sovai 0.2.67__tar.gz → 0.2.68__tar.gz

{sovai-0.2.67 → sovai-0.2.68}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: sovai
-Version: 0.2.67
+Version: 0.2.68
 Summary: python for asset management
 License: Proprietary License
 Author: Derek Snow

{sovai-0.2.67 → sovai-0.2.68}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [project]
 name = "sovai"
-version = "0.2.67"
+version = "0.2.68"
 description = "python for asset management"
 authors = [
     {name = "Derek Snow", email = "d.snow@outlook.com"}

{sovai-0.2.67 → sovai-0.2.68}/sovai/__init__.py

@@ -9,7 +9,7 @@ from .utils.file_management import update_data_files # Keep update_data_files im
 
 import warnings
 warnings.filterwarnings("ignore")
-
+update_data_files()
 
 try:
     from importlib.metadata import version

{sovai-0.2.67 → sovai-0.2.68}/sovai/get_data.py

@@ -993,6 +993,7 @@ def _get_ticker_codes_df():
         return pd.read_parquet("data/codes.parq")
     except Exception as e:
         logger.error(f"Failed to load ticker mapping data (data/codes.parq): {e}")
+        
         # Consider if this should raise a more specific error or if ValueError is appropriate
         raise ValueError(f"Cannot perform ticker mapping due to data loading failure: {e}")
 

sovai-0.2.68/sovai/utils/file_management.py

@@ -0,0 +1,58 @@
+import pandas as pd
+import os
+import time
+from datetime import timedelta
+from pathlib import Path
+
+# Cache configuration
+CACHE_DIR = Path("data")
+CACHE_TTL_SECONDS = timedelta(days=7).total_seconds()
+
+def _get_file_age_seconds(file_path: Path) -> float | None:
+    """Return file age in seconds, or None if it doesn't exist."""
+    try:
+        return time.time() - file_path.stat().st_mtime
+    except (FileNotFoundError, OSError):
+        return None
+
+
+def _cached_parquet(url: str, filename: str) -> pd.DataFrame:
+    """
+    Fetch a Parquet file from a URL into a local cache directory if missing or expired,
+    then load and return it as a pandas DataFrame.
+    """
+    CACHE_DIR.mkdir(parents=True, exist_ok=True)
+    path = CACHE_DIR / filename
+    age = _get_file_age_seconds(path)
+    if age is None or age > CACHE_TTL_SECONDS:
+        print(f"Downloading {url} to {path}")
+        df = pd.read_parquet(url)
+        df.to_parquet(path)
+    else:
+        print(f"Loading from cache: {path}")
+        df = pd.read_parquet(path)
+    return df
+
+
+def get_ticker_meta() -> pd.DataFrame:
+    """Return the tickers metadata, cached for 7 days."""
+    return _cached_parquet(
+        url="https://storage.googleapis.com/sovai-public/accounting/tickers_transformed.parq",
+        filename="tickers.parq"
+    )
+
+
+def get_codes_meta() -> pd.DataFrame:
+    """Return the codes metadata, cached for 7 days."""
+    return _cached_parquet(
+        url="https://storage.googleapis.com/sovai-public/sovai-master/output/df_codes.parquet",
+        filename="codes.parq"
+    )
+
+
+def update_data_files():
+    """
+    Manually pre-warm both ticker and code caches.
+    """
+    get_ticker_meta()
+    get_codes_meta()