Squashed 'utils_v2/' content from commit ddefb8fe

git-subtree-dir: utils_v2
git-subtree-split: ddefb8fec3a72ccff2cd85e75bd6687d0067c37b
This commit is contained in:
2025-01-07 18:51:09 +05:30
commit c7259cfe9f
186 changed files with 141972 additions and 0 deletions
View File
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
+295
View File
@@ -0,0 +1,295 @@
"""
AUTHOR:
Khushal P Soonderji
DATE:
Thursday, 5th Dec., 2024
OBJECTIVE:
To provide a way to retrieve pre-market data from NSE. This is typically available by 9:10 AM.
NOTE: This method involves web scraping. It is good for proof-of-concept development, but it is recommended that
more professional data-sources be used when the product starts becoming mature.
REFERENCES:
N/A
DOWNLOADS:
N/A
"""
# *****************************************************************************************************************
# ***** ****
# *** IMPORT ***
# ***** ****
# *****************************************************************************************************************
# To make sibling directories accessible for imports:
import sys
sys.path.append(".")
sys.path.append("..")
# System-level activities:
import io
# My utils:
from utils_v2.string import json
from utils_v2.date_time import date_time
# NSE-related utils:
from utils_v2.nse.controllers.base import AsyncNSEBase
from utils_v2.nse.models.api_call import NSEApiResponse
# To make REST-ful API calls:
import httpx
# To work with date and time:
import datetime
# To work with datatypes:
from typing import Any, List
# For asynchronous activities:
import asyncio
# *****************************************************************************************************************
# ***** ****
# *** MACROS / ONE-TIME INIT ***
# ***** ****
# *****************************************************************************************************************
# --- Nothing Yet
# *****************************************************************************************************************
# ***** ****
# *** VARIABLES ***
# ***** ****
# *****************************************************************************************************************
# --- Nothing Yet
# *****************************************************************************************************************
# ***** ****
# *** FUNCTIONS ***
# ***** ****
# *****************************************************************************************************************
class NSEPreMarket(AsyncNSEBase):
# Symbol names:
PRE_MARKET_KEY_NIFTY = "NIFTY"
PRE_MARKET_KEY_BANK_NIFTY = "BANKNIFTY"
PRE_MARKET_KEY_SME = "SME"
PRE_MARKET_KEY_FO = "FO"
PRE_MARKET_KEY_OTHERS = "OTHERS"
PRE_MARKET_KEY_ALL = "ALL"
def __init__(
self,
http_client: httpx.AsyncClient,
cookies_refresh_interval: int | float = 300,
debug = True,
debug_prefix = "NSE (CECal) | ",
debug_only_errors = True
):
# Pass on the initialization to the parent:
super().__init__(
base_url = r"https://www.nseindia.com/market-data/pre-open-market-cm-and-emerge-market",
data_url = r"https://www.nseindia.com/api/market-data-pre-open",
http_client = http_client,
cookies_refresh_interval = cookies_refresh_interval,
debug = debug,
debug_prefix = debug_prefix,
debug_only_errors = debug_only_errors
)
async def get_data(
self,
key: str,
return_raw: bool = False,
refresh_cookies: bool = True,
force_refresh_cookies: bool = False,
retry_count: int = 1,
backoff_seconds: int | float = 0.5,
backoff_multiplier: float = 1.1
) -> NSEApiResponse:
"""
To get the data of the pre-open market trading. Useful for finding gaps and expected unusual activity in the
trading hours.
:param key: The type of pre-market data that you want. Choose from the class variables.
:param return_raw: Whether you want the raw JSON from NSE or you want it formatted.
:param refresh_cookies: Whether, or not, you would like to refresh the cookies.
:param force_refresh_cookies: If set to True, cookies will be refreshed even if not timed out. If set to False,
cookies will be refreshed only when the interval passed to the constructor has elapsed since the last
successful refresh.
:param retry_count: The no. of times to try to get the data from the API.
:param backoff_seconds: The delay between unsuccessful API calls.
:param backoff_multiplier: The multiplier to add to the delay to change delay.
:return: The raw or formatted event calendar data in the 'data' field of the response model.
"""
# Make the API call:
api_response = await self.get(
params = {"key": key},
refresh_cookies = refresh_cookies,
force_refresh_cookies = force_refresh_cookies,
retry_count = retry_count,
backoff_seconds = backoff_seconds,
backoff_multiplier = backoff_multiplier
)
# If the API call was successful:
if api_response.httpCode in [200]:
api_response.success = True
if return_raw: api_response.data = await api_response.get_json()
else:
try: api_response.data = self.format_data(
raw_json = await api_response.get_json(),
timestamp = date_time.get_current_utc_date_time(as_string = False),
raise_exception = True
)
except Exception as exception:
api_response.exception = exception
api_response.success = False
# Done here:
return api_response
@staticmethod
def format_data(
raw_json: dict,
timestamp: datetime.datetime = None,
raise_exception: bool = False
) -> List[dict] | None:
"""
We format the data here to be able to retrieve it properly later.
:param raw_json: The raw data as scraped from NSE.
:param timestamp: The timestamp at which the data was scraped. This shall be useful for data retrieval from the
database, later.
:param raise_exception: If set to True, any exception will be propagated. If set to False, any exception will be
suppressed internally.
:return: The formatted data if successful, else None.
"""
# Can't do anything if the chain itself is null:
if raw_json is None: return raw_json
# Start by assuming failure:
formatted_data = None
# Ensure that we've got a proper timestamp:
if timestamp is None: timestamp = date_time.get_current_utc_date_time(as_string = False)
try:
# Start by extracting basic data:
formatted_data = {
"scrapeTs": timestamp,
"ts": date_time.to_timezone(
date_time.as_if_timezone(
date_time.parse_date_time(
input_value = raw_json["timestamp"],
date_formats = ["%d-%b-%Y %H:%M:%S"]
),
timezone = date_time.TIMEZONE_IST
),
timezone = date_time.TIMEZONE_UTC
),
"advances": raw_json["advances"],
"declines": raw_json["declines"],
"unchanged": raw_json["unchanged"],
"totalMarketCap": raw_json["totalmarketcap"],
"totalTradedValue": raw_json["totalTradedValue"],
"totalTradedVolume": raw_json["totalTradedVolume"],
"data": []
}
# Now we iterate through the symbol-wise data and extract what we need:
for raw_symbol_data in raw_json["data"]:
raw_symbol_metadata = raw_symbol_data["metadata"]
raw_symbol_detail = raw_symbol_data["detail"]["preOpenMarket"]
formatted_data["data"].append({
"symbol": raw_symbol_metadata["symbol"],
"ffmc": raw_symbol_metadata["marketCap"],
"trigger": raw_symbol_metadata["purpose"],
"yearHigh": raw_symbol_metadata["yearHigh"],
"yearLow": raw_symbol_metadata["yearLow"],
"prevClose": raw_symbol_metadata["previousClose"],
"premarketPrice": raw_symbol_metadata["iep"],
"chg": raw_symbol_metadata["change"],
"pChg": raw_symbol_metadata["pChange"],
"totalTradedVolume": raw_symbol_detail["totalTradedVolume"],
"totalBuyVolume": raw_symbol_detail["totalBuyQuantity"],
"totalSellVolume": raw_symbol_detail["totalSellQuantity"],
})
# Data sorting (descending order of percent change):
formatted_data["data"] = sorted(
formatted_data["data"],
key = lambda x: x["pChg"],
reverse = True
)
# If something goes wrong:
except Exception as exception:
formatted_data = None
if raise_exception: raise
# Done here:
return formatted_data
# *****************************************************************************************************************
# ***** ****
# *** MAIN PROGRAM ***
# ***** ****
# *****************************************************************************************************************
if __name__ == "__main__":
async def main():
# Create an HTTP client:
test_client = httpx.AsyncClient(
limits = httpx.Limits(
max_connections = 100, # ............ Maximum number of connections allowed in the pool.
max_keepalive_connections = 50, # ... Maximum number of connections that can be kept alive.
),
timeout = httpx.Timeout(
pool = 120.0, # .... Time to wait for a free connection from the pool.
connect = 2.5, # ... Time to wait for establishing a connection to the server.
write = 10.0, # .... Time to wait for sending data.
read = 9.9 # ....... Time to wait for receiving data.
)
)
# Create an instance of the scraper, and refresh its cookies:
my_nse = NSEPreMarket(http_client = test_client)
# Get and show the data:
api_response = await my_nse.get_data(key = my_nse.PRE_MARKET_KEY_ALL, return_raw = False)
print("SUMMARY:", api_response.to_markdown(), "\n---\n\n")
if api_response.success: print("PRE-MARKET DATA:", json.to_string(api_response.data, default = str))
if api_response.exception: raise api_response.exception
print("COUNT:", len(api_response.data["data"]))
asyncio.run(main())