11import json
22from collections import defaultdict
3+ from concurrent .futures import ThreadPoolExecutor , as_completed
34
45import pandas as pd
56
910
1011
1112class History :
13+ BATCH_SIZE = 200 # Alpaca API limit for multi-symbol requests
14+
1215 def __init__ (self , data_url : str , headers : dict [str , str ], asset : Assets ) -> None :
1316 """Initializes an instance of the History class.
1417
@@ -51,7 +54,7 @@ def check_if_stock(self, symbol: str) -> AssetModel:
5154 ###########################################
5255 def get_stock_data (
5356 self ,
54- symbol : str ,
57+ symbol : str | list [ str ] ,
5558 start : str ,
5659 end : str ,
5760 timeframe : str = "1d" ,
@@ -61,28 +64,59 @@ def get_stock_data(
6164 sort : str = "asc" ,
6265 adjustment : str = "raw" ,
6366 ) -> pd .DataFrame :
64- """Retrieves historical stock data for a given symbol within a specified date range and timeframe.
67+ """Retrieves historical stock data for one or more symbols within a specified date range and timeframe.
6568
6669 Args:
67- symbol: The stock symbol to fetch data for.
70+ symbol: The stock symbol(s) to fetch data for. Can be a single symbol string or list of symbols .
6871 start: The start date for historical data in the format "YYYY-MM-DD".
6972 end: The end date for historical data in the format "YYYY-MM-DD".
7073 timeframe: The timeframe for the historical data. Default is "1d".
7174 feed: The data feed source. Default is "sip".
7275 currency: The currency for historical data. Default is "USD".
73- limit: The number of data points to fetch. Default is 1000.
76+ limit: The number of data points to fetch per symbol . Default is 1000.
7477 sort: The sort order for the data. Default is "asc".
7578 adjustment: The adjustment for historical data. Default is "raw".
7679
7780 Returns:
78- A pandas DataFrame containing the historical stock data for the given symbol and time range.
81+ A pandas DataFrame containing the historical stock data for the given symbol(s) and time range.
7982
8083 Raises:
8184 ValueError: If the given timeframe is not one of the allowed values.
8285 """
83- self .check_if_stock (symbol )
86+ # Handle single symbol or list of symbols
87+ is_single = isinstance (symbol , str )
88+ if is_single :
89+ assert isinstance (symbol , str ) # Type guard for mypy
90+ symbols_list : list [str ] = [symbol ]
91+ single_symbol : str = symbol
92+ else :
93+ assert isinstance (symbol , list ) # Type guard for mypy
94+ symbols_list = symbol
95+ single_symbol = "" # Won't be used in multi-symbol case
96+
97+ # Validate symbols are stocks
98+ for sym in symbols_list :
99+ self .check_if_stock (sym )
84100
85- url = f"{ self .data_url } { symbol }
101+ # If more than BATCH_SIZE symbols, need to batch the requests
102+ if not is_single and len (symbols_list ) > self .BATCH_SIZE :
103+ return self ._get_batched_stock_data (
104+ symbols_list ,
105+ start ,
106+ end ,
107+ timeframe ,
108+ feed ,
109+ currency ,
110+ limit ,
111+ sort ,
112+ adjustment ,
113+ )
114+
115+ # Determine if using single or multi-symbol endpoint
116+ if is_single :
117+ url = f"{ self .data_url } { single_symbol }
118+ else :
119+ url = f"{ self .data_url }
86120
87121 timeframe_mapping : dict = {
88122 "1m" : "1Min" ,
@@ -111,8 +145,105 @@ def get_stock_data(
111145 "feed" : feed ,
112146 "sort" : sort ,
113147 }
114- symbol_data = self .get_historical_data (symbol , url , params )
115- return self .preprocess_data (symbol_data , symbol )
148+
149+ # Add symbols parameter for multi-symbol request
150+ if not is_single :
151+ params ["symbols" ] = "," .join (symbols_list )
152+
153+ symbol_data = self .get_historical_data (symbols_list , url , params , is_single )
154+
155+ # Process data based on single or multi-symbol
156+ if is_single :
157+ return self .preprocess_data (symbol_data [single_symbol ], single_symbol )
158+ return self .preprocess_multi_data (symbol_data )
159+
160+ def _get_batched_stock_data (
161+ self ,
162+ symbols : list [str ],
163+ start : str ,
164+ end : str ,
165+ timeframe : str ,
166+ feed : str ,
167+ currency : str ,
168+ limit : int ,
169+ sort : str ,
170+ adjustment : str ,
171+ ) -> pd .DataFrame :
172+ """Handle large symbol lists by batching requests.
173+
174+ Args:
175+ symbols: List of symbols to fetch data for.
176+ start: The start date for historical data.
177+ end: The end date for historical data.
178+ timeframe: The timeframe for the historical data.
179+ feed: The data feed source.
180+ currency: The currency for historical data.
181+ limit: The number of data points to fetch per symbol.
182+ sort: The sort order for the data.
183+ adjustment: The adjustment for historical data.
184+
185+ Returns:
186+ A pandas DataFrame containing the historical stock data for all symbols.
187+ """
188+ # Split symbols into batches
189+ batches = [
190+ symbols [i : i + self .BATCH_SIZE ]
191+ for i in range (0 , len (symbols ), self .BATCH_SIZE )
192+ ]
193+
194+ # Use ThreadPoolExecutor for concurrent batch requests
195+ all_dfs = []
196+ with ThreadPoolExecutor (max_workers = 5 ) as executor :
197+ futures = []
198+ for batch in batches :
199+ future = executor .submit (
200+ self .get_stock_data ,
201+ batch ,
202+ start ,
203+ end ,
204+ timeframe ,
205+ feed ,
206+ currency ,
207+ limit ,
208+ sort ,
209+ adjustment ,
210+ )
211+ futures .append (future )
212+
213+ for future in as_completed (futures ):
214+ try :
215+ df = future .result ()
216+ if not df .empty :
217+ all_dfs .append (df )
218+ except Exception as e :
219+ # Log error but continue with other batches
220+ print (f"Error fetching batch: { e } )
221+
222+ if all_dfs :
223+ return pd .concat (all_dfs , ignore_index = True ).sort_values (["symbol" , "date" ])
224+ return pd .DataFrame ()
225+
226+ @staticmethod
227+ def preprocess_multi_data (
228+ symbols_data : dict [str , list [defaultdict ]],
229+ ) -> pd .DataFrame :
230+ """Preprocess data for multiple symbols.
231+
232+ Args:
233+ symbols_data: A dictionary mapping symbols to their bar data.
234+
235+ Returns:
236+ A pandas DataFrame containing the preprocessed historical stock data for all symbols.
237+ """
238+ all_dfs = []
239+ for symbol , data in symbols_data .items ():
240+ if data : # Only process if data exists
241+ df = History .preprocess_data (data , symbol )
242+ all_dfs .append (df )
243+
244+ if all_dfs :
245+ return pd .concat (all_dfs , ignore_index = True ).sort_values (["symbol" , "date" ])
246+ return pd .DataFrame ()
116247
117248 ###########################################
118249 # /////////// PreProcess Data \\\\\\\\\\\ #
@@ -169,35 +300,51 @@ def preprocess_data(symbol_data: list[defaultdict], symbol: str) -> pd.DataFrame
169300 # ///////// Get Historical Data \\\\\\\\\ #
170301 ###########################################
171302 def get_historical_data (
172- self , symbol : str , url : str , params : dict
173- ) -> list [defaultdict ]:
174- """Retrieves historical data for a given symbol.
303+ self , symbols : list [ str ] , url : str , params : dict , is_single : bool
304+ ) -> dict [ str , list [defaultdict ] ]:
305+ """Retrieves historical data for given symbol(s) .
175306
176307 Args:
177- symbol (str): The symbol for which to retrieve historical data.
178- url (str): The URL to send the request to.
179- params (dict): Additional parameters to include in the request.
308+ symbols: List of symbols for which to retrieve historical data.
309+ url: The URL to send the request to.
310+ params: Additional parameters to include in the request.
311+ is_single: Whether this is a single-symbol request.
180312
181313 Returns:
182- list[defaultdict]: A list of historical data for the given symbol .
314+ dict[str, list[defaultdict]] : A dictionary mapping symbols to their historical data .
183315 """
184316 page_token = None
185317 symbols_data = defaultdict (list )
318+
186319 while True :
187- params ["page_token" ] = page_token
320+ if page_token :
321+ params ["page_token" ] = page_token
322+
188323 response = json .loads (
189324 Requests ()
190325 .request (method = "GET" , url = url , headers = self .headers , params = params )
191326 .text
192327 )
193328
194- if not response .get ("bars" ):
195- raise Exception (
196- f"No historical data found for { symbol }
197- )
329+ # Handle single vs multi-symbol response format
330+ if is_single :
331+ if not response .get ("bars" ):
332+ raise Exception (
333+ f"No historical data found for { symbols [0 ]}
334+ )
335+ symbols_data [symbols [0 ]].extend (response .get ("bars" , []))
336+ else :
337+ # Multi-symbol response has bars nested under symbol keys
338+ bars = response .get ("bars" , {})
339+ if not bars :
340+ raise Exception (
341+ f"No historical data found for symbols: { ', ' .join (symbols )}
342+ )
343+ for symbol , symbol_bars in bars .items ():
344+ symbols_data [symbol ].extend (symbol_bars )
198345
199- symbols_data [symbol ].extend (response .get ("bars" , []))
200- page_token = response .get ("next_page_token" , "" )
346+ page_token = response .get ("next_page_token" )
201347 if not page_token :
202348 break
203- return symbols_data [symbol ]
349+
350+ return symbols_data
0 commit comments