最新コードを追加

README.md

@@ -3,4 +3,17 @@
 ```sh
 python -m venv venv
 .\venv\Scripts\activate
-```
+```
+
+
+curl https://api.openai.com/v1/audio/transcriptions \
+  -H "Authorization: Bearer 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" \
+  -F "file=@part_001.m4a" \
+  -F "model=gpt-4o-transcribe" \
+  -F "response_format=text" \
+  -F "language=ja" \
+  -o part_001.txt
+
+ffmpeg -i voice.m4a -f segment -segment_time 600 -c copy part_%03d.m4a
+
+# フォーマットを揃える

docments/big_query.md

@@ -0,0 +1,12 @@
+## Big Queryは「データウェアハウス型」
+
+Big QueryはOLAP(分析)用途の列指向データベースなので、
+RDB(MySQL,PostgreSQLなど)とは違い、行の一意性や整合性制約を持たない設計です。
+
+## BigQueryに存在しない制約
+
+* PRIMARY KEY
+* FOREIGN KEY
+* UNIQUE
+* CHECK
+* AUTO_INCREMENT

docments/windows_disk.md

@@ -0,0 +1,66 @@
+
+
+```powershell
+# 一時ファイルの削除
+Remove-Item -Path "$env:TEMP\*" -Recurse -Force -ErrorAction SilentlyContinue
+```
+
+
+移動できるデータを他ドライブへ
+
+
+```powershell
+Get-ChildItem C:\ -Recurse -ErrorAction SilentlyContinue |
+  Where-Object { -not $_.PSIsContainer } |
+  Sort-Object Length -Descending |
+  Select-Object FullName, @{Name="Size(GB)";Expression={"{0:N2}" -f ($_.Length / 1GB)}} -First 20
+```
+
+
+サードパーティツールで整理
+
+WinDirStat
+→ ディスクの使用状況を可視化して、大きいファイルを一目で把握。
+
+```
+winget install WinDirStat
+windirstat
+```
+
+CCleaner
+→ 一時ファイルや不要レジストリの掃除に便利。
+
+
+```
+Remove-Item "C:\adobeTemp\*" -Recurse -Force -ErrorAction SilentlyContinue
+
+
+ren "C:\adobeTemp" "adobeTemp_old"
+```
+
+attrib -h -s -r "C:\adobeTemp"
+
+ren C:\adobeTemp C:\adobeTemp_old
+ren "C:\adobeTemp" "adobeTemp_old"
+
+mkdir "F:\adobeTemp"
+
+mklink /J C:\adobeTemp F:\adobeTemp
+
+
+ren C:\adobeTemp C:\adobeTemp_old
+リンク先のフォルダを作成(Dドライブに新しい保存場所を用意）
+
+powershell
+Copy code
+mkdir D:\adobeTemp
+ジャンクションを作成
+管理者権限のコマンドプロンプトで：
+
+cmd
+Copy code
+mklink /J C:\adobeTemp D:\adobeTemp
+
+
+cd "C:\Users\r_yam\AppData\Local\Docker\wsl\data"
+optimize-vhd -Path .\ext4.vhdx -Mode Full

example/example_bigquery.py

@@ -0,0 +1,110 @@
+import sys
+import os
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),"..", "src")))
+
+from lib.custom_logger import get_logger
+logger = get_logger(level=10)
+
+from providers.google_cloud_bigquery_provider import GoogleCloudBigQueryProvider
+
+
+def example_bigquery():
+    try:
+        # GoogleCloudBigQueryProviderのインスタンスを作成
+        provider = GoogleCloudBigQueryProvider(
+            cred_path="keys/google_service_accout.json",
+        )
+
+        dss = provider.get_datasets()
+        for ds in dss:
+            logger.info(f"Dataset ID: {ds.dataset_id}")
+
+        # データセットを作成する
+        dataset_id = "example_dataset"
+        if provider.is_exists_dataset(dataset_id):
+            logger.info(f"Dataset {dataset_id} already exists.")
+        else:
+            dataset = provider.create_dataset(dataset_id)
+            logger.info(f"Dataset {dataset_id} created at {dataset.created}.")
+
+        table_id = provider.full_table_id(dataset_id, "example_table")
+        # テーブルを作成する
+        if provider.is_exists_table(table_id):
+            logger.info(f"Table {table_id} already exists.")
+        else:
+            logger.info(f"Creating table {table_id}...")
+            schema = [
+                provider.addSchemaField("device_code", "string", "REQUIRED", description="Device code"),
+                provider.addSchemaField("time_stamp", "timestamp", "REQUIRED", description="Timestamp"),
+            ]
+            table = provider.create_table(
+                table_id=table_id,
+                schema=schema
+            )
+
+        # tables = provider.get_tables(dataset_id)
+        # for table in tables:
+        #     logger.info(f"Table ID: {table.table_id}")
+            
+        # テーブル情報を挿入する
+        # provider.insert_rows(
+        #     table_id=table_id,
+        #     rows=[
+        #         {"device_code": "device_001", "time_stamp": "2025-01-01 12:00:00"},
+        #         {"device_code": "device_002", "time_stamp": "2025-01-01 12:05:00"},
+        #         {"device_code": "device_003", "time_stamp": "2025-01-01 12:10:00"},
+        #     ]
+        # )
+        # テーブル情報を確認する(JOBあり・SQL発行)
+        # job_qyuery = provider.excute_query(
+        #     query=f"SELECT * FROM `{table_id}` where device_code='device_001'",
+        # )
+        # results = job_qyuery.result()
+        # for row in results:
+        #     logger.info(f"Row: {row}")
+
+
+        # テーブル情報を確認する(JOBあり)
+        # rows = provider.list_rows(
+        #     table_id=table_id,
+        # )
+        # for row in rows:
+        #     logger.info(f"Row: {row}")
+
+        # バッファ状況の確認
+        # buffer_status = provider.get_streaming_buffer_info(table_id=table_id)
+        # logger.info(f"Streaming Buffer Info: {buffer_status}")
+        # テーブルを更新する
+        # query_job = provider.update_query(
+        #     table_id=table_id,
+        #     values={
+        #         "device_code": "'device_999'"
+        #     },
+        #     were_clause="device_code='device_002'"
+        # )
+        # query_job.result()  # ジョブの完了を待機
+        
+
+        # # テーブル情報を確認する(JOBあり)
+        # job = provider.select_query(
+        #     table_id=table_id,
+        # )
+        # results = job.result()
+        # for row in results:
+        #     logger.info(f"Row: {row}")
+
+        # # テーブルのレコードを削除する
+        # provider.delete_query(
+        #     table_id=table_id,
+        #     were_clause="device_code='device_012'"
+        # )
+        # テーブルを削除する
+        # provider.delete_table(table_id=table_id)
+                
+
+    except Exception as e:
+        logger.error(f"Error in example_bigquery: {e}")
+
+
+
+example_bigquery()

example/example_bigquery_model.py

@@ -0,0 +1,74 @@
+import sys
+import os
+from datetime import datetime
+
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),"..", "src")))
+from typing import ClassVar, Optional
+
+from lib.custom_logger import get_logger
+logger = get_logger(level=10)
+from dataclasses import dataclass
+
+from models.bigquery_base_model import BigQueryBaseModel
+from providers.google_cloud_bigquery_provider import GoogleCloudBigQueryProvider
+
+@dataclass
+class ExamplModel(BigQueryBaseModel):
+    device_code: str
+    time_stamp: datetime
+
+    table_id: ClassVar[str] = "example_dataset.example_table"
+    key_field: ClassVar[Optional[str]] = "device_code"
+
+def example_model():
+    logger.info("Starting example_bigquery_model function.")
+    provider = GoogleCloudBigQueryProvider(
+        cred_path="keys/google_service_accout.json",
+    )
+    ExamplModel.set_provider(provider)
+
+    # テーブル内の全レコードを取得する
+    # records = ExamplModel.fetch_all()
+    # logger.info(f"Total records: {len(records)}")
+    # for record in records:
+    #     logger.info(f"Record: {record}")
+
+    # レコードを生成する
+    # new_record = ExamplModel(
+    #     device_code="device_010",
+    #     time_stamp=datetime(2025, 1, 1, 15, 0, 0)
+    # )
+    # new_record.create()
+
+    # レコードを大量に生成する
+    # ExamplModel.insert(
+    #     rows=[
+    #     ExamplModel(device_code="device_011",time_stamp=datetime(2025, 1, 1, 15, 0, 0)),        
+    #     ExamplModel(device_code="device_012",time_stamp=datetime(2025, 1, 1, 15, 0, 0)),        
+    # ])
+
+    # テーブルのストリームバッファ情報を取得する
+    if ExamplModel.is_streaming_buffer():
+        logger.info("Table is currently receiving streaming data.")
+        logger.info("not updated or delete in the streaming buffer.")
+    else:
+        logger.info("Table is not receiving streaming data.")
+        # 特定の条件でレコードを削除する
+        # ExamplModel.delete(where=[("device_code", "=", "device_010")])
+        # レコードを更新する
+        ExamplModel.update(
+            values={"device_code": "device_011_updated"},
+            where=[("device_code", "=", "device_011")]
+        )
+
+
+    
+    # records = ExamplModel.list(where=[("device_code", "=", "device_001")])
+    # logger.info(f"Records with device_code='device_001': {records if records else 'No records found'}")
+
+
+    # record = ExamplModel.first()
+    # logger.info(f"First record: {record if record else 'No record found'}")
+
+
+example_model()

example/example_duck_db.py

@@ -0,0 +1,25 @@
+import sys
+import os
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),"..", "src")))
+
+from lib.custom_logger import get_logger
+logger = get_logger(level=10)
+
+from providers.duck_db_provider import DuckDBProvider
+
+def example_duckdb():
+    logger.info("Starting example_duckdb function.")    
+    file_path = "./data_science/data/y=*/news/news_*.csv"
+    provider = DuckDBProvider()
+    sql = f"""
+        SELECT * 
+        FROM read_csv_auto('{file_path}' , HEADER=TRUE, IGNORE_ERRORS=TRUE)
+        """        
+    result = provider.query_df(sql)
+    print("latest published_parsed:", result)
+
+    # forで1件ずつ表示
+    for idx, row in result.iterrows():
+        logger.info(f"title:{row['title']}")
+
+example_duckdb()

example/example_duck_gcs.py

@@ -0,0 +1,87 @@
+"""
+HMACキーがひつようになる
+* Google Cloud Consoleで発行する
+* https://console.cloud.google.com/storage/settings
+* 「相互運用性(Interoperability)」タブを開く
+* 「HMACキーを作成」ボタンを押す
+* 使いたいサービスアカウントを選択
+"""
+
+import sys
+import os
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),"..", "src")))
+
+from dotenv import load_dotenv
+load_dotenv()
+
+
+from lib.custom_logger import get_logger
+logger = get_logger(level=10)
+
+import duckdb
+from providers.google_cloud_storage_provider import  GoogleCloudStorageProvider
+from providers.duck_db_provider import DuckDBProvider
+
+
+def example_init_google_cloud_storage():
+
+    logger.info("Starting example_google_cloud_storage function.")    
+    gcs_provider = GoogleCloudStorageProvider(
+        cred_path="./keys/google_service_accout.json",
+    )
+    csv_data = """id,name,age,city,score,created_at
+1,Alice,25,Tokyo,88,2025-10-01T09:00:00Z
+2,Bob,30,Osaka,75,2025-10-02T09:30:00Z
+3,Charlie,28,Nagoya,92,2025-10-03T10:00:00Z
+4,David,35,Fukuoka,64,2025-10-04T11:15:00Z
+5,Eva,22,Sapporo,80,2025-10-05T12:45:00Z
+"""    
+    gcs_provider.write_item("datasource-example-251018", 
+                            "example/y=2025/m=10/example.csv", 
+                            csv_data.encode("utf-8"),"text/csv")
+
+    buckets = gcs_provider.get_buckets()
+    logger.info(f"Buckets: {buckets}")
+
+
+def example_duckdb_cloud_raw():
+    logger.info("Starting example_duckdb_cloud_raw function.")    
+
+    # DuckDB接続
+    con = duckdb.connect()
+    con.sql(f"""
+        CREATE OR REPLACE SECRET gcs_creds (
+            TYPE gcs,
+            KEY_ID {os.getenv('GCP_STORAGE_HMAC_ACCESS_KEY')},
+            SECRET {os.getenv('GCP_STORAGE_HMAC_SECRET_KEY')}
+        );
+    """)
+ 
+    query = f"""
+    SELECT * FROM read_csv_auto('gs://datasource-example-251018/example/y=2025/m=10/example.csv');
+    """
+    result = con.execute(query).df()
+    logger.info(f"Read {len(result)} rows from GCS file.")
+
+
+def example_duckdb_cloud_class():
+    logger.info("Starting example_duckdb_cloud_class function.")    
+    # DuckDB接続
+    provider = DuckDBProvider()
+    provider.setup_gcs(
+        access_key=os.getenv('GCP_STORAGE_HMAC_ACCESS_KEY'),
+        secret_key=os.getenv('GCP_STORAGE_HMAC_SECRET_KEY'),
+    )
+    bucket_name = "datasource-example-251018"
+    object_name = "example/y=2025/m=*/example.csv"
+
+    query = f"""
+    SELECT * FROM {provider.get_gs_csv_name(bucket_name, object_name)};
+    """
+    result = provider.query_df(query)
+    logger.info(f"Read {len(result)} rows from GCS file using DuckDBProvider.")
+
+
+# example_init_google_cloud_storage()
+# example_duckdb_cloud_raw()
+example_duckdb_cloud_class()

example/example_duck_model.py

@@ -0,0 +1,30 @@
+import sys
+import os
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),"..", "src")))
+
+from typing import ClassVar, Optional
+
+from lib.custom_logger import get_logger
+logger = get_logger(level=10)
+from dataclasses import dataclass
+
+from models.duck_base_model import DuckBaseModel
+from providers.duck_db_provider import DuckDBProvider
+
+@dataclass
+class NewsModel(DuckBaseModel):
+    title: str
+    file_glob: ClassVar[str] = "./data_science/data/y=2025/news/news_*.csv"
+    default_order_by: ClassVar[Optional[str]] = "pub_date DESC"
+
+
+def example_duckdb_model():
+    logger.info("Starting example_duckdb_model function.")
+    provider = DuckDBProvider()
+    provider.connect()
+    NewsModel.provider = provider
+
+    news = NewsModel.first()
+    logger.info(f"Total news: {news if news else 'No news found'}")
+
+# example_duckdb_model()

example/example_rss.py

@@ -0,0 +1,47 @@
+# openai_rss_reader_stdlib.py
+# pip install feedparser
+
+import feedparser
+from email.utils import parsedate_to_datetime
+
+FEEDS = [
+    "https://platform.openai.com/docs/release-notes.rss",  # リリースノート
+    "https://openai.com/blog/rss.xml",                     # ブログ
+]
+
+def to_ts(entry: feedparser.FeedParserDict):
+    # published_parsed / updated_parsed があれば直接datetime化
+    if entry.get("published_parsed"):
+        return parsedate_to_datetime(entry.published)
+    if entry.get("updated_parsed"):
+        return parsedate_to_datetime(entry.updated)
+    return None
+
+def uniq_key(entry: feedparser.FeedParserDict):
+    return entry.get("id") or entry.get("guid") or entry.get("link")
+
+def fetch_all(feeds):
+    items = []
+    seen = set()
+    for url in feeds:
+        d: feedparser.FeedParserDict = feedparser.parse(url)
+        for e in d.entries:
+            k = uniq_key(e)
+            if not k or k in seen:
+                continue
+            seen.add(k)
+            ts = to_ts(e)
+            items.append({
+                "title": e.get("title", "(no title)"),
+                "link": e.get("link"),
+                "published": e.get("published") or e.get("updated"),
+                "ts": ts,
+                "source": url,
+            })
+    # ts が None のものは末尾に回す
+    items.sort(key=lambda x: (x["ts"] is not None, x["ts"]), reverse=True)
+    return items
+
+if __name__ == "__main__":
+    for i, it in enumerate(fetch_all(FEEDS), 1):
+        print(f"{i:02d}. {it['title']}\n    {it['link']}\n    {it['published']}  [{it['source']}]\n")

requirements.txt

@@ -2,13 +2,26 @@ matplotlib
 requests
 pyttsx3
 
-
+# pylib
+pandas
 # firebase_provider
 firebase-admin>=7.1.0
 # google cloud storage
 google-cloud-storage
+# google cloud big query
+google-cloud-bigquery
+# google cloud big query option
+google-cloud-bigquery-storage 
+# bigquery to dataframe
+db-dtypes 
 # onedrive
 msal
 
 # common
 python-dotenv
+# RSS
+feedparser
+# model
+pydantic
+# duckdb
+duckdb==1.3.2

src/lib/custom_logger.py

@@ -28,7 +28,6 @@ class CustomLogger(Singleton):
 
     def __init__(self, name='main', log_file=None, level=logging.INFO):
         if hasattr(self, '_initialized') and self._initialized:
-            self.logger.setLevel(level)
             return  # すでに初期化済みなら何もしない
 
         self.logger = logging.getLogger(name)

src/lib/singleton.py

@@ -2,7 +2,7 @@
 This implementation is thread-safe and ensures that only one instance of the class is created.
 
 Singleton が提供するのは「同じインスタンスを返す仕組み」
-* __init__() は毎回呼ばれる（多くの人が意図しない動作）
+* __init__() は毎回呼ばれる(多くの人が意図しない動作）
 * __init__の2回目は_initialized というフラグは 使う側で管理する必要がある。
 """
 

src/lib/time_schedule.py

@@ -14,7 +14,7 @@ class TaskScheduler:
 # インスタンスを作成し、スレッドを起動
 scheduler = TaskScheduler()
 
-# インスタンスを破棄（しかしスレッドは続行）
+# インスタンスを破棄(しかしスレッドは続行）
 # del scheduler
 
 # メインスレッドは他の作業を続ける

src/models/bigquery_base_model.py

@@ -0,0 +1,510 @@
+import os
+import base64
+from typing import Any, Dict, Optional,Type,TypeVar,Tuple,Sequence,Union,List,Iterable,ClassVar
+from enum import Enum
+from decimal import Decimal
+
+
+from datetime import datetime, date, time, timezone
+from dataclasses import dataclass, fields, is_dataclass
+
+import pandas as pd
+from providers.google_cloud_bigquery_provider import GoogleCloudBigQueryProvider
+
+from lib.custom_logger import get_logger
+logger = get_logger()
+
+T = TypeVar("T", bound="BigQueryBaseModel")
+
+Filter = Tuple[str, str, Any]
+
+def _build_where_clause_bq_without_prams(
+    where: Optional[Union[str, Dict[str, Any], Sequence[Tuple[str, str, Any]]]]
+)->str:
+    """
+    BigQuery 向け WHERE 句の生成(パラメータなし)
+    Returns:
+        where_sql
+    """
+    if where is None:
+        return ""
+
+    clauses = []
+
+    if isinstance(where, str):
+        # 文字列のまま(完全SQL指定)
+        return f"WHERE {where}"
+
+    if isinstance(where, dict):
+        for col, val in where.items():
+            if isinstance(val, (list, tuple, set)):
+                placeholders = ", ".join([f"'{str(v).replace('\'', '\\\'')}'" for v in val])
+                clauses.append(f"{col} IN ({placeholders})")
+            elif val is None:
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} = '{str(val).replace('\'', '\\\'')}'")
+    else:
+        # sequence of (col, op, val)
+        for col, op, val in where:
+            if (op.upper() == "IN") and isinstance(val, (list, tuple, set)):
+                placeholders = ", ".join([f"'{str(v).replace('\'', '\\\'')}'" for v in val])
+                clauses.append(f"{col} IN ({placeholders})")
+            elif val is None and op in ("=", "=="):
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} {op} '{str(val).replace('\'', '\\\'')}'")
+
+    where_sql = "WHERE " + " AND ".join(clauses)
+    return where_sql
+
+def _build_where_clause_bq(
+    where: Optional[Union[str, Dict[str, Any], Sequence[Tuple[str, str, Any]]]]
+) -> Tuple[str, List[Dict[str, Any]]]:
+    """
+    BigQuery 向け WHERE 句とパラメータの生成
+    Returns:
+        (where_sql, query_params)
+    """
+    if where is None:
+        return "", []
+
+    params = []
+    clauses = []
+
+    if isinstance(where, str):
+        # 文字列のまま(完全SQL指定)
+        return f"WHERE {where}", params
+
+    if isinstance(where, dict):
+        for i, (col, val) in enumerate(where.items()):
+            param_name = f"param{i}"
+            if isinstance(val, (list, tuple, set)):
+                placeholders = []
+                for j, v in enumerate(val):
+                    pname = f"{param_name}_{j}"
+                    placeholders.append(f"@{pname}")
+                    params.append({"name": pname, "parameterType": {"type": "STRING"}, "parameterValue": {"value": str(v)}})
+                clauses.append(f"{col} IN ({', '.join(placeholders)})")
+            elif val is None:
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} = @{param_name}")
+                params.append({"name": param_name, "parameterType": {"type": "STRING"}, "parameterValue": {"value": str(val)}})
+    else:
+        # sequence of (col, op, val)
+        for i, (col, op, val) in enumerate(where):
+            param_name = f"param{i}"
+            if (op.upper() == "IN") and isinstance(val, (list, tuple, set)):
+                placeholders = []
+                for j, v in enumerate(val):
+                    pname = f"{param_name}_{j}"
+                    placeholders.append(f"@{pname}")
+                    params.append({"name": pname, "parameterType": {"type": "STRING"}, "parameterValue": {"value": str(v)}})
+                clauses.append(f"{col} IN ({', '.join(placeholders)})")
+            elif val is None and op in ("=", "=="):
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} {op} @{param_name}")
+                params.append({"name": param_name, "parameterType": {"type": "STRING"}, "parameterValue": {"value": str(val)}})
+
+    where_sql = "WHERE " + " AND ".join(clauses)
+    return where_sql, params
+
+
+def to_bq_json(v):
+    if isinstance(v, datetime):
+        # TIMESTAMP 用(UTCに正規化して Z を付ける)
+        if v.tzinfo is None:
+            v = v.replace(tzinfo=timezone.utc)
+        return v.astimezone(timezone.utc).isoformat().replace("+00:00", "Z")
+    if isinstance(v, date) and not isinstance(v, datetime):
+        # DATE 用
+        return v.isoformat()
+    if isinstance(v, time):
+        # TIME 用
+        # v.isoformat() は microseconds を含むことあり → そのままでOK
+        return v.isoformat()
+    if isinstance(v, Decimal):
+        # NUMERIC/BIGNUMERIC に文字列で渡すのが安全
+        return str(v)
+    if isinstance(v, bytes):
+        # BYTES は base64 文字列
+        return base64.b64encode(v).decode("ascii")
+    if isinstance(v, Enum):
+        return v.value
+    if isinstance(v, list):
+        return [to_bq_json(x) for x in v]
+    if isinstance(v, dict):
+        return {k: to_bq_json(val) for k, val in v.items()}
+    return v
+
+@dataclass
+class BigQueryBaseModel:
+    
+    table_id: ClassVar[str]      # 例: "datasetid.table_name"
+    provider: ClassVar[GoogleCloudBigQueryProvider | None] = None
+    key_field: ClassVar[Optional[str]] = None  # 例: "id" 主キーに相当するフィールド名
+    columns_select: ClassVar[str] = "*"  # 例: "col1, col2, col3"
+    default_order_by: ClassVar[Optional[str]] = None
+
+    @classmethod
+    def set_provider(cls, provider: GoogleCloudBigQueryProvider):
+        cls.provider = provider
+
+    @classmethod
+    def fetch_all(cls: Type[T]) -> List[T]:
+        """テーブル内の全レコードを取得する"""
+        if cls.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+        
+        rows = cls.provider.list_rows(table_id=cls.table_id)        
+        result: List[T] = []
+        for row in rows:
+            data = dict(row)
+            obj: T = cls(**data)  # type: ignore[arg-type]
+            result.append(obj)
+        return result
+
+    def create(self):
+        """レコードをテーブルに挿入する"""
+        if self.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+        
+        data = {f.name: getattr(self, f.name) for f in fields(self)}
+        data = to_bq_json(data)
+
+        logger.debug(f"Inserting data: {data}")
+        self.provider.insert_rows(
+            table_id=self.table_id,
+            rows=[data]
+        )
+    
+    @classmethod
+    def insert(cls, rows: List[T]):
+        """複数レコードをテーブルに挿入する"""
+        if cls.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+        
+        bq_rows = []
+        for row in rows:
+            data = {f.name: getattr(row, f.name) for f in fields(row)}
+            bq_rows.append(to_bq_json(data))
+
+        logger.info(f"Inserting data: {len(bq_rows)}")
+        logger.debug(f"Inserting data details: {bq_rows[:10]}...")
+        cls.provider.insert_rows(
+            table_id=cls.table_id,
+            rows=bq_rows
+        )
+    
+    @classmethod
+    def is_streaming_buffer(cls) -> bool:
+        """テーブルのストリームバッファ情報を取得する"""
+        if cls.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+        
+        buf_info = cls.provider.get_streaming_buffer_info(table_id=cls.table_id)    
+        logger.debug(f"Streaming buffer info: {buf_info}")
+        
+        return buf_info is not None
+
+    def save(self):
+        """レコードをテーブルに更新する"""
+        if self.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+    
+        if self.is_streaming_buffer():
+            raise RuntimeError("Cannot update records in streaming buffer.")
+
+        if self.key_field is None:
+            raise RuntimeError("key_field is not set.")
+        
+        key_value = getattr(self, self.key_field)
+        if key_value is None:
+            raise ValueError(f"Key field '{self.key_field}' value is None.")
+
+        data = {f.name: getattr(self, f.name) for f in fields(self)}
+        data = to_bq_json(data)
+
+        set_clauses = []
+        for k, v in data.items():
+            if k == self.key_field:
+                continue
+            if isinstance(v, str):
+                v_str = f"'{v.replace('\'', '\\\'')}'"
+            else:
+                v_str = str(v)
+            set_clauses.append(f"{k}={v_str}")
+        set_clause = ", ".join(set_clauses)
+
+        where_clause = f"{self.key_field}='{str(key_value).replace('\'', '\\\'')}'"
+
+        logger.debug(f"Updating data: SET {set_clause} WHERE {where_clause}")
+        self.provider.update_query(
+            table_id=self.table_id,
+            values=set_clause,
+            were_clause=where_clause
+        )
+
+    def destroy(self):   
+        """レコードをテーブルから削除する"""
+        if self.provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+    
+        if self.is_streaming_buffer():
+            raise RuntimeError("Cannot delete records in streaming buffer.")
+
+        if self.key_field is None:
+            raise RuntimeError("key_field is not set.")
+        
+        key_value = getattr(self, self.key_field)
+        if key_value is None:
+            raise ValueError(f"Key field '{self.key_field}' value is None.")
+
+        where_clause = f"{self.key_field}='{str(key_value).replace('\'', '\\\'')}'"
+
+        logger.debug(f"Deleting data: WHERE {where_clause}")
+        self.provider.delete_query(
+            table_id=self.table_id,
+            were_clause=where_clause
+        )
+
+    # -------------------------------------
+    # select
+    @classmethod
+    def select(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+        limit: Optional[int] = None,
+        offset: Optional[int] = None,
+    ):
+        params_dict = []
+        where_sql, params_dict = _build_where_clause_bq(where)
+        # where_sql = _build_where_clause_bq_without_prams(where)
+        order = order_by or cls.default_order_by
+        order_sql = f"ORDER BY {order}" if order else ""
+        limit_sql = f"LIMIT {int(limit)}" if limit is not None else ""
+        offset_sql = f"OFFSET {int(offset)}" if offset is not None else ""
+        sql = f"""
+            SELECT {cls.columns_select}
+            FROM `{cls.table_id}`
+            {where_sql}
+            {order_sql}
+            {limit_sql}
+            {offset_sql}
+        """
+        # logger.debug(f"Select Query: {sql} with params_dict: {params_dict}")
+        if provider is None:
+            provider = cls.provider
+        if provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+
+        params = []
+        for p in params_dict:
+            param = provider.addQueryParameter(
+                name=p["name"],
+                value=p["parameterValue"]["value"],
+                param_type=p["parameterType"]["type"]
+            )
+            params.append(param)
+        # パラメータバインド：GoogleCloudBigQueryProvider の生 client を使用
+        # query_job = provider.excute_query(sql, params=params)
+        query_job = provider.excute_query(sql, params=params)
+        return query_job
+            
+
+    @classmethod
+    def select_df(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+        limit: Optional[int] = None,
+        offset: Optional[int] = None,
+    ) -> pd.DataFrame:
+        """DataFrame で取得
+        
+        Notes:
+            Google Cloud BigQuery から直接 DataFrame を取得するには db-dtypes パッケージが必要
+        """
+        query_job = cls.select(
+            provider=provider,
+            where=where,
+            order_by=order_by,
+            limit=limit,
+            offset=offset,
+        )        
+        return query_job.to_dataframe()
+    
+    @classmethod
+    def list(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+        limit: Optional[int] = None,
+        offset: Optional[int] = None,
+    ) -> List[T]:
+        job = cls.select(provider=provider, where=where, order_by=order_by, limit=limit, offset=offset)
+        results = job.result()
+        instances: List[T] = []
+        for row in results:
+            data = dict(row)
+            instance = cls(**data)  # type: ignore
+            instances.append(instance)
+
+        # df = cls.select_df(provider=provider, where=where, order_by=order_by, limit=limit, offset=offset)
+        # if not is_dataclass(cls):
+        #     raise TypeError(f"{cls.__name__} is not a dataclass.")
+        # instances: List[T] = []
+        # for _, row in df.iterrows():
+        #     data = {f.name: row[f.name] for f in fields(cls) if f.name in row}
+        #     instance = cls(**data)  # type: ignore
+        #     instances.append(instance)
+        return instances
+
+    @classmethod
+    def first(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+    ) -> Optional[T]:
+        results = cls.list(provider=provider, where=where, order_by=order_by, limit=1)
+        return results[0] if results else None
+
+    @classmethod
+    def count(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+    ) -> int:
+        where_sql, params_dict = _build_where_clause_bq(where)
+        sql = f"""
+            SELECT COUNT(*) as cnt
+            FROM `{cls.table_id}`
+            {where_sql}
+        """
+        # パラメータバインド：GoogleCloudBigQueryProvider の生 client を使用
+        if provider is None:
+            provider = cls.provider
+        if provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+
+        params = []
+        for p in params_dict:
+            param = provider.addQueryParameter(
+                name=p["name"],
+                value=p["parameterValue"]["value"],
+                param_type=p["parameterType"]["type"]
+            )
+            params.append(param)
+
+        provider.excute_query(
+            sql,
+            params=params
+        )
+
+        query_job = provider.excute_query(sql, params=params)
+        results = query_job.result()
+        for row in results:
+            return row["cnt"]
+        return 0
+
+    @classmethod
+    def exists(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+    ) -> bool:
+        if provider is None:
+            provider = cls.provider
+        return cls.count(provider, where=where)  > 0
+
+    @classmethod
+    def delete(
+        cls: Type[T],
+        provider: GoogleCloudBigQueryProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+    ):
+        where_sql, params_dict = _build_where_clause_bq(where)
+        sql = f"""
+            DELETE FROM `{cls.table_id}`
+            {where_sql}
+        """
+        if provider is None:
+            provider = cls.provider
+        if provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+
+        params = []
+        for p in params_dict:
+            param = provider.addQueryParameter(
+                name=p["name"],
+                value=p["parameterValue"]["value"],
+                param_type=p["parameterType"]["type"]
+            )
+            params.append(param)
+
+        job = provider.excute_query(
+            sql,
+            params=params
+        )
+        job.result()  # ジョブの完了を待機
+
+    @classmethod
+    def update(
+        cls: Type[T],
+        values: dict[str, Any],
+        *,
+        where: Union[str, Dict[str, Any], Sequence[Filter]] ,
+        provider: GoogleCloudBigQueryProvider = None,        
+    ):
+        params_dicts = []
+        where_sql, params_dicts = _build_where_clause_bq(where)
+        set_clauses = []
+        for k, v in values.items():
+            param_name = f"set_{k}"
+            set_clauses.append(f"{k}=@{param_name}")
+            params_dicts.append({
+                "name": param_name,
+                "parameterType": {"type": "STRING"},
+                "parameterValue": {"value": str(v)}
+            })
+        
+        sql = f"""
+            UPDATE `{cls.table_id}`
+            SET {', '.join(set_clauses)}
+            {where_sql}
+        """
+        logger.debug(f"update Query: {sql} with params_dict: {params_dicts}")
+
+        if provider is None:
+            provider = cls.provider
+        if provider is None:
+            raise RuntimeError("GoogleCloudBigQueryProvider not set. Call .set_provider().")
+
+        params = []
+        for p in params_dicts:
+            param = provider.addQueryParameter(
+                name=p["name"],
+                value=p["parameterValue"]["value"],
+                param_type=p["parameterType"]["type"]
+            )
+            params.append(param)
+
+        job = provider.excute_query(
+            sql,
+            params=params
+        )
+        job.result()  # ジョブの完了を待機

src/models/duck_base_model.py

@@ -0,0 +1,184 @@
+import os
+from typing import Any, Dict, Optional,Type,TypeVar,Tuple,Sequence,Union,List,Iterable,ClassVar
+from dataclasses import dataclass, fields, is_dataclass
+
+import pandas as pd
+from providers.duck_db_provider import DuckDBProvider
+
+DATA_BASEPATH=os.getenv("DATA_BASEPATH","./data")
+from lib.custom_logger import get_logger
+logger = get_logger()
+
+T = TypeVar("T", bound="DuckBaseModel")
+
+Filter = Tuple[str, str, Any]
+
+def _build_where_clause(where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]]
+                    ) -> Tuple[str, List[Any]]:
+    if where is None:
+        return "", []
+    params: List[Any] = []
+
+    if isinstance(where, str):
+        return f"WHERE {where}", params
+
+    clauses: List[str] = []
+    if isinstance(where, dict):
+        for col, val in where.items():
+            if isinstance(val, (list, tuple, set)):
+                placeholders = ", ".join(["?"] * len(val))
+                clauses.append(f"{col} IN ({placeholders})")
+                params.extend(list(val))
+            elif val is None:
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} = ?")
+                params.append(val)
+    else:
+        # sequence of (col, op, value)
+        for col, op, val in where:
+            if (op.upper() == "IN") and isinstance(val, (list, tuple, set)):
+                placeholders = ", ".join(["?"] * len(val))
+                clauses.append(f"{col} IN ({placeholders})")
+                params.extend(list(val))
+            elif val is None and op in ("=", "=="):
+                clauses.append(f"{col} IS NULL")
+            else:
+                clauses.append(f"{col} {op} ?")
+                params.append(val)
+
+    if not clauses:
+        return "", []
+    return "WHERE " + " AND ".join(clauses), params
+
+
+@dataclass
+class DuckBaseModel:
+    file_glob: ClassVar[str]                        # 例: "./data_science/data/y=2025/news/news_*.csv"
+    provider: ClassVar[DuckDBProvider | None] = None
+    columns_select: ClassVar[str] = "*"             # 例: "published_parsed, title, link"
+    default_order_by: ClassVar[Optional[str]] = None
+
+    # read_csv_auto オプション(必要ならサブクラスで上書き）
+    hive_partitioning: ClassVar[bool] = True
+    union_by_name: ClassVar[bool] = True
+    header: ClassVar[bool] = True
+    ignore_errors: ClassVar[bool] = True
+
+
+    @classmethod
+    def set_provider(cls, provider: DuckDBProvider):
+        cls.provider = provider
+
+
+
+
+    # ---- FROM 句(CSV読込） ----
+    @classmethod
+    def _from_clause(cls,file_glob=None) -> str:
+        path =  file_glob or cls.file_glob
+        path = path.replace("'", "''")
+        path =f"{DATA_BASEPATH}{path}"
+
+        # path = cls.file_glob.replace("'", "''")
+        hp = 1 if cls.hive_partitioning else 0
+        un = 1 if cls.union_by_name else 0
+        hd = "TRUE" if cls.header else "FALSE"
+        ig = "TRUE" if cls.ignore_errors else "FALSE"
+        return (f"read_csv_auto('{path}', HEADER={hd}, IGNORE_ERRORS={ig}, "
+                f"hive_partitioning={hp}, union_by_name={un})")
+
+    @classmethod
+    def select_df(
+        cls: Type[T],
+        provider: DuckDBProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+        limit: Optional[int] = None,
+        offset: Optional[int] = None,
+        file_glob=None,
+    ) -> List[T]:
+        where_sql, params = _build_where_clause(where)
+        order = order_by or cls.default_order_by
+        order_sql = f"ORDER BY {order}" if order else ""
+        limit_sql = f"LIMIT {int(limit)}" if limit is not None else ""
+        offset_sql = f"OFFSET {int(offset)}" if offset is not None else ""
+        sql = f"""
+            SELECT {cls.columns_select}
+            FROM {cls._from_clause(file_glob=file_glob)}
+            {where_sql}
+            {order_sql}
+            {limit_sql}
+            {offset_sql}
+        """
+        # パラメータバインド：DuckDBProvider の生 connection を使用
+        if provider is None:
+            provider = cls.provider
+
+        cur = provider.con.execute(sql, params)
+        return cur.df()
+    
+    @classmethod
+    def list(
+        cls: Type[T],
+        provider: DuckDBProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+        limit: Optional[int] = None,
+        offset: Optional[int] = None,
+        file_glob=None,
+    ) -> List[T]:
+        df = cls.select_df(provider=provider, where=where, order_by=order_by, limit=limit, offset=offset,file_glob=file_glob)
+        if not is_dataclass(cls):
+            raise TypeError(f"{cls.__name__} is not a dataclass.")
+        instances: List[T] = []
+        for _, row in df.iterrows():
+            data = {f.name: row[f.name] for f in fields(cls) if f.name in row}
+            instance = cls(**data)  # type: ignore
+            instances.append(instance)
+        return instances
+
+    @classmethod
+    def first(
+        cls: Type[T],
+        provider: DuckDBProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+        order_by: Optional[str] = None,
+    ) -> Optional[T]:
+        results = cls.list(provider=provider, where=where, order_by=order_by, limit=1)
+        return results[0] if results else None
+
+    @classmethod
+    def count(
+        cls: Type[T],
+        provider: DuckDBProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+    ) -> int:
+        where_sql, params = _build_where_clause(where)
+        sql = f"""
+            SELECT COUNT(*) AS cnt
+            FROM {cls._from_clause()}
+            {where_sql}
+        """
+        # パラメータバインド：DuckDBProvider の生 connection を使用
+        if provider is None:
+            provider = cls.provider
+
+        cur = provider.con.execute(sql, params)
+        row = cur.fetchone()
+        return row["cnt"] if row else 0
+
+    @classmethod
+    def exists(
+        cls: Type[T],
+        provider: DuckDBProvider = None,
+        *,
+        where: Optional[Union[str, Dict[str, Any], Sequence[Filter]]] = None,
+    ) -> bool:
+        if provider is None:
+            provider = cls.provider
+        return cls.count(provider, where=where)  > 0

src/models/firestore_base_model.py

@@ -45,7 +45,7 @@ class FirestoreBaseModel:
         return cls._col().document(doc_id)
 
     def _as_mutable_dict(self) -> Dict[str, Any]:
-        """dataclass → dict（id と None は必要に応じて処理）"""
+        """dataclass → dict(id と None は必要に応じて処理)"""
         data = asdict(self)
         data.pop("_doc_id", None)
         data.pop("collection_name", None)
@@ -54,7 +54,7 @@ class FirestoreBaseModel:
 
     @classmethod
     def _from_dict(cls: Type[T], data: Dict[str, Any]) -> T:
-        """dict -> モデル(data内の'id'を拾う）"""
+        """dict -> モデル(data内の'id'を拾う)"""
         data = dict(data)
         _id = data.pop("id", None)
         obj: T = cls(**data)  # type: ignore[arg-type]
@@ -88,7 +88,7 @@ class FirestoreBaseModel:
         return self._doc_id 
 
     def update_fields(self, changes: Dict[str, Any], auto_timestamp: bool = True) -> None:
-        """部分更新（set(merge=True)の糖衣）"""
+        """部分更新(set(merge=True)の糖衣)"""
         if not self._doc_id:
             raise ValueError("Cannot update without id.")
         payload = dict(changes)
@@ -205,7 +205,7 @@ class FirestoreBaseModel:
     ) -> int:
         """
         条件でヒットしたドキュメントを一括削除。削除件数を返す。
-        ※ list_documents に依存（“複合インデックスなし”ポリシーはそちらの制約に従う）
+        ※ list_documents に依存(“複合インデックスなし”ポリシーはそちらの制約に従う)
         """
         from providers.firestore_provider import FireStoreProvider
         rows = FireStoreProvider.list_documents(

src/providers/duck_db_provider.py

@@ -0,0 +1,73 @@
+import duckdb
+
+
+class DuckDBProvider:
+
+    def __init__(self, db_path: str = ":memory:", read_only: bool = False):
+        self.con = self.connect(db_path, read_only)
+
+    def connect(self, db_path: str = ":memory:", read_only: bool = False):
+        return duckdb.connect(database=db_path, read_only=read_only)
+
+    def setup_gcs(self, access_key: str,secret_key: str):
+        """GCSのシークレットを設定する"""
+        if not self.con:
+            raise ValueError("DuckDB is not connected.")
+        self.con.sql(f"""
+            CREATE OR REPLACE SECRET gcs_creds (
+                TYPE gcs,
+                KEY_ID '{access_key}',
+                SECRET '{secret_key}'
+            );
+        """)
+
+
+    def close(self):
+        """接続を閉じる"""
+        if self.con:
+            self.con.close()
+
+    def query_df(self, sql: str):
+        """SQLクエリを実行してDataFrameで返す"""        
+        return self.con.execute(sql).df()            
+
+    def max_value(
+        self,
+        file_path: str,
+        column: str,
+        hive_partitioning: bool = True,
+        union_by_name: bool = True,
+    ) -> any:
+        """CSVファイルの指定列の最大値を取得する"""
+        query = f"""
+            SELECT MAX({column}) AS max_{column}
+            FROM read_csv_auto('{file_path}', 
+            hive_partitioning={1 if hive_partitioning else 0},
+            union_by_name={1 if union_by_name else 0}
+            )
+        """
+        result = self.con.execute(query).fetchone()[0]
+        return result
+    
+    def load_file(self, file_glob: str, table: str):
+        """CSVを読み込みテーブル化"""
+        sql = f"""
+        CREATE OR REPLACE TABLE {table} AS
+        SELECT *
+        FROM read_csv_auto('{file_glob}', HEADER=TRUE, IGNORE_ERRORS=TRUE)
+        """
+        self.con.execute(sql)
+
+    @staticmethod
+    def get_gs_csv_name(
+                        backet: str, 
+                        object_name: str,
+                        hive_partitioning: bool = True,
+                        union_by_name: bool = True,
+                        ) -> str:
+        
+        return f"""read_csv_auto('gs://{backet}/{object_name}',
+        hive_partitioning={1 if hive_partitioning else 0},
+        union_by_name={1 if union_by_name else 0}
+        )
+        """

src/providers/firestore_provider.py

@@ -78,12 +78,12 @@ class FireStoreProvider():
         start_after: Optional[Union[google.cloud.firestore.DocumentSnapshot, Dict[str, Any]]] = None,
     ) -> List[Dict[str, Any]]:
         """
-        コレクションのドキュメントを取得（フィルタ/並び替え/制限/ページング対応）
+        コレクションのドキュメントを取得(フィルタ/並び替え/制限/ページング対応）
 
         Args:
             collection_name: コレクション名
             filters: [(field, op, value), ...]
-            order_by: 並び順のフィールド名 or その配列（'-created_at' のように先頭'-'で降順）
+            order_by: 並び順のフィールド名 or その配列('-created_at' のように先頭'-'で降順）
             limit: 取得件数の上限
             start_after: ドキュメントスナップショット、または order_by で並べた最後の値の辞書
 
@@ -189,13 +189,13 @@ class FireStoreProvider():
         merge: bool = True,
     ) -> None:
         """
-        ドキュメントを更新。merge=True なら部分更新（推奨）。
+        ドキュメントを更新。merge=True なら部分更新(推奨）。
         """
         ref = cls.get_doc_ref(collection_name, doc_id)
         if merge:
             ref.set(data, merge=True)
         else:
-            # 全置換（存在しないと作成される）
+            # 全置換(存在しないと作成される）
             ref.set(data, merge=False)
         logger.info(f"Updated document: {collection_name}/{doc_id} (merge={merge})")
 

src/providers/google_cloud_bigquery_provider.py

@@ -0,0 +1,662 @@
+import os
+import time
+from datetime import datetime
+from typing import Optional, Dict,  Union,Sequence
+from enum import Enum
+
+from google.cloud import bigquery
+from google.cloud.bigquery.table import RowIterator, TableListItem
+from google.cloud.bigquery.dataset import DatasetListItem
+from google.cloud.bigquery.table import StreamingBuffer
+
+from google.oauth2 import service_account
+
+from lib.custom_logger import get_logger
+logger = get_logger()
+
+class BigQueryFieldType(str,Enum):
+    STRING = "STRING"
+    BYTES = "BYTES"
+    INTEGER = "INTEGER"
+    FLOAT = "FLOAT"
+    BOOLEAN = "BOOLEAN"
+    TIMESTAMP = "TIMESTAMP"
+    DATE = "DATE"
+    TIME = "TIME"
+    DATETIME = "DATETIME"
+    RECORD = "RECORD"
+    NUMERIC = "NUMERIC"
+    BIGNUMERIC = "BIGNUMERIC"
+    GEOGRAPHY = "GEOGRAPHY"
+    JSON = "JSON"
+
+class BogQueryFieldMode(str,Enum):
+    NULLABLE = "NULLABLE" # NULL許可
+    REQUIRED = "REQUIRED" # 必須
+    REPEATED = "REPEATED" # 配列
+
+
+class GoogleCloudBigQueryProvider:
+    """Cloud BigQuery操作プロバイダ
+
+    Notes:
+        - このクラスはGoogle Cloud BigQueryの基本的な操作を提供します。
+        - 認証にはサービスアカウントキーまたはApplication Default Credentials(ADC)を使用します。
+        - BigQuery Storage Read APIを使用するなら別のライブラリを使用してください(高速読み込み)
+        - 必要な権限
+            - データセット操作を行う場合はBigQuery データオーナーの権限が必要です。
+                - bigquery.dataViewer / bigquery.dataEditor / bigquery.readSessionUserなど部分でも可能
+            - JOB操作を行う場合はBigQuery ジョブユーザーの権限が必要です。
+            - フル機能を使う場合はBigQuery データオーナー及びBigQuery ジョブユーザーの権限が必要です。
+    """
+
+    def __init__(self, cred_path: Optional[str] = None, project: Optional[str] = None):
+        try:
+            if cred_path:
+                creds = service_account.Credentials.from_service_account_file(cred_path)
+                # プロジェクト未指定なら credentials から取得
+                effective_project = project or creds.project_id
+                self._client = bigquery.Client(
+                    project=effective_project, credentials=creds
+                )
+                logger.info(f"Bigquery client initialized with service account file. project={effective_project}")
+            elif os.getenv("GOOGLE_APPLICATION_CREDENTIALS_JSON"):
+                cred_json = os.getenv("GOOGLE_APPLICATION_CREDENTIALS_JSON")
+                creds = service_account.Credentials.from_service_account_info(cred_json)
+                effective_project = project or creds.project_id
+                self._client = bigquery.Client(
+                    project=effective_project, credentials=creds
+                )
+                logger.info("Bigquery client initialized with credentials from environment variable.")
+            else:
+                self._client = bigquery.Client(project=project)
+                logger.info("Bigquery client initialized with default credentials (ADC).")
+        except Exception as e:
+            logger.error(f"Bigquery initialization failed: {e}")
+            raise
+
+    def get_datasets(self)-> list[DatasetListItem]: 
+        """データセット一覧を取得する"""
+        try:
+            datasets = list(self._client.list_datasets())
+            return datasets
+        except Exception as e:
+            logger.error(f"Failed to list datasets: {e}")
+            raise
+            
+    def create_dataset(
+            self, dataset_id: str,
+            location: Optional[str] ="asia-northeast1") -> bigquery.Dataset:
+        """
+        データセットを作成する
+        
+        Args:
+            dataset_id (str): 作成するデータセットID
+            location (Optional[str]): データセットのロケーション
+        Returns:
+            bigquery.Dataset: 作成されたデータセットオブジェクト
+        Notes:
+            - 既に存在するデータセットIDを指定した場合、例外が発生します。
+            - 必要な権限
+                - bigquery.datasets.create
+            - 命名規則
+                - 英小文字 (a-z)、数字 (0-9)、およびアンダースコア (_) のみ使用可能
+        """
+        try:
+            dataset_ref = self._client.dataset(dataset_id)
+            dataset = bigquery.Dataset(dataset_ref)
+            if location:
+                dataset.location = location
+            dataset = self._client.create_dataset(dataset)
+            logger.info(f"Dataset {dataset_id} created.")
+            return dataset
+        except Exception as e:
+            logger.error(f"Failed to create dataset {dataset_id}: {e}")
+            raise
+
+    def is_exists_dataset(self, dataset_id: str) -> bool:
+        """データセットが存在するか確認する"""
+        try:
+            self._client.get_dataset(dataset_id)
+            return True
+        except Exception:
+            return False
+        
+    def delete_dataset(self, dataset_id: str, delete_contents: bool = False):
+        """
+        データセットを削除する
+        
+        Args:
+            dataset_id (str): 削除するデータセットID
+            delete_contents (bool): データセット内のテーブルも削除する場合はTrue
+        Notes:
+            - 必要な権限
+                - bigquery.datasets.delete
+            - delete_contents=Falseの場合、データセット内にテーブルが存在すると削除に失敗します。
+                - その場合はdelete_contents=Trueを指定してください。
+                - `google.api_core.exceptions.Conflict`が発生します。
+        """
+        try:
+            self._client.delete_dataset(dataset_id, delete_contents=delete_contents)
+            logger.info(f"Dataset {dataset_id} deleted.")
+        except Exception as e:
+            logger.error(f"Failed to delete dataset {dataset_id}: {e}")
+            raise
+
+
+    def full_table_id(self, dataset_id: str, table_name: str) -> str:
+        """テーブルIDを構築する"""
+        project = self._client.project
+        return f"{project}.{dataset_id}.{table_name}"
+
+    def is_exists_table(self,table_id: str) -> bool:
+        """テーブルが存在するか確認する"""
+        try:
+            self._client.get_table(table_id)
+            return True
+        except Exception:
+            return False
+
+    @staticmethod
+    def addSchemaField(
+            name:str,
+            field_type:Union[str, BigQueryFieldType],
+            field_mode:Union[str, BogQueryFieldMode]=BogQueryFieldMode.NULLABLE ,
+            default_value_expression:Optional[str]=None,
+            description:Optional[str]=None
+        )-> bigquery.SchemaField:
+        if isinstance(field_type, BigQueryFieldType):
+            field_type = field_type.value
+        elif isinstance(field_type, str):
+            field_type = field_type.upper()
+
+        if isinstance(field_mode, BogQueryFieldMode):
+            field_mode = field_mode.value
+        elif isinstance(field_mode, str):
+            field_mode = field_mode.upper()
+
+        return bigquery.SchemaField(
+            name=name, 
+            field_type=field_type, 
+            mode=field_mode , 
+            default_value_expression=default_value_expression,
+            description=description)
+
+    def create_table(
+            self,
+            table_id: str,
+            schema: list[bigquery.SchemaField]) -> bigquery.Table:
+        """
+        テーブルを作成する
+        
+        Args:
+            table_id (str): 作成するテーブルID(例: 'dataset.table')
+            schema (list[bigquery.SchemaField]): テーブルのスキーマ定義
+        Returns:
+            bigquery.Table: 作成されたテーブルオブジェクト
+        Notes:
+            - 既に存在するテーブルIDを指定した場合、例外が発生します。
+        """
+        try:
+            table = bigquery.Table(table_id, schema=schema)
+            table = self._client.create_table(table)
+            logger.info(f"Table {table_id} created.")
+            return table
+        except Exception as e:
+            logger.error(f"Failed to create table {table_id}: {e}")
+            raise
+
+
+    def delete_table(self, table_id: str, not_found_ok: bool = False):
+        """
+        テーブルを削除する
+        
+        Args:
+            table_id (str): 削除するテーブルID(例: 'dataset.table')
+            not_found_ok (bool): テーブルが存在しない場合でもエラーにしない場合はTrue
+        """
+        try:
+            self._client.delete_table(table_id, not_found_ok=not_found_ok)
+            logger.info(f"Table {table_id} deleted.")
+        except Exception as e:
+            logger.error(f"Failed to delete table {table_id}: {e}")
+            raise
+
+    def get_tables(self, dataset_id: str) -> list[TableListItem]:
+        """
+        データセット内のテーブル一覧を取得する
+        
+        Args:
+            dataset_id (str): 対象のデータセットID
+        Returns:
+            list[bigquery.TableListItem]: テーブル一覧
+        """
+        try:
+            tables = list(self._client.list_tables(dataset_id))
+            logger.info(f"Tables listed successfully from dataset {dataset_id}, count {len(tables)}.")
+            return tables
+        except Exception as e:
+            logger.error(f"Failed to list tables in dataset {dataset_id}: {e}")
+            raise
+    
+    def insert_rows(
+            self,
+            table_id: str,
+            rows: list[Dict[str, any]]) -> Sequence[dict]:
+        """
+        テーブルに行を挿入する
+        
+        Args:
+            table_id (str): 挿入先のテーブルID(例: 'dataset.table')
+            rows (list[Dict[str, any]]): 挿入する行データのリスト   
+        Returns:
+            Sequence[dict]: 挿入結果のレスポンスオブジェクト
+        Notes:
+            - rowsは辞書形式で指定します。キーがカラム名、値がカラムの値となります。
+                - 例: [{"column1": "value1", "column2": 123}, {"column1": "value2", "column2": 456}]
+        """
+        try:
+            errors = self._client.insert_rows_json(table_id, rows)
+            if errors:
+                logger.error(f"Errors occurred while inserting rows into {table_id}: {errors}")
+            else:
+                logger.info(f"Rows inserted successfully into {table_id}.")
+            return errors
+        except Exception as e:
+            logger.error(f"Failed to insert rows into table {table_id}: {e}")
+            raise
+
+    def list_rows(
+        self,
+        table_id: str,
+        selected_fields:Optional[list[str]] = None,
+        max_results:Optional[int] = None,
+        start_index:Optional[int] = None
+    ) -> RowIterator:
+        """
+        テーブルの行を一覧取得する
+        Args:
+            table_id (str): 対象のテーブルID(例: 'dataset.table')
+            selected_fields (Optional[list[str]]): 取得するカラム名のリスト。Noneの場合は全カラム取得
+            max_results (Optional[int]): 取得する行数の上限。Noneの場合は制限なし
+            start_index (Optional[int]): 取得開始インデックス。Noneの場合は先頭から取得
+        Returns:
+            RowIterator: 取得した行のイテレータ
+        """
+        try:
+            table = self._client.get_table(table_id)
+            _selected_fields:Optional[list[bigquery.SchemaField]] = None
+            if selected_fields:
+                _selected_fields = [field for field in table.schema if field.name in selected_fields]
+
+            rows = self._client.list_rows(
+                table,
+                selected_fields=_selected_fields,
+                max_results=max_results,
+                start_index=start_index
+            )
+            logger.info(f"Rows listed successfully from table {table_id}.")
+            return rows
+        except Exception as e:
+            logger.error(f"Failed to list rows from table {table_id}: {e}")
+            raise
+        
+    def list_rows_to_dataframe(
+        self,
+        table_id: str,
+        selected_fields:Optional[list[str]] = None,
+        max_results:Optional[int] = None,
+        start_index:Optional[int] = None
+    ):
+        try:
+            table = self._client.get_table(table_id)
+            df = self._client.list_rows(
+                table,
+                selected_fields=selected_fields,
+                max_results=max_results,
+                start_index=start_index
+            ).to_dataframe()
+            logger.info(f"Rows listed to DataFrame successfully from table {table_id}.")
+            return df       
+        except Exception as e:
+            logger.error(f"Failed to list rows to DataFrame from table {table_id}: {e}")
+            raise
+
+    # --------------------------------------------------------
+    def get_streaming_buffer_info(self,table_id: str):
+        """スリーミングバッファを取得する
+
+        Notes:
+            - スリーミングバッファは、ストリーミング挿入によってテーブルに追加されたデータが永続的なストレージに書き込まれる前に一時的に保存される場所です。
+            - スリーミングバッファの情報が存在しない場合、Noneを返します
+            - ストリーミングバッファ内のデータを直接削除・更新・クリアすることは不可能です。
+            - GCP側で完全に自動制御されており、手動でクリア/フラッシュ/削除する手段は 存在しません。        
+            - 通常:数秒〜数分 高頻度ストリーミング:〜30分 極端な負荷／多パーティション:最大約90分(GCP上限目安)
+        """
+        t = self._client.get_table(table_id)
+        # SDK属性 or 下位プロパティのどちらかに入っています
+        buf:StreamingBuffer = getattr(t, "streaming_buffer", None) or t._properties.get("streamingBuffer")
+        if not buf:
+            return None
+        # 例: {'estimatedRows': '123', 'estimatedBytes': '456789', 'oldestEntryTime': '1698234000000'}
+        logger.debug(f"Streaming Buffer Raw Info: {buf}")
+        return {
+            "estimated_rows": buf.estimated_rows,
+            "estimated_bytes": buf.estimated_bytes,
+            "oldest_entry_ms": buf.oldest_entry_time,
+            "raw": buf,
+        }
+
+    def excute_query(
+            self,
+            query: str,
+            params:list = None) -> bigquery.QueryJob:
+        """
+        クエリを実行する
+        
+        Args:
+            query (str): 実行するSQLクエリ
+        Returns:
+            bigquery.QueryJob: 実行結果のQueryJobオブジェクト
+        """
+        try:
+            if params and len(params) >0:
+                query_job = self._client.query(query, job_config=bigquery.QueryJobConfig(
+                    query_parameters=params
+                ))
+            else:
+                query_job = self._client.query(query)
+            logger.info("Query executed successfully.")
+            logger.debug(f"Query JOB: {query_job}")
+            return query_job
+        except Exception as e:
+            logger.error(f"Failed to execute query: {e}")
+            raise
+
+
+
+    @staticmethod
+    def addQueryParameter(
+            name:str,
+            value: any,
+            param_type:Union[str, BigQueryFieldType]=BigQueryFieldType.STRING
+        ) -> bigquery.ScalarQueryParameter:
+        """クエリパラメータを作成する"""
+        if isinstance(param_type, BigQueryFieldType):
+            param_type = param_type.value
+        elif isinstance(param_type, str):
+            param_type = param_type.upper()
+
+        return bigquery.ScalarQueryParameter(
+            name=name,
+            type_=param_type,
+            value=value
+        )
+
+    def select_query(
+        self,
+        table_id: str,
+        columns: list[str] = None,
+        were_clause: str = None,
+        params: list = None,
+        order_by: str = None,
+        limit: int = None
+    ):
+        """
+        指定したテーブルからデータを選択するクエリを実行する
+        
+        Args:
+            table_id (str): 対象のテーブルID(例: 'dataset.table')
+            columns (list[str]): 取得するカラム名のリスト。Noneの場合は全カラム取得
+            were_clause (str): WHERE句の条件式。Noneの場合は条件なし
+            order_by (str): 並び替えのカラム名。Noneの場合は並び替えなし
+            limit (int): 取得する行数の上限。Noneの場合は制限なし
+        """
+        try:
+            cols = ", ".join(columns) if columns else "*"
+            query = f"SELECT {cols} FROM `{table_id}`"
+            if were_clause:
+                query += f" WHERE {were_clause}"
+            if order_by:
+                query += f" ORDER BY {order_by}"
+            if limit:
+                query += f" LIMIT {limit}"
+            query_job = self.excute_query(query, params=params)            
+            logger.info("Select query executed successfully.")
+            
+            return query_job
+        except Exception as e:
+            logger.error(f"Failed to execute select query on table {table_id}: {e}")
+            raise
+    
+    def select_query_to_dataframe(
+        self,
+        table_id: str,
+        columns: list[str] = None,
+        were_clause: str = None,
+        params: list = None,
+        order_by: str = None,
+        limit: int = None
+    ):
+        try:
+            query_job= self.select_query(
+                table_id=table_id,
+                columns=columns,
+                were_clause=were_clause,
+                params=params,
+                order_by=order_by,
+                limit=limit
+            )
+            df = query_job.to_dataframe()
+            logger.info("Select query to DataFrame executed successfully.")
+            return df
+        except Exception as e:
+            logger.error(f"Failed to convert query result to DataFrame for table {table_id}: {e}")
+            raise
+
+    def select_query_to_dict(
+        self,
+        table_id: str,
+        columns: list[str] = None,
+        were_clause: str = None,
+        params: list = None,
+        order_by: str = None,
+        limit: int = None
+    ) -> list[Dict[str, any]]:
+        try:
+            query_job= self.select_query(
+                table_id=table_id,
+                columns=columns,
+                were_clause=were_clause,
+                params=params,
+                order_by=order_by,
+                limit=limit
+            )
+            results = query_job.result()
+            dicts = [dict(row) for row in results]
+            logger.info("Select query to dicts executed successfully.")
+            return dicts
+        except Exception as e:
+            logger.error(f"Failed to convert query result to dicts for table {table_id}: {e}")
+            raise
+
+    def isert_query(
+        self,
+        table_id: str,
+        values: dict[str, str],
+        params: list = None
+    ) -> bigquery.QueryJob:
+        """
+        挿入クエリ(INSERT)を実行する
+        """
+        try:
+            query = f"INSERT INTO `{table_id}` ({', '.join(values.keys())})"
+            query += f" VALUES ({', '.join([str(v) for v in values.values()])})"
+            query_job = self.excute_query(query, params=params)            
+            logger.info("Insert query executed successfully.")
+            return query_job
+        except Exception as e:
+            logger.error(f"Failed to execute insert query on table {table_id}: {e}")
+            raise
+
+    def update_query(
+        self,
+        table_id: str,
+        values: dict[str, str],
+        were_clause: str,
+        params: list = None
+    ) -> bigquery.QueryJob:
+        """
+        更新クエリ(UPDATE)を実行する
+
+        Args:
+            table_id (str): 対象のテーブルID(例: 'dataset.table')
+            values (dict[str, str]): 更新するカラムと値の辞書   
+            were_clause (str): WHERE句の条件式。全件の場合は"TRUE"を指定
+        """
+        try:
+            query = f"UPDATE `{table_id}`"
+            query += f" SET {', '.join([f'{k} = {v}' for k, v in values.items()])}"
+            if were_clause:
+                query += f" WHERE {were_clause}"                
+            logger.debug(f"Update Query: {query}")
+            query_job = self.excute_query(query, params=params)            
+            logger.info("Update query executed successfully.")
+            return query_job.result()
+        except Exception as e:
+            logger.error(f"Failed to execute update query on table {table_id}: {e}")
+            raise
+    
+    def delete_query(
+        self,
+        table_id: str,
+        were_clause:str,
+        params: list = None
+    ) -> bigquery.QueryJob:
+        """
+        削除クエリ(DELETE)を実行する
+        """
+        try:
+            query = f"DELETE FROM `{table_id}`"
+            if were_clause:
+                query += f" WHERE {were_clause}"                
+            query_job = self.excute_query(query, params=params)            
+            logger.info("Delete query executed successfully.")
+            return query_job.result()
+        except Exception as e:
+            logger.error(f"Failed to execute delete query on table {table_id}: {e}")
+            raise
+    
+    def upsert_query(
+        self,
+        table_id: str,
+        insert_values: dict[str, str],
+        update_values: dict[str, str],
+        key_columns: list[str],
+        params: list = None
+    ) -> bigquery.QueryJob:
+        """
+        UPSERTクエリ(MERGE)を実行する
+        """
+        try:
+            query = f"MERGE INTO `{table_id}` T"
+            query += f" USING (SELECT {', '.join([str(v) for v in insert_values.values()])}) S"
+            query += f" ON {' AND '.join([f'T.{col} = S.{col}' for col in key_columns])}"
+            query += " WHEN MATCHED THEN"
+            query += f" UPDATE SET {', '.join([f'T.{k} = {v}' for k, v in update_values.items()])}"
+            query += " WHEN NOT MATCHED THEN"
+            query += f" INSERT ({', '.join(insert_values.keys())})"
+            query += f" VALUES ({', '.join([str(v) for v in insert_values.values()])})"
+            query_job = self.excute_query(query, params=params)            
+            logger.info("Upsert query executed successfully.")
+            return query_job
+        except Exception as e:
+            logger.error(f"Failed to execute upsert query on table {table_id}: {e}")
+            raise
+
+    def import_csv(
+        self,
+        table_id: str,
+        csv_file_path: str,
+        skip_leading_rows: int = 1,
+        write_disposition: str = "WRITE_APPEND",
+        field_delimiter: str = ",",
+        autodetect: bool = True,
+        schema: list[bigquery.SchemaField] = None
+    ) -> bigquery.LoadJob:
+        """
+        CSVファイルからテーブルにデータを挿入する
+        
+        Args:
+            table_id (str): 挿入先のテーブルID(例: 'dataset.table')
+            csv_file_path (str): 挿入元のCSVファイルパス GCSも対応可能(gs://bucket/path/to/file.csv)
+            skip_leading_rows (int): ヘッダ行をスキップする行数
+            write_disposition (str): 書き込みモード (例: "WRITE_APPEND", "WRITE_TRUNCATE", "WRITE_EMPTY")
+            field_delimiter (str): フィールド区切り文字
+            autodetect (bool): スキーマ自動検出を有効にするかどうか
+            schema (list[bigquery.SchemaField]): スキーマ定義。autodetect=Falseの場合に必要
+        Returns:
+            bigquery.LoadJob: ロードジョブオブジェクト
+        """
+        try:
+            job_config = bigquery.LoadJobConfig(
+                skip_leading_rows=skip_leading_rows,
+                write_disposition=write_disposition,
+                field_delimiter=field_delimiter,
+                autodetect=autodetect,
+                schema=schema
+            )
+            with open(csv_file_path, "rb") as source_file:
+                load_job = self._client.load_table_from_file(
+                    source_file,
+                    table_id,
+                    job_config=job_config
+                )
+            load_job.result()  # ジョブの完了を待機
+            logger.info(f"Data loaded successfully from {csv_file_path} to table {table_id}.")
+            return load_job
+        except Exception as e:
+            logger.error(f"Failed to load data from {csv_file_path} to table {table_id}: {e}")
+            raise
+
+
+
+    def export_csv_to_gcs(
+        self,
+        table_id: str,
+        bucket: str,
+        prefix: str,
+        csv_file_name: str =None,
+        location: str = None
+    ) -> bigquery.ExtractJob:
+        """
+        テーブルからGCSのCSVファイルにデータをエクスポートする
+        
+        Args:
+            table_id (str): エクスポート元のテーブルID(例: 'dataset.table')
+            bucket (str): エクスポート先のGCSバケット名
+            prefix (str): エクスポート先のGCSプレフィックス(フォルダパス)
+            csv_file_name (str): エクスポート先のCSVファイル名。Noneの場合は自動生成
+            location (str): ジョブのロケーション
+        Returns:
+            bigquery.ExtractJob: エクストラクトジョブオブジェクト
+        """
+        try:            
+            if not csv_file_name:
+                _timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")                
+                csv_file_name = f"{table_id.replace('.','_')}_{_timestamp}.csv"
+            
+            destination_uri = f"gs://{bucket}/{prefix}/{csv_file_name}"
+            
+            extract_job = self._client.extract_table(
+                table_id,
+                destination_uri,
+                location=location
+            )
+            extract_job.result()  # ジョブの完了を待機
+            logger.info(f"Data exported successfully from table {table_id} to {destination_uri}.")
+            return extract_job
+        except Exception as e:
+            logger.error(f"Failed to export data from table {table_id} to {destination_uri}: {e}")
+            raise
+    

src/providers/google_cloud_storage_provider.py

@@ -10,6 +10,9 @@ from google.oauth2 import service_account
 from lib.custom_logger import get_logger
 logger = get_logger()
 
+import zipfile
+from pathlib import Path
+
 class GoogleCloudStorageProvider:
 
 
@@ -75,6 +78,17 @@ class GoogleCloudStorageProvider:
 
     def write_item(self, bucket: str, object_name: str, data: Union[bytes, BinaryIO, str],
                 content_type: str | None = None) -> Dict[str, Any]:
+        """
+        オブジェクトを書き込む
+        
+        Args:
+            bucket (str): バケット名
+            object_name (str): オブジェクト名
+            data (Union[bytes, BinaryIO, str]): 書き込むデータ
+            content_type (Optional[str]): コンテンツタイプ(MIMEタイプ）
+        Returns:
+            Dict[str, Any]: 書き込んだオブジェクトの情報
+        """
         blob = self._blob(bucket, object_name)
         if content_type is None:
             content_type = mimetypes.guess_type(object_name)[0] or "application/octet-stream"
@@ -100,4 +114,57 @@ class GoogleCloudStorageProvider:
 
     def generate_signed_url(self, bucket: str, object_name: str, method: str = "GET",
                             expires: timedelta = timedelta(hours=1)) -> str:
-        return self._blob(bucket, object_name).generate_signed_url(expiration=expires, method=method)
+        return self._blob(bucket, object_name).generate_signed_url(expiration=expires, method=method)
+
+    def zip_items(
+            self,
+            bucket: str,
+            object_names: List[str],
+        ) -> bytes:
+        """
+        複数のGCSオブジェクトを1つのZIPにまとめ、ZIPバイナリ(bytes)を返す
+
+        Args:
+            bucket (str): バケット名
+            object_names (List[str]): 対象オブジェクトのリスト
+        Returns:
+            bytes: ZIPファイルのバイナリ
+        """
+        out = io.BytesIO()
+        with zipfile.ZipFile(out, mode="w", compression=zipfile.ZIP_DEFLATED) as zf:
+            for obj in object_names:
+                blob = self._blob(bucket, obj)
+                if not blob.exists():
+                    raise FileNotFoundError(f"Object not found: gs://{bucket}/{obj}")
+
+                buf = io.BytesIO()
+                blob.download_to_file(buf)
+                buf.seek(0)
+                arcname = Path(obj).name
+                zf.writestr(arcname, buf.read())
+
+            zf.comment = f"bucket={bucket}, files={len(object_names)}".encode()
+
+        return out.getvalue()
+
+    def upload_folder(self, bucket: str, folder_path: str, gcs_prefix: str = ""):
+        """
+        ローカルフォルダをGCSに再帰的にアップロードする
+
+        Args:
+            bucket (str): バケット名
+            folder_path (str): ローカルフォルダのパス
+            gcs_prefix (str): GCS上のプレフィックス(フォルダパス)
+        """
+        _bucket = self._bucket(bucket)
+    
+        for root, _, files in os.walk(folder_path):
+            for file in files:
+                local_file_path = os.path.join(root, file)
+                # フォルダ構造を保つように相対パスを生成
+                relative_path = os.path.relpath(local_file_path, folder_path)
+                gcs_object_name = os.path.join(gcs_prefix, relative_path).replace("\\", "/")
+                
+                blob = _bucket.blob(gcs_object_name)
+                blob.upload_from_filename(local_file_path)
+                logger.info(f"Uploaded {local_file_path} to gs://{bucket}/{gcs_object_name}")

src/providers/one_drive_provider.py

@@ -31,8 +31,8 @@ class OneDriveProvider:
             client_id: Azure Portal のアプリ(公開クライアント)の Client ID
             authority: 'https://login.microsoftonline.com/{tenant}' (未指定は 'common')
             scopes: 例 ["Files.ReadWrite", "User.Read", "offline_access"]
-            token_cache_path: MSAL のシリアライズ済みトークンキャッシュ保存先（任意）
-            access_token: 既に取得済みの Bearer トークンを直接使いたい場合（任意）
+            token_cache_path: MSAL のシリアライズ済みトークンキャッシュ保存先(任意）
+            access_token: 既に取得済みの Bearer トークンを直接使いたい場合(任意）
         """
         self.client_id = client_id or os.getenv("MS_CLIENT_ID") or ""
         self.authority = authority or self.DEFAULT_AUTHORITY
@@ -64,7 +64,7 @@ class OneDriveProvider:
                 f.write(self._token_cache.serialize())
 
     def ensure_token(self):
-        """有効な Access Token を確保（キャッシュ→デバイスコード）。"""
+        """有効な Access Token を確保(キャッシュ→デバイスコード）。"""
         if self._access_token:
             return self._access_token
 
@@ -102,7 +102,7 @@ class OneDriveProvider:
         """
         Graph のパス表記: /me/drive/root:/foo/bar:/children のように使用。
         先頭に / を付けず、URL エンコードは requests 側に任せる前提で
-        空白や日本語は安全のため quote することを推奨（今回は簡易化）。
+        空白や日本語は安全のため quote することを推奨(今回は簡易化）。
         """
         if not path or path.strip() in ["/", "."]:
             return ""
@@ -156,7 +156,7 @@ class OneDriveProvider:
 
     def create_folder(self, folder_path: str) -> Dict[str, Any]:
         """
-        中間フォルダも順次作成（簡易実装）。
+        中間フォルダも順次作成(簡易実装）。
         """
         parts = [p for p in self._normalize_path(folder_path).split("/") if p]
         cur = ""
@@ -300,7 +300,7 @@ class OneDriveProvider:
         return r.json()
 
     # -----------------------
-    # 共有リンク（擬似 Signed URL）
+    # 共有リンク(擬似 Signed URL）
     # -----------------------
     def generate_share_link(
         self,
@@ -316,7 +316,7 @@ class OneDriveProvider:
         url = f"{self._item_by_path_url(path)}:/createLink"
         body: Dict[str, Any] = {"type": link_type, "scope": scope}
         if password:
-            body["password"] = password  # パスワード保護（ポリシーにより可否）
+            body["password"] = password  # パスワード保護(ポリシーにより可否）
         r: requests.Response = self._session.post(url, headers=self._headers(), json=body)
         r.raise_for_status()
         data:dict = r.json()

src/providers/rss_reader_client.py

@@ -0,0 +1,41 @@
+import feedparser
+from feedparser import FeedParserDict
+from pydantic import BaseModel
+
+class Feed(BaseModel):
+    url: str
+    title: str | None = None
+    company: str | None = None
+    description: str | None = None
+    language: str | None = None
+    tags: list[str] | None = None
+    
+
+class RSSItem(BaseModel):
+    title: str
+    link: str
+    guid: str | None = None 
+    published: str | None = None
+    ts: str | None = None
+    source: str | None = None
+    description: str | None = None
+
+
+class RSSReaderClient:
+    """RSSリーダークライアント"""
+    @classmethod
+    def fetch(cls, feed: Feed, from_date: str, to_date: str) -> list[RSSItem]:
+        """指定されたフィードから記事を取得する"""
+        items = []
+        d: FeedParserDict = feedparser.parse(feed.url)
+        for e in d.entries:
+            items.append(RSSItem(
+                title=e.get("title", "(no title)"),
+                link=e.get("link"),
+                guid=e.get("id") or e.get("guid") or e.get("link"),
+                published=e.get("published") or e.get("updated"),
+                ts=e.get("published") or e.get("updated"),
+                source=feed.url,
+                description=e.get("summary") or e.get("description"),
+            ))
+        return items