FinMind 2.0.8__tar.gz → 2.0.10__tar.gz

{finmind-2.0.8 → finmind-2.0.10}/FinMind/data/data_loader.py

@@ -1217,6 +1217,38 @@ class DataLoader(FinMindApi):
         )
         return futures_daily
 
+    def taiwan_futures_kbar(
+        self,
+        futures_id: str = "",
+        date: str = "",
+        timeout: int = None,
+    ) -> pd.DataFrame:
+        """get 期貨分K資料
+        :param futures_id: 期貨代號("TX")
+        :param date (str): 日期("2024-01-02")
+        :param timeout (int): timeout seconds, default None
+
+        :return: 期貨分K TaiwanFuturesKBar
+        :rtype pd.DataFrame
+        :rtype column date (str): 日期
+        :rtype column futures_id (str): 期貨代碼
+        :rtype column contract_date (str): 到期月份
+        :rtype column minute (str): 分鐘時間
+        :rtype column open (float): 開盤價
+        :rtype column high (float): 最高價
+        :rtype column low (float): 最低價
+        :rtype column close (float): 收盤價
+        :rtype column volume (int): 成交量
+        """
+        futures_kbar = self.get_data(
+            dataset=Dataset.TaiwanFuturesKBar,
+            data_id=futures_id,
+            start_date=date,
+            end_date=date,
+            timeout=timeout,
+        )
+        return futures_kbar
+
     def taiwan_option_daily(
         self,
         option_id: str = "",
@@ -2088,6 +2120,7 @@ class DataLoader(FinMindApi):
         use_object: bool = False,
     ) -> pd.DataFrame:
         """get 台股分 K 資料表
+        :param stock_id (str): 股票代碼, 加權指數請帶 "TAIEX"
         :param timeout (int): timeout seconds, default None
         :param use_object (bool): 是否透過 signed URL 下載整日 parquet 資料物件,
             設為 True 時忽略 stock_id, stock_id_list, use_async 參數, default False

{finmind-2.0.8 → finmind-2.0.10}/FinMind/data/finmind_api.py

@@ -7,7 +7,11 @@ import requests
 from loguru import logger
 
 from FinMind.schema.data import Dataset
-from FinMind.utility.request import async_request_get, request_get
+from FinMind.utility.request import (
+    async_request_get,
+    download_storage_object,
+    request_get,
+)
 
 logger.remove()
 logger.add(sys.stderr, level="INFO")
@@ -398,11 +402,19 @@ class FinMindApi:
         data = pd.DataFrame(self._extract_data(data))
         return data
 
-    def get_object(self, dataset: str, date: str, timeout: int) -> pd.DataFrame:
+    def get_object(
+        self,
+        dataset: str,
+        date: str,
+        timeout: int,
+        max_retry_times: int = 10,
+    ) -> pd.DataFrame:
         """透過 signed URL 下載整日資料物件 (parquet 格式)
+        傳輸中斷時會自動重試，並以 HTTP Range 從中斷處續傳
         :param dataset (str): 資料集名稱 (e.g. "TaiwanStockPriceTick")
         :param date (str): 資料日期 ("2025-01-06")
         :param timeout (int): request timeout 秒數
+        :param max_retry_times (int): 最多重試次數
 
         :return: 該日期所有資料
         :rtype: pd.DataFrame
@@ -412,16 +424,11 @@ class FinMindApi:
             "dataset": dataset,
             "date": date,
         }
-        response = request_get(
+        content = download_storage_object(
             self.__session,
             url,
             params=params,
             timeout=timeout,
+            max_retry_times=max_retry_times,
         )
-        status_code = response.status_code
-        if status_code == 200:
-            return pd.read_parquet(io.BytesIO(response.content))
-        else:
-            resp_json = response.json()
-            logger.info(resp_json)
-        return pd.DataFrame()
+        return pd.read_parquet(io.BytesIO(content))

{finmind-2.0.8 → finmind-2.0.10}/FinMind/schema/data.py

@@ -30,6 +30,7 @@ class Dataset(str, Enum):
     TaiwanOptionTick = "TaiwanOptionTick"
     TaiwanStockPriceBidAsk = "TaiwanStockPriceBidAsk"
     TaiwanFuturesDaily = "TaiwanFuturesDaily"
+    TaiwanFuturesKBar = "TaiwanFuturesKBar"
     TaiwanOptionDaily = "TaiwanOptionDaily"
     TaiwanStockNews = "TaiwanStockNews"
     TaiwanStockTotalReturnIndex = "TaiwanStockTotalReturnIndex"

{finmind-2.0.8 → finmind-2.0.10}/FinMind/strategies/base.py

@@ -116,11 +116,12 @@ class Trader:
         self.UnrealizedProfit = capital_gains - sell_fee - sell_tax
         self.EverytimeProfit = self.UnrealizedProfit + self.RealizedProfit
 
-    @staticmethod
     def __have_enough_money(
-        trader_fund: int, trade_price: float, trade_volume: float
+        self, trader_fund: int, trade_price: float, trade_volume: float
     ) -> bool:
-        return trader_fund >= (trade_price * trade_volume)
+        buy_price = trade_price * trade_volume
+        buy_fee = max(20.0, buy_price * self.fee)
+        return trader_fund >= (buy_price + buy_fee)
 
     @staticmethod
     def __have_enough_volume(hold_volume: float, trade_volume: float) -> bool:
@@ -266,7 +267,7 @@ class BackTest:
         indicators_info: typing.Union[IndicatorsInfo, typing.Dict[str, str]],
     ):
         indicators_info = (
-            indicators_info.dict()
+            indicators_info.model_dump()
             if isinstance(indicators_info, IndicatorsInfo)
             else indicators_info
         )
@@ -323,7 +324,7 @@ class BackTest:
         ],
     ):
         return [
-            rule.dict() if isinstance(rule, AddBuySellRule) else rule
+            rule.model_dump() if isinstance(rule, AddBuySellRule) else rule
             for rule in rule_list
         ]
 
@@ -584,11 +585,15 @@ class BackTest:
         gain_cash = (
             cash_div * trader.hold_volume + gain_stock_frac * 10
         )  # 將小數部分加進 gain_cash
+        # 配股「除權」不是一筆交易，總持有成本不會改變，
+        # 只是同一筆成本攤到更多股數上（每股成本被稀釋）。
+        # 因此 origin_cost 必須在 hold_volume 增加「之前」計算，
+        # 否則總成本會憑空跟著配股一起變大。
+        origin_cost = trader.hold_cost * trader.hold_volume
         trader.hold_volume += gain_stock_div
         # 在 UnrealizedProfit & RealizedProfit
         # 避免重複計算 gain_cash
-        origin_cost = trader.hold_cost * trader.hold_volume
-        # 持有成本不變，將配息歸類在，已實現損益
+        # 現金股利不調整持有成本，改歸類在已實現損益
         # new_cost = origin_cost - gain_cash
         trader.hold_cost = (
             origin_cost / trader.hold_volume if trader.hold_volume != 0 else 0
@@ -698,7 +703,7 @@ class BackTest:
     @property
     def final_stats(self) -> pd.Series:
         self._final_stats = pd.Series(
-            FinalStats(**self._final_stats.to_dict()).dict()
+            FinalStats(**self._final_stats.to_dict()).model_dump()
         )
         return self._final_stats
 
@@ -706,7 +711,7 @@ class BackTest:
     def trade_detail(self) -> pd.DataFrame:
         self._trade_detail = pd.DataFrame(
             [
-                TradeDetail(**row_dict).dict()
+                TradeDetail(**row_dict).model_dump()
                 for row_dict in self._trade_detail.to_dict("records")
             ]
         )
@@ -716,7 +721,7 @@ class BackTest:
     def compare_market_detail(self) -> pd.DataFrame:
         self._compare_market_detail = pd.DataFrame(
             [
-                CompareMarketDetail(**row_dict).dict()
+                CompareMarketDetail(**row_dict).model_dump()
                 for row_dict in self._compare_market_detail.to_dict("records")
             ]
         )
@@ -725,7 +730,9 @@ class BackTest:
     @property
     def compare_market_stats(self) -> pd.Series:
         self._compare_market_stats = pd.Series(
-            CompareMarketStats(**self._compare_market_stats.to_dict()).dict()
+            CompareMarketStats(
+                **self._compare_market_stats.to_dict()
+            ).model_dump()
         )
         return self._compare_market_stats
 

{finmind-2.0.8 → finmind-2.0.10}/FinMind/strategies/utils.py

@@ -35,9 +35,18 @@ def get_asset_underlying_type(stock_id: str, data_loader: DataLoader) -> str:
     return sorted(industry_category.values)[0]
 
 
+# ETF 的證交稅是 0.1%，一般股票是 0.3%。TaiwanStockInfo 對上櫃 ETF 的
+# industry_category 有「上櫃指數股票型基金(ETF)」與「上櫃ETF」兩種寫法，
+# 同一檔可能只出現其中一種，兩種都要對應到 ETF 稅率。
+UNDERLYING_TRADING_TAX = {
+    "ETF": 0.001,
+    "上櫃指數股票型基金(ETF)": 0.001,
+    "上櫃ETF": 0.001,
+}
+
+
 def get_underlying_trading_tax(underlying_type: str) -> float:
-    mapping = {"ETF": 0.001, "上櫃指數股票型基金(ETF)": 0.001}
-    return mapping.get(underlying_type, 0.003)
+    return UNDERLYING_TRADING_TAX.get(underlying_type, 0.003)
 
 
 def calculate_datenbr(day1: str, day2: str) -> int:

finmind-2.0.10/FinMind/utility/request.py

@@ -0,0 +1,277 @@
+import asyncio
+import concurrent.futures
+import hashlib
+import os
+import ssl
+import time
+from typing import Dict, List, Union
+
+import nest_asyncio
+import requests
+from loguru import logger
+from tqdm import tqdm
+
+nest_asyncio.apply()
+
+
+def request_get(
+    session: requests.Session,
+    url: str,
+    params: Dict[str, Union[int, str, float]] = None,
+    timeout: int = 60,
+    max_retry_times: int = 10,
+    verbose: bool = False,
+):
+    """
+    單次 request，支援 retry 與 log
+    """
+    response = None
+    for retry_times in range(1, max_retry_times + 1):
+        try:
+            response = session.get(
+                url,
+                verify=True,
+                params=params,
+                timeout=timeout,
+            )
+            if response.status_code == 504:
+                if verbose:
+                    logger.warning(
+                        f"status_code=504, retry {retry_times}/{max_retry_times}"
+                    )
+                time.sleep(retry_times * 0.1)
+            else:
+                break
+        except (
+            requests.ConnectionError,
+            ssl.SSLError,
+            requests.exceptions.ReadTimeout,
+            requests.exceptions.ChunkedEncodingError,
+        ) as exc:
+            if verbose:
+                logger.warning(
+                    f"Error: {exc}, retry {retry_times}/{max_retry_times}"
+                )
+            time.sleep(retry_times * 0.1)
+        except Exception as exc:
+            if verbose:
+                logger.warning(
+                    f"Unexpected error: {exc}, retry {retry_times}/{max_retry_times}"
+                )
+            time.sleep(retry_times * 0.1)
+    if response and response.status_code != 200:
+        raise Exception(
+            f"Final response status: {response.status_code}, text: {response.text}"
+        )
+    return response
+
+
+RETRYABLE_ERRORS = (
+    requests.ConnectionError,
+    ssl.SSLError,
+    requests.exceptions.ReadTimeout,
+    requests.exceptions.ChunkedEncodingError,
+)
+REDIRECT_STATUS = (301, 302, 303, 307, 308)
+
+
+def download_storage_object(
+    session: requests.Session,
+    url: str,
+    params: Dict[str, Union[int, str, float]] = None,
+    timeout: int = 60,
+    max_retry_times: int = 10,
+    verbose: bool = False,
+) -> bytes:
+    """
+    下載整日資料物件，傳輸中斷時以 Range + If-Match 續傳
+
+    storage_objects 回 307 導向有效期限約 3 分鐘的下載網址，
+    所以每次（含重試）都重新取得網址；已收到的內容保留，
+    只要求剩下的位元組，並以 If-Match 確認仍是同一版本，
+    版本不同（412）才從頭下載。完成後檢查長度，ETag 不含 "-" 時再比對 MD5。
+    """
+    buffer = bytearray()
+    etag = None
+    total = None
+    for retry_times in range(1, max_retry_times + 1):
+        try:
+            response = session.get(
+                url,
+                verify=True,
+                params=params,
+                timeout=timeout,
+                allow_redirects=False,
+            )
+        except RETRYABLE_ERRORS as exc:
+            if verbose:
+                logger.warning(
+                    f"Error: {exc}, retry {retry_times}/{max_retry_times}"
+                )
+            time.sleep(retry_times * 0.5)
+            continue
+        if response.status_code == 200:
+            return response.content
+        if response.status_code == 504:
+            time.sleep(retry_times * 0.5)
+            continue
+        if response.status_code not in REDIRECT_STATUS:
+            raise Exception(
+                f"Final response status: {response.status_code}, text: {response.text}"
+            )
+        object_url = response.headers["Location"]
+
+        headers = {}
+        if buffer and etag:
+            headers = {"Range": f"bytes={len(buffer)}-", "If-Match": etag}
+        try:
+            # 下載網址本身帶簽章，不能帶 session 的 Authorization header
+            with requests.get(
+                object_url, headers=headers, stream=True, timeout=timeout
+            ) as object_response:
+                status_code = object_response.status_code
+                if status_code == 200:
+                    buffer = bytearray()
+                    etag = object_response.headers.get("ETag")
+                    content_length = object_response.headers.get(
+                        "Content-Length"
+                    )
+                    total = int(content_length) if content_length else None
+                elif status_code == 412:
+                    # 物件在兩次請求之間已更新，已收到的內容作廢
+                    buffer, etag, total = bytearray(), None, None
+                    continue
+                elif status_code != 206:
+                    if verbose:
+                        logger.warning(
+                            f"status_code={status_code}, retry {retry_times}/{max_retry_times}"
+                        )
+                    time.sleep(retry_times * 0.5)
+                    continue
+                for chunk in object_response.iter_content(chunk_size=1 << 20):
+                    buffer.extend(chunk)
+        except RETRYABLE_ERRORS as exc:
+            if verbose:
+                logger.warning(
+                    f"Download interrupted at {len(buffer)} bytes: {exc}, "
+                    f"retry {retry_times}/{max_retry_times}"
+                )
+            time.sleep(retry_times * 0.5)
+            continue
+
+        if total is not None and len(buffer) < total:
+            continue
+        too_long = total is not None and len(buffer) > total
+        if too_long or not _match_etag(buffer, etag):
+            if verbose:
+                logger.warning(
+                    f"Downloaded object failed verification, retry {retry_times}/{max_retry_times}"
+                )
+            buffer, etag, total = bytearray(), None, None
+            continue
+        return bytes(buffer)
+    raise Exception(
+        f"Download storage object failed after {max_retry_times} retries: "
+        f"received {len(buffer)} of {total} bytes"
+    )
+
+
+def _match_etag(content: bytearray, etag: str) -> bool:
+    """ETag 不含 "-" 時等於整個物件的 MD5；分段上傳的 ETag 無法以 MD5 驗證"""
+    if not etag:
+        return True
+    etag = etag.strip('"')
+    if "-" in etag:
+        return True
+    return hashlib.md5(content).hexdigest() == etag
+
+
+async def _loop_run_get(
+    executor: concurrent.futures.ThreadPoolExecutor,
+    loop: asyncio.AbstractEventLoop,
+    session: requests.Session,
+    url: str,
+    params: Dict[str, Union[str, int, float]],
+    timeout: int = 60,
+    max_retry_times: int = 10,
+    verbose: bool = False,
+):
+    """
+    將同步 request 包成 async
+    """
+    return await loop.run_in_executor(
+        executor,
+        request_get,
+        session,
+        url,
+        params,
+        timeout,
+        max_retry_times,
+        verbose,
+    )
+
+
+def async_request_get(
+    session: requests.Session,
+    url: str,
+    params_list: List[Dict[str, Union[str, int, float]]],
+    timeout: int = 60,
+    max_retry_times: int = 10,
+    verbose: bool = False,
+    auto_tune: bool = True,
+    max_concurrency: int = None,
+    batch_size: int = 10,
+):
+    """
+    批量 async request，支援自動根據機器資源調整 max_concurrency / batch_size
+    """
+    # 自動調整
+    if auto_tune:
+        cpu_count = os.cpu_count() or 1
+        max_concurrency = max_concurrency or max(1, cpu_count * 2)
+    else:
+        max_concurrency = max_concurrency or 10
+
+    async def runner():
+        semaphore = asyncio.Semaphore(max_concurrency)
+        loop = asyncio.get_event_loop()
+        executor = concurrent.futures.ThreadPoolExecutor(
+            max_workers=max_concurrency
+        )
+
+        async def _limited_run(params):
+            async with semaphore:
+                resp = await _loop_run_get(
+                    executor,
+                    loop,
+                    session,
+                    url=url,
+                    params=params,
+                    timeout=timeout,
+                    max_retry_times=max_retry_times,
+                    verbose=verbose,
+                )
+                return resp
+
+        results = []
+        pbar = tqdm(total=len(params_list))
+
+        # 分 batch 建立 task
+        for i in range(0, len(params_list), batch_size):
+            batch = params_list[i : i + batch_size]
+            tasks = [asyncio.create_task(_limited_run(p)) for p in batch]
+
+            for coro in asyncio.as_completed(tasks):
+                try:
+                    res = await coro
+                    results.append(res)
+                except Exception as exc:
+                    if verbose:
+                        logger.error(f"Task failed: {exc}")
+                pbar.update(1)
+
+        pbar.close()
+        executor.shutdown()
+        return results
+
+    return asyncio.run(runner())

{finmind-2.0.8/finmind.egg-info → finmind-2.0.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: finmind
-Version: 2.0.8
+Version: 2.0.10
 Summary: financial mining
 Author-email: linsam <finmind.tw@gmail.com>
 License:                                  Apache License
@@ -245,8 +245,8 @@ Dynamic: license-file
 </p>
 
 <p align="center">
-  <a href="https://travis-ci.org/FinMind/FinMind"><img src="https://travis-ci.org/FinMind/FinMind.svg?branch=master" alt="Build Status"></a>
-  <a href="https://github.com/linsamtw/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
+  <a href="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml"><img src="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml/badge.svg?branch=master" alt="Build Status"></a>
+  <a href="https://github.com/FinMind/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
   <a href="https://finmind.github.io/"><img src="https://readthedocs.org/projects/finminddoc/badge/?version=latest" alt="Documentation Status"></a>
   <a href="https://gitter.im/FinMindTW/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge"><img src="https://badges.gitter.im/FinMindTW/community.svg" alt="Gitter"></a>
   <a href="https://badge.fury.io/py/FinMind"><img src="https://badge.fury.io/py/FinMind.svg" alt="PyPI version"></a>
@@ -285,7 +285,7 @@ pip install FinMind
 
 ## 這是什麼?
 
-**FinMind** 是超過 50 種金融開源數據 [50 datasets](https://finmind.github.io/)。
+**FinMind** 提供台灣與國際市場的金融開源數據，完整資料集清單請見[官方文件](https://finmind.github.io/)。
 包含
 
 * 技術面 : 台股股價 daily、即時報價、歷史 tick、PER、PBR、每5秒委託成交統計、加權指數、當日沖銷交易標的及成交量值。
@@ -299,8 +299,9 @@ pip install FinMind
 
 ## What is this?
 
-**FinMind** is open source of more
-than [50 datasets](https://finmind.github.io/), including
+**FinMind** provides open financial datasets for Taiwan and international
+markets. See the [official documentation](https://finmind.github.io/) for the
+current dataset list, including
 
 Taiwan stock trade data daily, Taiwan stock trade data (5 seconds) (2019-05-29 ~
 now, more than 30 million data in total), Financial Statements, Balance Sheet,
@@ -321,7 +322,7 @@ without having to collect the data by yourself.
 
 ## License
 
-- [License Detail](https://github.com/linsamtw/FinMind/blob/master/LICENSE)
+- [License Detail](https://github.com/FinMind/FinMind/blob/master/LICENSE)
 
 - 本專案提供的所有內容均用於教育、非商業用途。資料僅供參考，使用者依本資料交易發生交易損失需自行負責，本專案不對資料內容錯誤、更新延誤或傳輸中斷負任何責任。
 
@@ -390,6 +391,11 @@ Email: FinMind.TW@gmail.com
 
 每週日早上零點至早上七點為維護時間，不提供服務。
 
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for the development setup, test commands,
+and pull request checklist.
+
 ## Note
 
 + 未來預計新增更多功能，包含個人化回測分析、chatbot 監控策略，

{finmind-2.0.8 → finmind-2.0.10}/README.md

@@ -3,8 +3,8 @@
 </p>
 
 <p align="center">
-  <a href="https://travis-ci.org/FinMind/FinMind"><img src="https://travis-ci.org/FinMind/FinMind.svg?branch=master" alt="Build Status"></a>
-  <a href="https://github.com/linsamtw/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
+  <a href="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml"><img src="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml/badge.svg?branch=master" alt="Build Status"></a>
+  <a href="https://github.com/FinMind/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
   <a href="https://finmind.github.io/"><img src="https://readthedocs.org/projects/finminddoc/badge/?version=latest" alt="Documentation Status"></a>
   <a href="https://gitter.im/FinMindTW/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge"><img src="https://badges.gitter.im/FinMindTW/community.svg" alt="Gitter"></a>
   <a href="https://badge.fury.io/py/FinMind"><img src="https://badge.fury.io/py/FinMind.svg" alt="PyPI version"></a>
@@ -43,7 +43,7 @@ pip install FinMind
 
 ## 這是什麼?
 
-**FinMind** 是超過 50 種金融開源數據 [50 datasets](https://finmind.github.io/)。
+**FinMind** 提供台灣與國際市場的金融開源數據，完整資料集清單請見[官方文件](https://finmind.github.io/)。
 包含
 
 * 技術面 : 台股股價 daily、即時報價、歷史 tick、PER、PBR、每5秒委託成交統計、加權指數、當日沖銷交易標的及成交量值。
@@ -57,8 +57,9 @@ pip install FinMind
 
 ## What is this?
 
-**FinMind** is open source of more
-than [50 datasets](https://finmind.github.io/), including
+**FinMind** provides open financial datasets for Taiwan and international
+markets. See the [official documentation](https://finmind.github.io/) for the
+current dataset list, including
 
 Taiwan stock trade data daily, Taiwan stock trade data (5 seconds) (2019-05-29 ~
 now, more than 30 million data in total), Financial Statements, Balance Sheet,
@@ -79,7 +80,7 @@ without having to collect the data by yourself.
 
 ## License
 
-- [License Detail](https://github.com/linsamtw/FinMind/blob/master/LICENSE)
+- [License Detail](https://github.com/FinMind/FinMind/blob/master/LICENSE)
 
 - 本專案提供的所有內容均用於教育、非商業用途。資料僅供參考，使用者依本資料交易發生交易損失需自行負責，本專案不對資料內容錯誤、更新延誤或傳輸中斷負任何責任。
 
@@ -148,6 +149,11 @@ Email: FinMind.TW@gmail.com
 
 每週日早上零點至早上七點為維護時間，不提供服務。
 
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for the development setup, test commands,
+and pull request checklist.
+
 ## Note
 
 + 未來預計新增更多功能，包含個人化回測分析、chatbot 監控策略，

{finmind-2.0.8 → finmind-2.0.10/finmind.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: finmind
-Version: 2.0.8
+Version: 2.0.10
 Summary: financial mining
 Author-email: linsam <finmind.tw@gmail.com>
 License:                                  Apache License
@@ -245,8 +245,8 @@ Dynamic: license-file
 </p>
 
 <p align="center">
-  <a href="https://travis-ci.org/FinMind/FinMind"><img src="https://travis-ci.org/FinMind/FinMind.svg?branch=master" alt="Build Status"></a>
-  <a href="https://github.com/linsamtw/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
+  <a href="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml"><img src="https://github.com/FinMind/FinMind/actions/workflows/python-package.yml/badge.svg?branch=master" alt="Build Status"></a>
+  <a href="https://github.com/FinMind/FinMind/blob/master/LICENSE"><img src="https://img.shields.io/github/license/FinMind/FinMind" alt="license"></a>
   <a href="https://finmind.github.io/"><img src="https://readthedocs.org/projects/finminddoc/badge/?version=latest" alt="Documentation Status"></a>
   <a href="https://gitter.im/FinMindTW/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge"><img src="https://badges.gitter.im/FinMindTW/community.svg" alt="Gitter"></a>
   <a href="https://badge.fury.io/py/FinMind"><img src="https://badge.fury.io/py/FinMind.svg" alt="PyPI version"></a>
@@ -285,7 +285,7 @@ pip install FinMind
 
 ## 這是什麼?
 
-**FinMind** 是超過 50 種金融開源數據 [50 datasets](https://finmind.github.io/)。
+**FinMind** 提供台灣與國際市場的金融開源數據，完整資料集清單請見[官方文件](https://finmind.github.io/)。
 包含
 
 * 技術面 : 台股股價 daily、即時報價、歷史 tick、PER、PBR、每5秒委託成交統計、加權指數、當日沖銷交易標的及成交量值。
@@ -299,8 +299,9 @@ pip install FinMind
 
 ## What is this?
 
-**FinMind** is open source of more
-than [50 datasets](https://finmind.github.io/), including
+**FinMind** provides open financial datasets for Taiwan and international
+markets. See the [official documentation](https://finmind.github.io/) for the
+current dataset list, including
 
 Taiwan stock trade data daily, Taiwan stock trade data (5 seconds) (2019-05-29 ~
 now, more than 30 million data in total), Financial Statements, Balance Sheet,
@@ -321,7 +322,7 @@ without having to collect the data by yourself.
 
 ## License
 
-- [License Detail](https://github.com/linsamtw/FinMind/blob/master/LICENSE)
+- [License Detail](https://github.com/FinMind/FinMind/blob/master/LICENSE)
 
 - 本專案提供的所有內容均用於教育、非商業用途。資料僅供參考，使用者依本資料交易發生交易損失需自行負責，本專案不對資料內容錯誤、更新延誤或傳輸中斷負任何責任。
 
@@ -390,6 +391,11 @@ Email: FinMind.TW@gmail.com
 
 每週日早上零點至早上七點為維護時間，不提供服務。
 
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for the development setup, test commands,
+and pull request checklist.
+
 ## Note
 
 + 未來預計新增更多功能，包含個人化回測分析、chatbot 監控策略，

finmind-2.0.8/FinMind/utility/request.py

@@ -1,156 +0,0 @@
-import asyncio
-import concurrent.futures
-import os
-import ssl
-import time
-from typing import Dict, List, Union
-
-import nest_asyncio
-import requests
-from loguru import logger
-from tqdm import tqdm
-
-nest_asyncio.apply()
-
-
-def request_get(
-    session: requests.Session,
-    url: str,
-    params: Dict[str, Union[int, str, float]] = None,
-    timeout: int = 60,
-    max_retry_times: int = 10,
-    verbose: bool = False,
-):
-    """
-    單次 request，支援 retry 與 log
-    """
-    response = None
-    for retry_times in range(1, max_retry_times + 1):
-        try:
-            response = session.get(
-                url,
-                verify=True,
-                params=params,
-                timeout=timeout,
-            )
-            if response.status_code == 504:
-                if verbose:
-                    logger.warning(
-                        f"status_code=504, retry {retry_times}/{max_retry_times}"
-                    )
-                time.sleep(retry_times * 0.1)
-            else:
-                break
-        except (
-            requests.ConnectionError,
-            ssl.SSLError,
-            requests.exceptions.ReadTimeout,
-            requests.exceptions.ChunkedEncodingError,
-        ) as exc:
-            if verbose:
-                logger.warning(
-                    f"Error: {exc}, retry {retry_times}/{max_retry_times}"
-                )
-            time.sleep(retry_times * 0.1)
-        except Exception as exc:
-            if verbose:
-                logger.warning(
-                    f"Unexpected error: {exc}, retry {retry_times}/{max_retry_times}"
-                )
-            time.sleep(retry_times * 0.1)
-    if response and response.status_code != 200:
-        raise Exception(
-            f"Final response status: {response.status_code}, text: {response.text}"
-        )
-    return response
-
-
-async def _loop_run_get(
-    executor: concurrent.futures.ThreadPoolExecutor,
-    loop: asyncio.AbstractEventLoop,
-    session: requests.Session,
-    url: str,
-    params: Dict[str, Union[str, int, float]],
-    timeout: int = 60,
-    max_retry_times: int = 10,
-    verbose: bool = False,
-):
-    """
-    將同步 request 包成 async
-    """
-    return await loop.run_in_executor(
-        executor,
-        request_get,
-        session,
-        url,
-        params,
-        timeout,
-        max_retry_times,
-        verbose,
-    )
-
-
-def async_request_get(
-    session: requests.Session,
-    url: str,
-    params_list: List[Dict[str, Union[str, int, float]]],
-    timeout: int = 60,
-    max_retry_times: int = 10,
-    verbose: bool = False,
-    auto_tune: bool = True,
-    max_concurrency: int = None,
-    batch_size: int = 10,
-):
-    """
-    批量 async request，支援自動根據機器資源調整 max_concurrency / batch_size
-    """
-    # 自動調整
-    if auto_tune:
-        cpu_count = os.cpu_count() or 1
-        max_concurrency = max_concurrency or max(1, cpu_count * 2)
-    else:
-        max_concurrency = max_concurrency or 10
-
-    async def runner():
-        semaphore = asyncio.Semaphore(max_concurrency)
-        loop = asyncio.get_event_loop()
-        executor = concurrent.futures.ThreadPoolExecutor(
-            max_workers=max_concurrency
-        )
-
-        async def _limited_run(params):
-            async with semaphore:
-                resp = await _loop_run_get(
-                    executor,
-                    loop,
-                    session,
-                    url=url,
-                    params=params,
-                    timeout=timeout,
-                    max_retry_times=max_retry_times,
-                    verbose=verbose,
-                )
-                return resp
-
-        results = []
-        pbar = tqdm(total=len(params_list))
-
-        # 分 batch 建立 task
-        for i in range(0, len(params_list), batch_size):
-            batch = params_list[i : i + batch_size]
-            tasks = [asyncio.create_task(_limited_run(p)) for p in batch]
-
-            for coro in asyncio.as_completed(tasks):
-                try:
-                    res = await coro
-                    results.append(res)
-                except Exception as exc:
-                    if verbose:
-                        logger.error(f"Task failed: {exc}")
-                pbar.update(1)
-
-        pbar.close()
-        executor.shutdown()
-        return results
-
-    return asyncio.run(runner())