StockSense-AI / backend /stock_data.py
PROSTLE's picture
fix: implement robust historical fallbacks for market data
d5687d1
Raw History Blame Contribute Delete
13.3 kB
import yfinance as yf
import pandas as pd
import ta
from datetime import datetime
def _get_yf_session():
"""Create a yfinance session with proper headers to bypass IP blocking."""
import requests
session = requests.Session()
session.headers.update({
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
})
return session
def fetch_stock_data(ticker: str, period: str = "2y", interval: str = "1d") -> pd.DataFrame:
session = _get_yf_session()
stock = yf.Ticker(ticker, session=session)
df = stock.history(period=period, interval=interval)
if df.empty:
raise ValueError(f"No data found for ticker: {ticker}")
df.reset_index(inplace=True)
return df
def add_technical_indicators(df: pd.DataFrame) -> pd.DataFrame:
close = df["Close"]
high = df["High"]
low = df["Low"]
volume = df["Volume"]
df["SMA_20"] = ta.trend.sma_indicator(close, window=20)
df["SMA_50"] = ta.trend.sma_indicator(close, window=50)
df["EMA_20"] = ta.trend.ema_indicator(close, window=20)
df["RSI"] = ta.momentum.rsi(close, window=14)
macd = ta.trend.MACD(close)
df["MACD"] = macd.macd()
df["MACD_Signal"] = macd.macd_signal()
df["MACD_Hist"] = macd.macd_diff()
bb = ta.volatility.BollingerBands(close, window=20, window_dev=2)
df["BB_Upper"] = bb.bollinger_hband()
df["BB_Lower"] = bb.bollinger_lband()
df["BB_Width"] = bb.bollinger_wband()
df["ATR"] = ta.volatility.average_true_range(high, low, close, window=14)
# New Technical Indicators Added for Accuracy
df["Stoch_k"] = ta.momentum.stoch(high, low, close)
df["Stoch_d"] = ta.momentum.stoch_signal(high, low, close)
df["OBV"] = ta.volume.on_balance_volume(close, volume)
# VWAP Calculation
df["VWAP"] = (df["Volume"] * (df["High"] + df["Low"] + df["Close"]) / 3).cumsum() / df["Volume"].cumsum()
df["Volume_SMA_20"] = ta.trend.sma_indicator(volume, window=20)
df.dropna(inplace=True)
return df
def get_stock_info(ticker: str) -> dict:
session = _get_yf_session()
stock = yf.Ticker(ticker, session=session)
info = stock.info or {}
# Reliability Fallback: If price is 0, try to get it from history
price = info.get("currentPrice", info.get("regularMarketPrice", 0))
if not price or price == 0:
hist = stock.history(period="5d")
if not hist.empty:
price = round(float(hist["Close"].iloc[-1]), 2)
return {
"name": info.get("shortName", ticker),
"sector": info.get("sector", "N/A"),
"currency": info.get("currency", "INR"),
"current_price": price,
"market_cap": info.get("marketCap", 0),
"day_high": info.get("dayHigh", price),
"day_low": info.get("dayLow", price),
"prev_close": info.get("previousClose", price),
"open": info.get("open", price),
"volume": info.get("volume", 0),
"fifty_two_week_high": info.get("fiftyTwoWeekHigh", price),
"fifty_two_week_low": info.get("fiftyTwoWeekLow", price),
}
def fetch_live_price(ticker: str) -> dict:
from datetime import datetime
session = _get_yf_session()
stock = yf.Ticker(ticker, session=session)
# Prefer fast_info for more robust fallback live data
fast = stock.fast_info
info = stock.info or {}
prev_close = fast.get("previousClose", info.get("previousClose", 0))
current = fast.get("lastPrice", info.get("currentPrice", info.get("regularMarketPrice", 0)))
# Fallback to history if live data returns zero
if not current or current == 0 or not prev_close or prev_close == 0:
hist = stock.history(period="5d")
if not hist.empty and len(hist) >= 2:
current = current or round(float(hist["Close"].iloc[-1]), 2)
prev_close = prev_close or round(float(hist["Close"].iloc[-2]), 2)
elif not hist.empty:
current = current or round(float(hist["Close"].iloc[-1]), 2)
prev_close = prev_close or current
change = round(current - prev_close, 2) if current and prev_close else 0
change_pct = round((change / prev_close) * 100, 2) if prev_close else 0
return {
"ticker": ticker,
"price": current,
"change": change,
"change_pct": change_pct,
"direction": "up" if change >= 0 else "down",
"day_high": info.get("dayHigh", current),
"day_low": info.get("dayLow", current),
"volume": info.get("volume", 0),
"timestamp": datetime.now().isoformat(),
}
TIMEFRAME_MAP = {
"1d": ("1d", "5m"),
"5d": ("5d", "15m"),
"1wk": ("5d", "15m"),
"1mo": ("1mo", "1h"),
"3mo": ("3mo", "1d"),
"6mo": ("6mo", "1d"),
"1y": ("1y", "1d"),
"2y": ("2y", "1wk"),
}
def fetch_chart_data(ticker: str, timeframe: str) -> list[dict]:
period, interval = TIMEFRAME_MAP.get(timeframe, ("1mo", "1h"))
session = _get_yf_session()
stock = yf.Ticker(ticker, session=session)
df = stock.history(period=period, interval=interval)
if df.empty:
raise ValueError(f"No chart data for {ticker}")
df.reset_index(inplace=True)
records = []
date_col = "Datetime" if "Datetime" in df.columns else "Date"
for _, row in df.iterrows():
records.append({
"time": str(row[date_col]),
"open": round(float(row["Open"]), 2),
"high": round(float(row["High"]), 2),
"low": round(float(row["Low"]), 2),
"close": round(float(row["Close"]), 2),
"volume": int(row["Volume"]),
})
return records
# ─────────────────────────────────────────────────
# INTRADAY DATA & FEATURES (for Auto Trade only)
# ─────────────────────────────────────────────────
import numpy as np
def fetch_intraday_data(ticker: str, period: str = "5d", interval: str = "1m") -> pd.DataFrame:
"""Fetch 1-min intraday bars (max 7 days with yfinance)."""
session = _get_yf_session()
stock = yf.Ticker(ticker, session=session)
df = stock.history(period=period, interval=interval)
if df.empty:
raise ValueError(f"No intraday data for {ticker}")
df.reset_index(inplace=True)
col = "Datetime" if "Datetime" in df.columns else "Date"
df.rename(columns={col: "Datetime"}, inplace=True)
df["Datetime"] = pd.to_datetime(df["Datetime"])
return df
def _supertrend(df: pd.DataFrame, period: int = 10, multiplier: float = 3.0):
"""Compute Supertrend indicator (period, multiplier)."""
hl2 = (df["High"] + df["Low"]) / 2
atr = ta.volatility.average_true_range(df["High"], df["Low"], df["Close"], window=period)
upper_band = hl2 + multiplier * atr
lower_band = hl2 - multiplier * atr
supertrend = pd.Series(0.0, index=df.index)
direction = pd.Series(1, index=df.index) # 1 = bullish, -1 = bearish
for i in range(1, len(df)):
if df["Close"].iloc[i] > upper_band.iloc[i - 1]:
direction.iloc[i] = 1
elif df["Close"].iloc[i] < lower_band.iloc[i - 1]:
direction.iloc[i] = -1
else:
direction.iloc[i] = direction.iloc[i - 1]
if direction.iloc[i] == 1 and lower_band.iloc[i] < lower_band.iloc[i - 1]:
lower_band.iloc[i] = lower_band.iloc[i - 1]
if direction.iloc[i] == -1 and upper_band.iloc[i] > upper_band.iloc[i - 1]:
upper_band.iloc[i] = upper_band.iloc[i - 1]
supertrend.iloc[i] = lower_band.iloc[i] if direction.iloc[i] == 1 else upper_band.iloc[i]
return supertrend, direction
def add_intraday_indicators(df: pd.DataFrame) -> pd.DataFrame:
"""Add all intraday trading indicators to a DataFrame of 1-min bars."""
close = df["Close"]
high = df["High"]
low = df["Low"]
volume = df["Volume"].astype(float)
# Moving averages
df["EMA_9"] = ta.trend.ema_indicator(close, window=9)
df["EMA_21"] = ta.trend.ema_indicator(close, window=21)
df["SMA_50"] = ta.trend.sma_indicator(close, window=50)
# Price vs MAs (percentage distance)
df["price_vs_ema9"] = ((close - df["EMA_9"]) / df["EMA_9"]) * 100
df["price_vs_ema21"] = ((close - df["EMA_21"]) / df["EMA_21"]) * 100
df["price_vs_sma50"] = ((close - df["SMA_50"]) / df["SMA_50"]) * 100
# Returns
df["returns_1bar"] = close.pct_change(1) * 100
df["returns_2bar"] = close.pct_change(2) * 100
df["returns_3bar"] = close.pct_change(3) * 100
df["returns_5bar"] = close.pct_change(5) * 100
df["returns_15bar"] = close.pct_change(15) * 100
# Momentum
df["RSI"] = ta.momentum.rsi(close, window=14)
macd_obj = ta.trend.MACD(close, window_fast=12, window_slow=26, window_sign=9)
df["MACD"] = macd_obj.macd()
df["MACD_Signal"] = macd_obj.macd_signal()
df["MACD_Hist"] = macd_obj.macd_diff()
# Volatility
df["ATR"] = ta.volatility.average_true_range(high, low, close, window=14)
bb = ta.volatility.BollingerBands(close, window=20, window_dev=2)
df["BB_Upper"] = bb.bollinger_hband()
df["BB_Lower"] = bb.bollinger_lband()
df["BB_Width"] = bb.bollinger_wband()
bb_range = df["BB_Upper"] - df["BB_Lower"]
df["BB_Position"] = np.where(bb_range > 0, (close - df["BB_Lower"]) / bb_range, 0.5)
# VWAP
cumvol = volume.cumsum()
cumtp = (((high + low + close) / 3) * volume).cumsum()
df["VWAP"] = np.where(cumvol > 0, cumtp / cumvol, close)
df["price_vs_vwap"] = ((close - df["VWAP"]) / df["VWAP"]) * 100
# Volume ratio
vol_sma20 = volume.rolling(window=20).mean()
df["volume_ratio"] = np.where(vol_sma20 > 0, volume / vol_sma20, 1.0)
# Supertrend
st_val, st_dir = _supertrend(df, period=10, multiplier=3.0)
df["Supertrend"] = st_val
df["Supertrend_Dir"] = st_dir # 1=bullish, -1=bearish
df["supertrend_dist"] = ((close - st_val) / close) * 100
# Time features
if "Datetime" in df.columns:
dt = pd.to_datetime(df["Datetime"])
df["minute_of_day"] = dt.dt.hour * 60 + dt.dt.minute
df["is_first_hour"] = (dt.dt.hour == 9).astype(int)
df["is_last_hour"] = (dt.dt.hour >= 15).astype(int)
else:
df["minute_of_day"] = 0
df["is_first_hour"] = 0
df["is_last_hour"] = 0
df.dropna(inplace=True)
return df
def fetch_india_vix() -> dict:
"""Fetch India VIX current level and recent change."""
try:
stock = yf.Ticker("^INDIAVIX")
hist = stock.history(period="7d")
if hist.empty:
return {"vix_level": 15.0, "vix_change_5d": 0.0, "available": False}
current = float(hist["Close"].iloc[-1])
if len(hist) >= 5:
prev = float(hist["Close"].iloc[-5])
else:
prev = float(hist["Close"].iloc[0])
change = round(current - prev, 2)
return {"vix_level": round(current, 2), "vix_change_5d": change, "available": True}
except Exception:
return {"vix_level": 15.0, "vix_change_5d": 0.0, "available": False}
def fetch_nifty_trend() -> dict:
"""Fetch Nifty 50 trend vs EMA-9 and EMA-21."""
try:
stock = yf.Ticker("^NSEI")
hist = stock.history(period="1mo", interval="1d")
if len(hist) < 21:
return {"nifty_price": 0, "trend": "flat", "above_ema9": False,
"above_ema21": False, "available": False}
close = hist["Close"]
ema9 = ta.trend.ema_indicator(close, window=9)
ema21 = ta.trend.ema_indicator(close, window=21)
price = float(close.iloc[-1])
e9 = float(ema9.iloc[-1])
e21 = float(ema21.iloc[-1])
above_ema9 = price > e9
above_ema21 = price > e21
if above_ema9 and above_ema21:
trend = "bullish"
elif not above_ema9 and not above_ema21:
trend = "bearish"
else:
trend = "flat"
return {
"nifty_price": round(price, 2),
"ema9": round(e9, 2),
"ema21": round(e21, 2),
"trend": trend,
"above_ema9": above_ema9,
"above_ema21": above_ema21,
"available": True,
}
except Exception:
return {"nifty_price": 0, "trend": "flat", "above_ema9": False,
"above_ema21": False, "available": False}
def detect_ipo_stock(ticker: str) -> dict:
"""Check if stock was listed recently (IPO < 180 days)."""
try:
stock = yf.Ticker(ticker)
hist = stock.history(period="max")
if hist.empty:
return {"is_ipo": False, "ipo_age_days": 999, "listing_date": None}
first_date = hist.index[0]
if hasattr(first_date, "date"):
first_date = first_date.date()
age_days = (datetime.now().date() - first_date).days
return {
"is_ipo": age_days < 180,
"ipo_age_days": age_days,
"listing_date": str(first_date),
}
except Exception:
return {"is_ipo": False, "ipo_age_days": 999, "listing_date": None}