neurostats-API 0.0.13__py3-none-any.whl → 0.0.15__py3-none-any.whl

neurostats_API/__init__.py

@@ -1 +1,13 @@
-__version__='0.0.13'
+__version__='0.0.15'
+
+from .fetchers import (
+    BalanceSheetFetcher,
+    CashFlowFetcher,
+    FinanceOverviewFetcher,
+    FinanceReportFetcher,
+    InstitutionFetcher,
+    MarginTradingFetcher,
+    MonthRevenueFetcher,
+    TechFetcher,
+    ProfitLoseFetcher
+)

neurostats_API/fetchers/__init__.py

@@ -2,6 +2,8 @@ from .base import StatsDateTime, StatsFetcher
 from .balance_sheet import BalanceSheetFetcher
 from .cash_flow import CashFlowFetcher
 from .finance_overview import FinanceOverviewFetcher
+from .tej_finance_report import FinanceReportFetcher
+from .tech import TechFetcher
 from .institution import InstitutionFetcher
 from .margin_trading import MarginTradingFetcher
 from .month_revenue import MonthRevenueFetcher

neurostats_API/fetchers/base.py

@@ -1,16 +1,16 @@
+import abc
 from pymongo import MongoClient
 import pandas as pd
 import json
 import pytz
 from datetime import datetime, timedelta, date
-from ..utils import StatsDateTime, StatsProcessor
+from ..utils import StatsDateTime, StatsProcessor, YoY_Calculator
 import yaml
 
 class StatsFetcher:
     def __init__(self, ticker, db_client):
         self.ticker = ticker
-        self.db = db_client[
-            "company"]  # Replace with your database name
+        self.db = db_client["company"]  # Replace with your database name
         self.collection = self.db["twse_stats"]
 
         self.timezone = pytz.timezone("Asia/Taipei")
@@ -37,7 +37,7 @@ class StatsFetcher:
         ]
 
     def collect_data(self, start_date, end_date):
-        pipeline = self.prepare_query(start_date, end_date)
+        pipeline = self.prepare_query()
 
         fetched_data = list(self.collection.aggregate(pipeline))
 
@@ -52,3 +52,101 @@ class StatsFetcher:
         season = (month - 1) // 3 + 1
 
         return StatsDateTime(date, year, month, day, season)
+
+class BaseTEJFetcher(abc.ABC):
+    def __init__(self):
+        self.client = None
+        self.db = None
+        self.collection = None
+
+    @abc.abstractmethod
+    def get(self):
+        pass
+    
+    def get_latest_data_time(self, ticker):
+        latest_data = self.collection.find_one(
+            {"ticker": ticker},
+            {"last_update": 1, "_id" : 0}
+        )
+
+        try:
+            latest_date = latest_data['last_update']["latest_data_date"]
+        except Exception as e:
+            latest_date = None
+        
+        return latest_date
+
+    def cal_YoY(self, data_dict: dict, start_year: int, end_year: int):
+        year_shifts = [1,3,5,10]
+        return_dict = {}
+        for year in range(start_year, end_year+1):
+            year_data = data_dict[str(year)]
+            year_keys = list(year_data.keys())
+            for key in year_keys:
+                if (key in 'season'):
+                    continue
+
+                if (isinstance(year_data[key], (int, float))):
+                    temp_dict = {"value": year_data[key]}
+
+                    for shift in year_shifts:
+                        this_value = year_data[key]
+                        try:
+                            past_year = str(year - shift)
+                            last_value = data_dict[past_year][key]['value']
+                            temp_dict[f"YoY_{shift}"] = YoY_Calculator.cal_growth(
+                                this_value, last_value, delta = shift
+                            )
+                        except Exception as e:
+                            temp_dict[f"YoY_{shift}"] = None
+                        
+                    year_data[key] = temp_dict
+
+                else:
+                    year_data.pop(key)
+            
+            return_dict[year] = year_data
+        
+        
+        return return_dict
+    
+    def cal_QoQ(self, data_dict):
+        return_dict = {}
+        for i, time_index in enumerate(data_dict.keys()):
+            year, season = time_index.split("Q")
+            year = int(year)
+            season = int(season)
+            if (season == 1):
+                last_year = year - 1
+                last_season = 4
+            else:
+                last_year = year
+                last_season = season - 1
+            
+            this_data = data_dict[time_index]
+            this_keys = list(this_data.keys())
+            for key in this_keys:
+                if (key in 'season'):
+                    continue
+
+                this_value = this_data[key]
+
+                if (isinstance(this_value, (int, float))):
+                    temp_dict = {"value": this_value}
+
+                    try:
+                        last_value = data_dict[f"{last_year}Q{last_season}"][key]['value']
+
+                        temp_dict['growth'] = YoY_Calculator.cal_growth(
+                            this_value, last_value, delta=1
+                        )
+                    except Exception as e:
+                        temp_dict['growth'] = None
+                        
+                    this_data[key] = temp_dict
+
+                else:
+                    this_data.pop(key)
+            return_dict[time_index] = this_data
+        return return_dict
+            

neurostats_API/fetchers/institution.py

@@ -164,25 +164,26 @@ class InstitutionFetcher(StatsFetcher):
                                         latest_daily_data['volume'])
         }
         # 一年內法人
-        annual_dates = [data['date'].strftime("%Y-%m-%d") for data in daily_datas]
+        annual_dates = [
+            data['date'].strftime("%Y-%m-%d") for data in daily_datas
+        ]
         annual_closes = {
-            data['date'].strftime("%Y-%m-%d") : data['close']
+            data['date'].strftime("%Y-%m-%d"): data['close']
             for data in daily_datas
             if (data['date'].strftime("%Y-%m-%d") in annual_dates)
         }
         annual_volumes = {
-            data['date'].strftime("%Y-%m-%d") : data['volume']
+            data['date'].strftime("%Y-%m-%d"): data['volume']
             for data in daily_datas
             if (data['date'].strftime("%Y-%m-%d") in annual_dates)
         }
         annual_trading = {
-            data['date'].strftime("%Y-%m-%d") : data
+            data['date'].strftime("%Y-%m-%d"): data
             for data in institution_tradings
-        } 
+        }
 
         annual_trading = {
             date: {
-                
                 "close": annual_closes[date],
                 "volume": annual_volumes[date],
                 **annual_trading[date]
@@ -217,13 +218,18 @@ class InstitutionFetcher(StatsFetcher):
                 self.target_institution(latest_trading,
                                         latest_table['institutional_investor'],
                                         key, volume)
-
-            for trade_type in ['buy', 'sell']:
-                for unit in ['stock', 'percentage']:
-                    latest_table['institutional_investor'][trade_type][
-                        unit] = (latest_table['foreign'][trade_type][unit] +
-                                 latest_table['prop'][trade_type][unit] +
-                                 latest_table['mutual'][trade_type][unit])
+        # 計算合計
+        for unit in ['stock', 'percentage']:
+            # 買進總和
+            latest_table['institutional_investor']['buy'][unit] = (
+                latest_table['foreign']['buy'][unit] +
+                latest_table['prop']['buy'][unit] +
+                latest_table['mutual']['buy'][unit])
+            # 賣出總和
+            latest_table['institutional_investor']['sell'][unit] = (
+                latest_table['foreign']['sell'][unit] +
+                latest_table['prop']['sell'][unit] +
+                latest_table['mutual']['sell'][unit])
 
         frames = []
         for category, trades in latest_table.items():

neurostats_API/fetchers/profit_lose.py

@@ -21,12 +21,6 @@ class ProfitLoseFetcher(StatsFetcher):
     def prepare_query(self, target_season):
         pipeline = super().prepare_query()
 
-        target_query = {
-            "year": "$$target_season_data.year",
-            "season": "$$target_season_data.season",
-            "balance_sheet": "$$$$target_season_data.balance_sheet"
-        }
-
         pipeline.append({
             "$project": {
                 "_id": 0,
@@ -121,6 +115,9 @@ class ProfitLoseFetcher(StatsFetcher):
                     elif ('YoY' in item_name):
                         if (isinstance(item, (float, int))):
                             item = StatsProcessor.cal_percentage(item)
+                    elif ('每股盈餘' in index_name):
+                        if (isinstance(item, (float, int))):
+                            item = StatsProcessor.cal_non_percentage(item,  postfix="元")
                     else:
                         if (isinstance(item, (float, int))):
                             item = StatsProcessor.cal_non_percentage(item,  postfix="千元")
@@ -153,6 +150,9 @@ class ProfitLoseFetcher(StatsFetcher):
                         target_index=target_index)
                     break
                 except Exception as e:
-                    continue
+                    return_dict[name] = StatsProcessor.slice_multi_col_table(
+                        total_table=total_table,
+                        mode=setting['mode'],
+                        target_index=target_index)
 
         return return_dict

neurostats_API/fetchers/tech.py

@@ -1,9 +1,16 @@
 from .base import StatsFetcher
 import pandas as pd
+import yfinance as yf
 
 class TechFetcher(StatsFetcher):
     
     def __init__(self, ticker:str, db_client):
+
+        """
+        The Capitalization-Weighted Index includes the following tickers:
+        ['GSPC', 'IXIC', 'DJI', 'TWII']   
+        """
+        
         super().__init__(ticker, db_client)
         self.full_ohlcv = self._get_ohlcv()
         self.basic_indexes = ['SMA5', 'SMA20', 'SMA60', 'EMA5', 'EMA20', 
@@ -40,16 +47,36 @@ class TechFetcher(StatsFetcher):
         )
 
     def _get_ohlcv(self):
-        query = {'ticker': self.ticker}
-        ticker_full = list(self.collection.find(query))
+        
+        if self.ticker in ['GSPC', 'IXIC', 'DJI', 'TWII']:
+            
+            full_tick = f'^{self.ticker}'
+            yf_ticker = yf.Ticker(full_tick)
+            origin_df = yf_ticker.history(period="10y")
+            origin_df = origin_df.reset_index()
+            origin_df["Date"] = pd.to_datetime(origin_df["Date"]).dt.date
+            df = origin_df.rename(
+                columns={
+                    "Date": "date",
+                    "Open": "open",
+                    "High": "high",
+                    "Low": "low",
+                    "Close": "close",
+                    "Volume": "volume"
+                }
+            )
+        else:
+
+            query = {'ticker': self.ticker}
+            ticker_full = list(self.collection.find(query))
 
-        if not ticker_full:
-            raise ValueError(f"No data found for ticker: {self.ticker}")
+            if not ticker_full:
+                raise ValueError(f"No data found for ticker: {self.ticker}")
 
-        if 'daily_data' not in ticker_full[0] or ticker_full[0]['daily_data'] is None:
-            raise KeyError("Missing 'daily_data' in the retrieved data")
+            if 'daily_data' not in ticker_full[0] or ticker_full[0]['daily_data'] is None:
+                raise KeyError("Missing 'daily_data' in the retrieved data")
 
-        df = pd.DataFrame(ticker_full[0]['daily_data'])
+            df = pd.DataFrame(ticker_full[0]['daily_data'])
 
         selected_cols = ['date','open','high','low','close','volume']
 

neurostats_API/fetchers/tej_finance_report.py

@@ -0,0 +1,313 @@
+from .base import BaseTEJFetcher
+from datetime import datetime
+from enum import Enum
+import pandas as pd
+from pymongo import MongoClient
+from ..utils import StatsProcessor, YoY_Calculator
+import warnings
+
+
+class FinanceReportFetcher(BaseTEJFetcher):
+    class FetchMode(Enum):
+        YOY = 1
+        QOQ = 2
+        YOY_NOCAL = 3
+        QOQ_NOCAL = 4
+
+    def __init__(self, mongo_uri, db_name = "company", collection_name = "TWN/AINVFQ1"):
+        self.client = MongoClient(mongo_uri)
+        self.db = self.client[db_name]
+        self.collection = self.db[collection_name]
+
+        self.check_index = {
+            'coid', 'mdate', 'key3', 'no',
+            'sem', 'merg', 'curr', 'annd',
+            'fin_ind', 'bp11', 'bp21', 'bp22',
+            'bp31', 'bp41', 'bp51', 'bp53',
+            'bp61', 'bp62', 'bp63', 'bp64',
+            'bp65', 'bf11', 'bf12', 'bf21',
+            'bf22', 'bf41', 'bf42', 'bf43',
+            'bf44', 'bf45', 'bf99', 'bsca',
+            'bsnca', 'bsta','bscl','bsncl',
+            'bstl','bsse','bslse','debt',
+            'quick','ppe','ar','ip12',
+            'ip22','ip31','ip51','iv41',
+            'if11','isibt','isni','isnip',
+            'eps','ispsd','gm','opi',
+            'nri','ri','nopi','ebit',
+            'cip31','cscfo','cscfi','cscff',
+            'person','shares','wavg','taxrate',
+            'r104','r115','r105','r106',
+            'r107','r108','r201','r112',
+            'r401','r402','r403','r404',
+            'r405','r408','r409','r410',
+            'r502','r501','r205','r505',
+            'r517','r512','r509','r608',
+            'r616','r610','r607','r613',
+            'r612','r609','r614','r611',
+            'r307','r304','r305','r306',
+            'r316','r834'
+        }
+
+    def get(
+        self, 
+        ticker, 
+        fetch_mode: FetchMode = FetchMode.QOQ,
+        start_date: str = None,
+        end_date: str = None,
+        report_type: str = "Q",
+        indexes: list = []
+    ):
+        """
+        基礎的query function
+        ticker(str): 股票代碼
+        start_date(str):  開頭日期範圍
+        end_date(str):  = 結束日期範圍
+        report_type(str): 報告型態 {"A", "Q", "TTM"}  
+        fetch_mode(class FetchMode): 
+           YoY : 起始日期到結束日期範圍內，特定該季的資料
+           QoQ : 起始日期到結束日期內，每季的資料(與上一季成長率)
+        indexes(List): 指定的index
+        """
+        # 確認indexes中是否有錯誤的index，有的話回傳warning
+        if (indexes and self.check_index):
+            indexes = set(indexes)
+            difference = indexes-self.check_index
+            if (difference):
+                warnings.warn(
+                    f"{list(difference)} 沒有出現在資料表中，請確認column名稱是否正確",
+                    UserWarning
+                )
+
+
+        if (fetch_mode in {
+                self.FetchMode.QOQ,
+                self.FetchMode.QOQ_NOCAL
+            }   
+        ):
+            if (not start_date):
+                warnings.warn("No start_date specified, use default date = \"2005-01-01\"", UserWarning)
+                start_date = datetime.strptime("2005-01-01", "%Y-%m-%d")      
+            if (not end_date):
+                warnings.warn("No end_date specified, use default date = today", UserWarning)
+                end_date = datetime.today()
+
+            assert (start_date <= end_date)
+            start_date = datetime.strptime(start_date, "%Y-%m-%d")
+            end_date = datetime.strptime(end_date, "%Y-%m-%d")
+
+            start_year = start_date.year
+            start_season = (start_date.month - 1) // 4 + 1
+            end_year = end_date.year
+            end_season = (end_date.month - 1) // 4 + 1
+
+            if (fetch_mode == self.FetchMode.QOQ):
+                use_cal = True
+            else:
+                use_cal = False
+
+            data_df = self.get_QoQ_data(
+                ticker=ticker,
+                start_year=start_year,
+                start_season=start_season,
+                end_year=end_year,
+                end_season=end_season,
+                report_type=report_type,
+                indexes=indexes,
+                use_cal=use_cal
+            )
+
+            return data_df
+
+        elif (fetch_mode in {
+                self.FetchMode.YOY,
+                self.FetchMode.YOY_NOCAL
+            }
+        ):
+            start_year = 2005
+            end_date = self.get_latest_data_time(ticker)
+            if (not end_date):
+                end_date = datetime.today()
+
+            end_year = end_date.year
+            season = (end_date.month - 1) // 4 + 1
+
+            if (fetch_mode == self.FetchMode.YOY):
+                use_cal = True
+            else:
+                use_cal = False
+
+            data_df = self.get_YoY_data(
+                ticker = ticker,
+                start_year = start_year,
+                end_year = end_year,
+                season = season,
+                report_type = report_type,
+                indexes = indexes
+            )
+        
+            return data_df
+
+    def get_QoQ_data(
+        self,
+        ticker,
+        start_year,
+        start_season,
+        end_year,
+        end_season,
+        report_type = "Q",
+        indexes = [],
+        use_cal = False
+    ):
+        """
+            取得時間範圍內每季資料
+        """
+        if (not indexes): # 沒有指定 -> 取全部
+            pipeline = [
+                { "$match": { "ticker": ticker } },
+                { "$unwind": "$data" },
+                { "$match": {
+                    "$or": [
+                        { "data.year": { "$gt": start_year, "$lt": end_year } },
+                        { "data.year": start_year, "data.season": { "$gte": start_season } },
+                        { "data.year": end_year, "data.season": { "$lte": end_season } }
+                    ]
+                }},
+                { "$project": {
+                    "data.year": 1,
+                    "data.season": 1,
+                    f"data.{report_type}": 1,
+                    "_id": 0
+                    } 
+                }
+            ]
+
+        
+        else: # 取指定index
+            project_stage = {
+                "data.year": 1,
+                "data.season": 1
+            }
+            for index in indexes:
+                project_stage[f"data.{report_type}.{index}"] = 1
+
+            pipeline = [
+                { "$match": { "ticker": ticker } },
+                { "$unwind": "$data" },
+                { "$match": {
+                    "$or": [
+                        { "data.year": { "$gt": start_year, "$lt": end_year } },
+                        { "data.year": start_year, "data.season": { "$gte": start_season } },
+                        { "data.year": end_year, "data.season": { "$lte": end_season } }
+                    ]
+                }},
+                { "$project": project_stage }
+            ]
+
+
+        fetched_data = self.collection.aggregate(pipeline).to_list()
+        
+        data_dict = StatsProcessor.list_of_dict_to_dict(
+            fetched_data,
+            keys = ["year", "season"],
+            delimeter = "Q",
+            data_key=report_type
+        )
+        if (use_cal):
+            data_with_QoQ = self.cal_QoQ(data_dict)
+            data_df = pd.DataFrame.from_dict(data_with_QoQ)
+        else:
+            data_df = pd.DataFrame.from_dict(data_dict)
+        return data_df
+    
+    def get_YoY_data(
+        self,
+        ticker,
+        start_year,
+        end_year,
+        season,
+        report_type = "Q",
+        indexes = [],
+        use_cal = False
+    ):
+        """
+        取得某季歷年資料
+        """
+        if (use_cal):
+            select_year = set()
+
+            for year in range(start_year, end_year + 1):
+                year_shifts = {
+                    year,
+                    year - 1,
+                    year - 3,
+                    year - 5,
+                    year - 10
+                }
+
+                select_year = select_year.union(year_shifts)
+            
+            select_year = sorted(list(select_year), reverse=True)
+        else:
+            select_year = [year for year in range(start_year, end_year + 1)]
+
+        if (not indexes): # 沒有指定 -> 取全部
+            pipeline = [
+                { "$match": { "ticker": ticker } },
+                { "$unwind": "$data" },
+                { "$match": {
+                    "$or":[
+                        {
+                            "$and": [
+                                { "data.year": {"$in": select_year }},
+                                { "data.season": {"$eq": season}}
+                            ]
+                        },
+                    ]
+                }},
+                { "$project": {
+                    "data.year": 1,
+                    "data.season": 1,
+                    f"data.{report_type}": 1, 
+                    "_id": 0 
+                    } 
+                }
+            ]
+
+        
+        else: # 取指定index
+            project_stage = {
+                "data.year": 1,
+                "data.season": 1
+            }
+            for index in indexes:
+                project_stage[f"data.{report_type}.{index}"] = 1
+
+            pipeline = [
+                { "$match": { "ticker": ticker } },
+                { "$unwind": "$data" },
+                { "$match": {
+                    "$and": [
+                                { "data.year": {"$in": select_year}},
+                                { "data.season": {"$eq": season}}
+                            ]
+                }},
+                { "$project": project_stage }
+            ]
+
+        fetched_data = self.collection.aggregate(pipeline).to_list()
+
+        # 處理計算YoY
+        data_dict = StatsProcessor.list_of_dict_to_dict(
+            fetched_data,
+            keys = ['year', 'season'],
+            data_key=report_type,
+            delimeter='Q'
+        )
+        if (use_cal):
+            data_with_YoY = self.cal_YoY(data_dict, start_year, end_year)
+            result_df = pd.DataFrame.from_dict(data_with_YoY)
+        else:
+            result_df = pd.DataFrame.from_dict(data_dict)
+
+        return result_df

neurostats_API/tools/balance_sheet.yaml

@@ -21,6 +21,7 @@ current_debt:
   mode: value_and_percentage
   target_index: 
     - 流動負債合計
+    - 流動負債總額
 
 non_current_debt:
   mode: value_and_percentage

neurostats_API/tools/profit_lose.yaml

@@ -3,38 +3,45 @@ profit_lose:  # 總營收表
 
 grand_total_profit_lose:
   mode: grand_total_values
-
+  
 revenue:
   mode: growth
   target_index: 
     - 營業收入合計
+    - 收入合計
     - 利息收入
 
 grand_total_revenue:
   mode: grand_total_growth
   target_index: 
     - 營業收入合計
+    - 收入合計
     - 利息收入
 
 gross_profit:
   mode: growth
   target_index: 
     -  營業毛利（毛損）淨額
+    -  利息淨收益
 
 grand_total_gross_profit:
   mode: grand_total_growth
   target_index: 
     - 營業毛利（毛損）淨額
-
+    - 利息淨收益
+ 
 gross_profit_percentage:
   mode: percentage
   target_index: 
     - 營業毛利（毛損）淨額
+    - 利息淨收益
 
 grand_total_gross_profit_percentage:
   mode: grand_total_percentage
   target_index: 
     - 營業毛利（毛損）淨額
+    - 利息淨收益
+
 # 營利
 operating_income:
   mode: growth
@@ -55,46 +62,61 @@ grand_total_operating_income_percentage:
   mode: grand_total_percentage
   target_index: 
     - 營業利益（損失）
+
 # 稅前淨利
 net_income_before_tax:
   mode: growth
   target_index: 
     - 稅前淨利（淨損）
+    - 繼續營業單位稅前損益
+    - 繼續營業單位稅前淨利（淨損）
 
 grand_total_net_income_before_tax:
   mode: grand_total_growth
   target_index: 
     - 稅前淨利（淨損）
+    - 繼續營業單位稅前損益
+    - 繼續營業單位稅前淨利（淨損）
 
 net_income_before_tax_percentage:
   mode: percentage
   target_index: 
     - 稅前淨利（淨損）
+    - 繼續營業單位稅前損益
+    - 繼續營業單位稅前淨利（淨損）
 
 grand_total_net_income_before_tax_percentage:
   mode: grand_total_percentage
   target_index: 
     - 稅前淨利（淨損）
+    - 繼續營業單位稅前損益
+    - 繼續營業單位稅前淨利（淨損）
+
 # 本期淨利
 net_income:
   mode: growth
   target_index: 
     - 本期淨利（淨損）
+    - 本期稅後淨利（淨損）
 
 grand_total_net_income:
   mode: grand_total_growth
   target_index: 
     - 本期淨利（淨損）
+    - 本期稅後淨利（淨損）
 
 net_income_percentage:
   mode: percentage
   target_index: 
     - 本期淨利（淨損）
+    - 本期稅後淨利（淨損）
 
 grand_total_income_percentage:
   mode: grand_total_percentage
   target_index: 
     - 本期淨利（淨損）
+    - 本期稅後淨利（淨損）
+
 # EPS
 EPS:
   mode: value
@@ -118,4 +140,4 @@ grand_total_EPS_growth:
   mode: grand_total_growth
   target_index: 
     - 基本每股盈餘
-    - 基本每股盈餘合計
+    - 基本每股盈餘合計

neurostats_API/utils/__init__.py

@@ -1,4 +1,5 @@
 from .datetime import StatsDateTime
 from .db_client import DBClient
 from .data_process import StatsProcessor
-from .fetcher import StatsFetcher
+from .fetcher import StatsFetcher
+from .calculate_value import YoY_Calculator

neurostats_API/utils/calculate_value.py

@@ -0,0 +1,26 @@
+class YoY_Calculator:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def cal_growth(cls, target_value: float, past_value: float, delta: int):
+        """
+        計算成長率以及年化成長率
+        target_value: float，這個時間的數值
+        past_value: float，過去的這個時間數值
+        delta: int，代表隔了幾年/季 delta > 1 時改以年化成長率計算
+        """
+        try:
+            if (delta > 1):
+                YoY = ((target_value / past_value)**(1 / delta)) - 1
+
+            else:
+                YoY = ((target_value - past_value) / past_value)
+
+        except Exception as e:
+            return None
+
+        if (isinstance(YoY, complex)): # 年化成長率有複數問題
+            return None
+
+        return YoY

neurostats_API/utils/data_process.py

@@ -20,7 +20,6 @@ class StatsProcessor:
     """
     1. 讀檔: txt / yaml
     2. 將巢狀dictionary / DataFrame扁平化
-
     """
 
     @classmethod
@@ -97,20 +96,47 @@ class StatsProcessor:
         """
         對Multicolumn的dataframe切出目標的index
         """
-        times = total_table.columns.get_level_values(0).unique()
         try:
             target_metrics = target_metric_dict[mode]
         except KeyError as e:
             return f"mode Error: Get mode should be {list(target_metric_dict.keys())} but get {mode}"
 
+        times = total_table.columns.get_level_values(0).unique()
         desired_order = [(time, value_name) for time in times
                          for value_name in target_metrics]
 
         if (target_index):
             target_index = target_index.split()
-            sliced_table = total_table.loc[
-                target_index, pd.IndexSlice[:,
-                                            target_metrics]][desired_order].T
+            try:
+                sliced_table = total_table.loc[
+                    target_index,
+                    pd.IndexSlice[:, target_metrics]][desired_order].T
+
+            except Exception as e:  # 沒辦法完整取得表格
+                # 先設立空表格
+                empty_index = pd.Index(desired_order)
+                empty_columns = pd.Index(target_index)
+                sliced_table = pd.DataFrame(index=empty_index,
+                                            columns=empty_columns)
+
+                try:
+                    # 提取有效的部分資料
+                    partial_table = total_table.loc[
+                        total_table.index.intersection(target_index),
+                        pd.IndexSlice[:, target_metrics]
+                    ]
+
+                    # 遍歷 partial_table 的索引和值，手動填入 sliced_table
+                    for row_index in partial_table.index:
+                        for col_index in partial_table.columns:
+                            if col_index in desired_order and row_index in target_index:
+                                sliced_table.loc[col_index, row_index] = partial_table.loc[row_index, col_index]
+
+                    # 確保 `sliced_table` 的排序符合 `desired_order`
+                    sliced_table = sliced_table.reindex(index=desired_order, columns=target_index)
+                except Exception as sub_e:
+                    pass
+
             if (mode == 'value_and_percentage'):  # 因應balance_sheet 頁面的格式
                 return_table = sliced_table.T
                 return_table.columns = [
@@ -146,7 +172,7 @@ class StatsProcessor:
     @classmethod
     def cal_percentage(cls, value, postfix="%"):
         if (isinstance(value, (float, int))):
-            value = np.round(value * 100 , 2).item()
+            value = np.round(value * 100, 2).item()
             value = f"{value:.2f}{postfix}"
 
             return value
@@ -157,7 +183,7 @@ class StatsProcessor:
     @classmethod
     def cal_non_percentage(cls, value, to_str=False, postfix="元"):
         if (isinstance(value, (float, int))):
-            
+
             value = np.round(value, 2).item()
             if (postfix == "千元"):
                 value *= 1000
@@ -178,7 +204,7 @@ class StatsProcessor:
 
         else:
             return value
-    
+
     @classmethod
     def cal_round_int(cls, value):
         """
@@ -188,3 +214,59 @@ class StatsProcessor:
             return int(np.round(value).item())
         else:
             return value
+    
+    @classmethod
+    def list_of_dict_to_dict(
+        cls,
+        data_list: list,
+        key: str = "",
+        keys: list = [],
+        delimeter: str = "_",
+        data_key: str = "Q"
+    ):
+        """
+        TEJ DB 用
+        List[Dict] -> Dict[Dict]
+        input:
+        data_list(List):
+            [
+                { "data":
+                    {
+                        "year": 2021...
+                        "season": 1,
+                        "Q": {}...
+                        
+                    }
+                }
+            ]
+        
+        key(str): 選擇哪一個key作為轉化後的index
+        delimeter(str): 多個key時要用甚麼分隔
+        return: 
+        {
+            "2021" : {# Q下的資料} ...
+        }
+
+        or  (keys = ['year', 'season'])
+        {
+            "2021Q2" : {}
+        }
+        """
+        assert (key or keys), "func list_of_dict_to_dict must have argument \"key\" or \"keys\""
+
+        return_dict = {}
+        if (key):
+            keys = [key]
+        for data in data_list:
+            data = data['data']
+
+            pop_keys = []
+
+            for key in keys:
+                assert (key in data.keys())
+                pop_keys.append(str(data.pop(key)))
+            
+            pop_key = delimeter.join(pop_keys)
+            return_dict[pop_key] = data[data_key]
+        
+        return return_dict

{neurostats_API-0.0.13.dist-info → neurostats_API-0.0.15.dist-info}/METADATA

@@ -1,12 +1,18 @@
 Metadata-Version: 2.1
 Name: neurostats-API
-Version: 0.0.13
+Version: 0.0.15
 Summary: The service of NeuroStats website
 Home-page: https://github.com/NeurowattStats/NeuroStats_API.git
 Author: JasonWang@Neurowatt
 Author-email: jason@neurowatt.ai
 Requires-Python: >=3.6
 Description-Content-Type: text/markdown
+Requires-Dist: numpy>=2.1.0
+Requires-Dist: pandas>=2.2.0
+Requires-Dist: pymongo
+Requires-Dist: pytz
+Requires-Dist: python-dotenv
+Requires-Dist: yfinance
 
 # neurostats_API
 
@@ -83,7 +89,7 @@ pip install neurostats-API
 ```Python 
 >>> import neurostats_API
 >>> print(neurostats_API.__version__)
-0.0.13
+0.0.15
 ```
 
 ### 得到最新一期的評價資料與歷年評價
@@ -667,7 +673,71 @@ fetcher.query()
 
 請注意`range`, `last_range`, `52week_range`這三個項目型態為字串，其餘為float
 
+
+## TEJ 相關
+### 會計師簽證財務資料
+```Python
+from neurostats_API import FinanceReportFetcher
+
+mongo_uri = <MongoDB 的 URI>
+db_name = 'company' # 連接的DB名稱
+collection_name = "TWN/AINVFQ1" # 連接的collection對象
+
+fetcher = FinanceReportFetcher(
+    mongo_uri = mongo_uri,
+    db_name = db_name,
+    collection_name = collection_name
+)
+
+data = fetcher.get(
+    ticker = "2330" # 任意的股票代碼
+    fetch_mode = fetcher.YOY_NOCAL # 取得模式
+    start_date = "2005-01-01",
+    end_date = "2024-12-31",
+    report_type = "Q",
+    indexes = []
+)
+```
+- `ticker`: 股票代碼
+
+- `fetch_mode` : 取得模式，為`fetcher.YOY_NOCAL` 或 `fetcher.QOQ_NOCAL`
+    - `YOY_NOCAL`: 以end_date為準，取得與end_date為準同季的歷年資料，時間範圍以start_date為準
+        > 例如`start_date = "2020-07-01"`, `end_date = "2024-01-01"`，會回傳2020~2024的第一季資料
+
+    - `QOQ_NOCAL`: 時間範圍內的每季資料
+
+- `start_date`: 開始日期，不設定時預設為`2005-01-01`
+
+- `end_date`: 結束日期，不設定時預設為資料庫最新資料的日期
+
+- `report_type`: 選擇哪種報告，預設為`Q`
+    - `A`: 當年累計
+    - `Q`: 當季數值
+    - `TTM`: 移動四季 (包括當季在內，往前累計四個季度)
+
+- `indexes`: 選擇的column，需要以TEJ提供的欄位名稱為準，不提供時或提供`[]`會回傳全部column
+   - 範例輸入: `['bp41', 'bp51']`
+
+[TEJ資料集連結](https://tquant.tejwin.com/%E8%B3%87%E6%96%99%E9%9B%86/)
+請看 `會計師簽證財務資料`
+
+#### 回傳資料
+fetch_mode設定為`YOY_NOCAL`與`QOQ_NOCAL`下
+為回傳pd.DataFrame，column名稱為<年份>Q<季>， row名稱為指定財報項目
+
 ## 版本紀錄
+## 0.0.15
+- TechFetcher中新增指數條件
+
+- 新增tej_fetcher索取TEJ相關的資料
+
+- package新增depensnecy，可以安裝需要的相關package
+
+## 0.0.14
+- 修改部分財報資料錯誤的乘以1000的問題
+
+- 新增例外處理: 若資料庫對於季資料一部分index缺失的情況下仍會盡可能去將資料蒐集並呈現
+
 ### 0.0.13
 - value_fetcher 新增獲得一序列評價的功能
 

{neurostats_API-0.0.13.dist-info → neurostats_API-0.0.15.dist-info}/RECORD

@@ -1,28 +1,30 @@
-neurostats_API/__init__.py,sha256=UDEAcv3wrFvy1-DjCv7qON8202kc3zUGHpfClyTJWhs,20
+neurostats_API/__init__.py,sha256=oEkbIWbrC6-8sBPEJQXg0QYoz3TNZtYXhSTEO6d0JcU,261
 neurostats_API/cli.py,sha256=UJSWLIw03P24p-gkBb6JSEI5dW5U12UvLf1L8HjQD-o,873
 neurostats_API/main.py,sha256=QcsfmWivg2Dnqw3MTJWiI0QvEiRs0VuH-BjwQHFCv00,677
-neurostats_API/fetchers/__init__.py,sha256=ylYEySHQxcAhUUWEOCGZfmaAg7Mir5MfuEhOjk3POJg,406
+neurostats_API/fetchers/__init__.py,sha256=B4aBwVzf_X-YieEf3fZteU0qmBPVIB9VjrmkyWhLK18,489
 neurostats_API/fetchers/balance_sheet.py,sha256=sQv4Gk5uoKURLEdh57YknOQWiyVwaXJ2Mw75jxNqUS0,5804
-neurostats_API/fetchers/base.py,sha256=NW2SFzrimyAIrdJx1LVmTazelyZOAtcj54kJKHc4Vaw,1662
+neurostats_API/fetchers/base.py,sha256=4YS8MJR3u9Sg6dKX7QoCYuqNeQaoYHIlvPm5x8VQ72U,4882
 neurostats_API/fetchers/cash_flow.py,sha256=TY7VAWVXkj5-mzH5Iu0sIE-oV8MvGmmDy0URNotNV1E,7614
 neurostats_API/fetchers/finance_overview.py,sha256=PxUdWY0x030olYMLcCHDBn068JLmCE2RTOce1dxs5vM,27753
-neurostats_API/fetchers/institution.py,sha256=rEcs5-JKbWjg8lSjG1V3YdHJQuPSN1GntnxvedESCeo,10984
+neurostats_API/fetchers/institution.py,sha256=UrcBc6t7u7CnEwUsf6YmLbbJ8VncdWpq8bCz17q2dgs,11168
 neurostats_API/fetchers/margin_trading.py,sha256=lQImtNdvaBoSlKhJvQ3DkH3HjSSgKRJz4ZZpyR5-Z4I,10433
 neurostats_API/fetchers/month_revenue.py,sha256=nixX2llzjCFr2m2YVjxrSfkBusnZPrPb2dRDq1XLGhw,4251
-neurostats_API/fetchers/profit_lose.py,sha256=xlLNsGSy4Azf4HyZyYaX3dFad-ACO-vuQToBooZi1_w,5698
-neurostats_API/fetchers/tech.py,sha256=wH1kkqiETQhF0HAhk-UIiucnZ3EiL85Q-yMWCcVOiFM,11395
+neurostats_API/fetchers/profit_lose.py,sha256=EN9Y0iamcAaHMZdjHXO6b_2buLnORssf8ZS7A0hi74s,5896
+neurostats_API/fetchers/tech.py,sha256=Hol1bcwJ_ERcnoTXNWlqqaWOuzdl7MeiAjCvzQMZDTg,12269
+neurostats_API/fetchers/tej_finance_report.py,sha256=VDP0Lx2ErCgIBBz7nbquC1ugkcnj6p7ehM2JtFInjsQ,10218
 neurostats_API/fetchers/value_invest.py,sha256=_eQxuEnIYvksb06QHixGK29Gnwr_3xmI6Tu7dv4J__E,5769
-neurostats_API/tools/balance_sheet.yaml,sha256=yTxrWh7m4K3LnaNunETidfNzl6S4Bf58VIg9U38XShQ,648
+neurostats_API/tools/balance_sheet.yaml,sha256=6XygNG_Ybb1Xkk1e39LMLKr7ATvaCP3xxuwFbgNl6dA,673
 neurostats_API/tools/cash_flow_percentage.yaml,sha256=fk2Z4eb1JjGFvP134eJatHacB7BgTkBenhDJr83w8RE,1345
 neurostats_API/tools/finance_overview_dict.yaml,sha256=B9nV75StXkrF3yv2-eezzitlJ38eEK86RD_VY6588gQ,2884
-neurostats_API/tools/profit_lose.yaml,sha256=dcO-0J0BC4p06XBNuowu8ux0NTbyZiOkGfy6szHF6fw,2402
+neurostats_API/tools/profit_lose.yaml,sha256=iyp9asYJ04vAxk_HBUDse_IBy5oVvYHpwsyACg5YEeg,3029
 neurostats_API/tools/seasonal_data_field_dict.txt,sha256=X8yc_el6p8BH_3FikTqBVFGsvWdXT6MHXLfKfi44334,8491
-neurostats_API/utils/__init__.py,sha256=FTYKRFzW2XVXdnSHXnS3mQQaHlKF9xGqrMsgZZ2kroc,142
-neurostats_API/utils/data_process.py,sha256=2yrO0iP1LHhF0uhXZ442PHQBI-Zd2xIqNItkNf5hKIc,6339
+neurostats_API/utils/__init__.py,sha256=0tJCRmlJq2aDwcNNW-oEaA9H0OxTJMFvjpVYtG4AvZU,186
+neurostats_API/utils/calculate_value.py,sha256=lUKSsWU76XRmDUcmi4eDjoQxjb3vWpAAKInF9w49VNI,782
+neurostats_API/utils/data_process.py,sha256=A--dzOsu42jRxqqCD41gTtjE5rhEBYmhB6y-AnCvo5U,8986
 neurostats_API/utils/datetime.py,sha256=XJya4G8b_-ZOaBbMXgQjWh2MC4wc-o6goQ7EQJQMWrQ,773
 neurostats_API/utils/db_client.py,sha256=OYe6yazcR4Aa6jYmy47JrryUeh2NnKGqY2K_lSZe6i8,455
 neurostats_API/utils/fetcher.py,sha256=VbrUhjA-GG5AyjPX2SHtFIbZM4dm3jo0RgZzuCbb_Io,40927
-neurostats_API-0.0.13.dist-info/METADATA,sha256=tj8M4sSgXPprPV472mYHdEY_QOBpcObMXXo5lLGSOic,25744
-neurostats_API-0.0.13.dist-info/WHEEL,sha256=bFJAMchF8aTQGUgMZzHJyDDMPTO3ToJ7x23SLJa1SVo,92
-neurostats_API-0.0.13.dist-info/top_level.txt,sha256=nSlQPMG0VtXivJyedp4Bkf86EOy2TpW10VGxolXrqnU,15
-neurostats_API-0.0.13.dist-info/RECORD,,
+neurostats_API-0.0.15.dist-info/METADATA,sha256=btfdGRam5QpUHFFiA_UPWYeZuAqAMYkEJ0Ufod399T4,27959
+neurostats_API-0.0.15.dist-info/WHEEL,sha256=bFJAMchF8aTQGUgMZzHJyDDMPTO3ToJ7x23SLJa1SVo,92
+neurostats_API-0.0.15.dist-info/top_level.txt,sha256=nSlQPMG0VtXivJyedp4Bkf86EOy2TpW10VGxolXrqnU,15
+neurostats_API-0.0.15.dist-info/RECORD,,