Harden factor loader zip parsing and fallback

factor_attribution.py

@@ -1,7 +1,9 @@
 from __future__ import annotations
 
 import io
+import zipfile
 from pathlib import Path
+from urllib.error import URLError
 
 import pandas as pd
 
@@ -11,7 +13,14 @@ def _download_kf_zip_bytes() -> bytes:
 
 
 def _parse_kf_daily_csv(raw_bytes: bytes) -> pd.DataFrame:
-    text = raw_bytes.decode("utf-8")
+    with zipfile.ZipFile(io.BytesIO(raw_bytes)) as archive:
+        member_name = next(
+            name
+            for name in archive.namelist()
+            if not name.endswith("/") and name.lower().endswith((".csv", ".txt"))
+        )
+        text = archive.read(member_name).decode("utf-8-sig")
+
     lines = [line for line in text.splitlines() if line.strip()]
     header_index = next(i for i, line in enumerate(lines) if "Mkt-RF" in line)
     table = "\n".join(lines[header_index:])
@@ -31,10 +40,12 @@ def load_external_us_factors(cache_dir: Path | str = "data/factors") -> pd.DataF
     cache_path.parent.mkdir(parents=True, exist_ok=True)
 
     try:
-        factors = _parse_kf_daily_csv(_download_kf_zip_bytes())
-        factors.to_csv(cache_path)
-        return factors
-    except Exception:
+        raw_bytes = _download_kf_zip_bytes()
+    except (URLError, TimeoutError, ConnectionError, OSError):
         if cache_path.exists():
             return pd.read_csv(cache_path, index_col=0, parse_dates=True)
         raise
+
+    factors = _parse_kf_daily_csv(raw_bytes)
+    factors.to_csv(cache_path)
+    return factors