[ENH] Remove yfinance as a dependency and implement data_loader

pypfopt/data/__init__.py

@@ -0,0 +1,3 @@
+from .data_loader import available_tickers, load_marketcaps, load_stockdata
+
+__all__ = ["load_stockdata", "available_tickers", "load_marketcaps"]

pypfopt/data/data_loader.py

@@ -0,0 +1,143 @@
+from importlib import resources
+
+import pandas as pd
+
+
+def _load_raw_data(filename: str, **read_csv_kwargs):
+    with resources.files(__package__).joinpath(filename).open("r") as f:
+        return pd.read_csv(f, **read_csv_kwargs)
+
+
+def load_stockdata(tickers: list = None, start: str = None, end: str = None):
+    """
+    Load example stock price data.
+
+    This function loads a synthetic stock price dataset included with the
+    package. The data can optionally be filtered by ticker symbols and
+    date range.
+
+    Parameters
+    ----------
+    tickers : list of str, optional
+        List of ticker symbols to include. If ``None``, all available
+        tickers are returned.
+    start : str, optional
+        Start date for filtering the dataset (inclusive). Should be
+        interpretable by ``pandas.to_datetime``.
+    end : str, optional
+        End date for filtering the dataset (inclusive). Should be
+        interpretable by ``pandas.to_datetime``.
+
+    Returns
+    -------
+    pandas.DataFrame
+        DataFrame of stock prices indexed by date. Columns correspond to
+        ticker symbols and values represent price levels.
+
+    Notes
+    -----
+    The dataset is bundled with the package and does not rely on external
+    data sources. It is intended for examples and tutorials.
+    """
+    df = _load_raw_data("stock_prices.csv", parse_dates=["date"])
+
+    if start is not None:
+        df = df[df["date"] >= pd.to_datetime(start)]
+    if end is not None:
+        df = df[df["date"] <= pd.to_datetime(end)]
+
+    if tickers is not None:
+        cols = ["date"] + tickers
+        df = df[cols]
+
+    return df.set_index("date")
+
+
+def load_marketcaps(tickers: list = None):
+    """
+    Load bundled example market capitalisation data.
+
+    This function loads synthetic market capitalisation values for the
+    example assets included in the package.
+
+    Parameters
+    ----------
+    tickers : list of str, optional
+        List of ticker symbols to return. If ``None``, market caps for all
+        available tickers are returned.
+
+    Returns
+    -------
+    dict
+        Dictionary mapping ticker symbols to market capitalisation values.
+
+    Notes
+    -----
+    The values are synthetic and provided solely for use in examples
+    demonstrating portfolio optimisation methods.
+    """
+    df = _load_raw_data("market_caps.csv")
+
+    if tickers is not None:
+        available = set(df["ticker"])
+        invalid = set(tickers) - available
+        if invalid:
+            raise ValueError(f"Invalid tickers: {invalid}")
+
+        df = df[df["ticker"].isin(tickers)]
+
+    return dict(zip(df["ticker"], df["market_cap"]))
+
+
+def available_tickers():
+    """
+    Return the list of available ticker symbols.
+
+    Returns
+    -------
+    list of str
+        Sorted list of ticker symbols present in the bundled example
+        dataset.
+
+    Notes
+    -----
+    These tickers correspond to the columns available in the example
+    stock price dataset returned by :func:`load_stockdata`.
+    """
+    cols = [
+        "AAPL",
+        "ACN",
+        "AMD",
+        "AMZN",
+        "BAC",
+        "BLK",
+        "COST",
+        "CVS",
+        "DIS",
+        "DPZ",
+        "F",
+        "GILD",
+        "INTU",
+        "JD",
+        "JPM",
+        "KO",
+        "LUV",
+        "MA",
+        "MCD",
+        "MSFT",
+        "NAT",
+        "NVDA",
+        "PBI",
+        "PFE",
+        "SBUX",
+        "SPY",
+        "TGT",
+        "TM",
+        "TSLA",
+        "UL",
+        "UNH",
+        "WMT",
+        "XOM",
+    ]
+
+    return cols

pypfopt/data/market_caps.csv

@@ -0,0 +1,34 @@
+ticker,market_cap
+AAPL,752207537100
+ACN,1901675041287
+AMD,1465327913913
+AMZN,1199323675973
+BAC,316257187682
+BLK,316209068070
+COST,120876806275
+CVS,1733021410820
+DIS,1204224448427
+DPZ,1417604792703
+F,46066066120
+GILD,1939970155063
+INTU,1665723068396
+JD,428616525803
+JPM,367740809578
+KO,370891997157
+LUV,611963274704
+MA,1051889081106
+MCD,866730312191
+MSFT,586002134695
+NAT,1225646524971
+NVDA,283290252000
+PBI,587828573827
+PFE,735891877370
+SBUX,914859618512
+SPY,1571426042979
+TGT,403349195405
+TM,1030897704635
+TSLA,1186867064879
+UL,97668573376
+UNH,1217051979543
+WMT,345195626756
+XOM,134777928005

pyproject.toml

@@ -61,8 +61,8 @@ all_extras = [
 
 # dev - the developer dependency set, for contributors and CI
 dev = [
-    "pytest>=9.0.2",
-    "yfinance>=1.2.0",
+    "pytest>=9.0.0",
+    "pytest-cov>=7.0.0",
 ]
 
 # notebook tests
@@ -93,6 +93,9 @@ requires = [
 [tool.setuptools.packages.find]
 exclude = ["example", "example.*", "tests", "tests.*"]
 
+[tool.setuptools.package-data]
+pypfopt = ["data/*.csv"]
+
 [tool.ruff]
 line-length = 88
 # Keep Ruff aligned with project target version