsyncIOMotorClient
+
+from db2_hj3415.valuation import MilData, _ops
+from utils_hj3415 import setup_logger
+
+
+mylogger = setup_logger(__name__, 'WARNING')
+
+COL_NAME = "mil"
+
+
+async def save(mil_data: MilData, client: AsyncIOMotorClient) -> dict:
+    return await _ops.save(COL_NAME, mil_data, client)
+
+
+async def save_many(many_data: dict[str, MilData], client: AsyncIOMotorClient) -> dict:
+    return await _ops.save_many(COL_NAME, many_data, client)

db2_hj3415-0.1.9/db2_hj3415/valuation/models.py

@@ -0,0 +1,193 @@
+from typing import Any
+
+from pydantic import BaseModel, Field, field_validator, field_serializer, ConfigDict, model_validator, SerializationInfo
+from datetime import datetime
+import math
+from utils_hj3415 import tools
+
+
+class RedData(BaseModel):
+    id: str | None = Field(default=None, alias="_id")
+    코드: str
+    날짜: datetime | None = Field(default=None)
+    종목명: str
+
+    사업가치: float | None
+    지배주주당기순이익: float | None
+    expect_earn: float | None
+
+    재산가치: float | None
+    유동자산: float | None
+    유동부채: float | None
+    투자자산: float | None
+    투자부동산: float | None
+
+    부채평가: float | None
+    발행주식수: int | None
+
+    자료제출일: list[str] | None = Field(default=None)
+    주가: float | None
+    red_price: float | None
+    score: int | None
+
+    @model_validator(mode='before')
+    @classmethod
+    def replace_nan_with_none(cls, values: dict) -> dict:
+        return {
+            k: (None if isinstance(v, float) and math.isnan(v) else v)
+            for k, v in values.items()
+        }
+
+    @field_serializer("날짜")
+    def serialize_날짜(self, value: datetime, info: SerializationInfo) -> str | datetime:
+        # JSON 응답용일 때만 문자열로 직렬화
+        if info.mode == 'json':
+            return value.isoformat()
+        return value
+
+    @field_validator("코드")
+    @classmethod
+    def validate_코드(cls, v):
+        if not tools.is_6digit(v):
+            raise ValueError(f"code는 6자리 숫자형 문자열이어야 합니다. (입력값: {v})")
+        return v
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        str_strip_whitespace=True,
+    )
+
+
+class Evaluation(BaseModel):
+    최근값: float | None = Field(default=None)
+    시계열: dict[str, float] = Field(default_factory=dict)
+    평가결과: dict[str, Any] = Field(default_factory=dict)
+    model_config = ConfigDict(
+        extra="allow"
+    )
+
+
+class MilData(BaseModel):
+    id: str | None = Field(default=None, alias="_id")
+    코드: str
+    날짜: datetime | None = Field(default=None)
+    종목명: str
+
+    주주수익률: float | None
+    이익지표: float | None
+
+    # 투자수익률
+    ROIC: Evaluation
+    ROE: Evaluation
+    ROA: Evaluation
+
+    # 가치지표
+    FCF: Evaluation
+    PFCF: Evaluation
+    PCR: Evaluation
+
+    @model_validator(mode='before')
+    @classmethod
+    def replace_nan_with_none(cls, values: dict) -> dict:
+        return {
+            k: (None if isinstance(v, float) and math.isnan(v) else v)
+            for k, v in values.items()
+        }
+
+    @field_serializer("날짜")
+    def serialize_날짜(self, value: datetime, info: SerializationInfo) -> str | datetime:
+        # JSON 응답용일 때만 문자열로 직렬화
+        if info.mode == 'json':
+            return value.isoformat()
+        return value
+
+    @field_validator("코드")
+    @classmethod
+    def validate_코드(cls, v):
+        if not tools.is_6digit(v):
+            raise ValueError(f"code는 6자리 숫자형 문자열이어야 합니다. (입력값: {v})")
+        return v
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        str_strip_whitespace=True,
+    )
+
+
+class BlueData(BaseModel):
+    id: str | None = Field(default=None, alias="_id")
+    코드: str
+    날짜: datetime | None = Field(default=None)
+    종목명: str
+
+    유동비율: float
+
+    재고자산회전율: Evaluation
+    이자보상배율: Evaluation
+    순운전자본회전율: Evaluation | None
+    순부채비율: Evaluation
+
+    자료제출일: list[str] | None = Field(default=None)
+
+    @model_validator(mode='before')
+    @classmethod
+    def replace_nan_with_none(cls, values: dict) -> dict:
+        return {
+            k: (None if isinstance(v, float) and math.isnan(v) else v)
+            for k, v in values.items()
+        }
+
+    @field_serializer("날짜")
+    def serialize_날짜(self, value: datetime, info: SerializationInfo) -> str | datetime:
+        # JSON 응답용일 때만 문자열로 직렬화
+        if info.mode == 'json':
+            return value.isoformat()
+        return value
+
+    @field_validator("코드")
+    @classmethod
+    def validate_코드(cls, v):
+        if not tools.is_6digit(v):
+            raise ValueError(f"code는 6자리 숫자형 문자열이어야 합니다. (입력값: {v})")
+        return v
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        str_strip_whitespace=True,
+    )
+
+
+class GrowthData(BaseModel):
+    id: str | None = Field(default=None, alias="_id")
+    코드: str
+    날짜: datetime | None = Field(default=None)
+    종목명: str
+
+    매출액증가율: Evaluation
+
+    @model_validator(mode='before')
+    @classmethod
+    def replace_nan_with_none(cls, values: dict) -> dict:
+        return {
+            k: (None if isinstance(v, float) and math.isnan(v) else v)
+            for k, v in values.items()
+        }
+
+    @field_serializer("날짜")
+    def serialize_날짜(self, value: datetime, info: SerializationInfo) -> str | datetime:
+        # JSON 응답용일 때만 문자열로 직렬화
+        if info.mode == 'json':
+            return value.isoformat()
+        return value
+
+    @field_validator("코드")
+    @classmethod
+    def validate_코드(cls, v):
+        if not tools.is_6digit(v):
+            raise ValueError(f"code는 6자리 숫자형 문자열이어야 합니다. (입력값: {v})")
+        return v
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        str_strip_whitespace=True,
+    )

db2_hj3415-0.1.9/db2_hj3415/valuation/red.py

@@ -0,0 +1,17 @@
+from motor.motor_asyncio import AsyncIOMotorClient
+
+from db2_hj3415.valuation import RedData, _ops
+from utils_hj3415 import setup_logger
+
+
+mylogger = setup_logger(__name__, 'WARNING')
+
+COL_NAME = "red"
+
+
+async def save(red_data: RedData, client: AsyncIOMotorClient) -> dict:
+    return await _ops.save(COL_NAME, red_data, client)
+
+
+async def save_many(many_data: dict[str, RedData], client: AsyncIOMotorClient) -> dict:
+    return await _ops.save_many(COL_NAME, many_data, client)

{db2_hj3415-0.1.7 → db2_hj3415-0.1.9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "flit_core.buildapi"
 
 [project]
 name = "db2_hj3415"
-version = "0.1.7"
+version = "0.1.9"
 authors = [{name = "Hyungjin Kim", email = "hj3415@gmail.com"}]
 description = "Gathering the stock data by playwright"
 readme = "README.md"
@@ -15,7 +15,6 @@ dependencies = [
     "pandas-stubs",
     "deepdiff",
     "utils_hj3415>=3.2.3",
-    "redis",
     "pydantic",
 ]
 

db2_hj3415-0.1.7/db2_hj3415/common/connection.py

@@ -1,29 +0,0 @@
-# MongoDB/Redis 연결
-import os
-from motor.motor_asyncio import AsyncIOMotorClient
-import redis.asyncio as redis
-from redis.asyncio.client import Redis
-
-# 싱글톤 몽고 클라이언트 정의
-MONGO_URI = os.getenv("MONGO_ADDR", "mongodb://localhost:27017")
-client: AsyncIOMotorClient = None
-
-def get_mongo_client() -> AsyncIOMotorClient:
-    global client
-    if client is None:
-        client = AsyncIOMotorClient(MONGO_URI)
-    return client
-
-def close_mongo_client():
-    if client:
-        client.close()
-
-
-async def get_redis() -> Redis:
-    client = redis.Redis(host="localhost", port=6379, decode_responses=True)
-    try:
-        await client.ping()
-        return client
-    except Exception as e:
-        print("Redis connection failed:", e)
-        return None

db2_hj3415-0.1.7/db2_hj3415/common/utils.py

@@ -1,24 +0,0 @@
-# 자주 쓰는 간단한 유틸 함수
-import pandas as pd
-import numpy as np
-
-def df_to_dict_replace_nan(df: pd.DataFrame) -> list[dict]:
-    # NaN → None으로 변환
-    return df.replace({np.nan: None}).to_dict(orient="records")
-
-import json
-from bson import json_util
-from pydantic import BaseModel
-
-def pretty_print(obj):
-    if isinstance(obj, BaseModel):
-        # Pydantic 모델이면 dict로 변환
-        data = obj.model_dump(by_alias=True)
-    elif isinstance(obj, list) and all(isinstance(o, BaseModel) for o in obj):
-        # 리스트 안에 BaseModel만 있다면 변환
-        data = [o.model_dump(by_alias=True) for o in obj]
-    else:
-        # 일반 dict나 기타 객체
-        data = obj
-
-    print(json.dumps(data, indent=2, ensure_ascii=False, default=json_util.default))

db2_hj3415-0.1.7/db2_hj3415/nfs/_ops.py

@@ -1,42 +0,0 @@
-from motor.motor_asyncio import AsyncIOMotorClient
-from db2_hj3415.nfs import DB_NAME
-
-
-async def get_all_codes(client: AsyncIOMotorClient) -> list[str]:
-    """
-    c103, c104, c106 컬렉션에 모두 존재하는 코드의 리스트를 반환함.
-
-    Args:
-        client (AsyncIOMotorClient): MongoDB 비동기 클라이언트 객체
-
-    Returns:
-        list[str]: c103, c104, c106 컬렉션에 공통으로 존재하는 종목 코드 리스트
-    """
-    db = client[DB_NAME]
-
-    collections = ['c103', 'c104', 'c106']
-
-    # 첫 컬렉션으로 초기화
-    s = set(await db[collections[0]].distinct("코드"))
-
-    for col in collections[1:]:
-        codes = await db[col].distinct("코드")
-        s &= set(codes)
-
-    return list(s)
-
-
-async def delete_code_from_all_collections(code: str, client: AsyncIOMotorClient) -> dict[str, int]:
-    db = client[DB_NAME]
-
-    collections = ['c101', 'c103', 'c104', 'c106', 'c108']
-
-    deleted_counts = {}
-
-    for col in collections:
-        result = await db[col].delete_many({"코드": code})
-        deleted_counts[col] = result.deleted_count
-
-    print(f"삭제된 도큐먼트 갯수: {deleted_counts}")
-    return deleted_counts
-