chore: Update storage to store backtest results as zip file

this will enable further additions to files without polluting the filesystem further.

freqtrade/optimize/optimize_reports/bt_storage.py

@@ -1,4 +1,6 @@
 import logging
+import zipfile
+from io import BytesIO, StringIO
 from pathlib import Path
 from typing import Any, TextIO
 
@@ -7,7 +9,7 @@ from pandas import DataFrame
 from freqtrade.constants import LAST_BT_RESULT_FN
 from freqtrade.enums.runmode import RunMode
 from freqtrade.ft_types import BacktestResultType
-from freqtrade.misc import file_dump_json
+from freqtrade.misc import dump_json_to_file, file_dump_json
 from freqtrade.optimize.backtest_caching import get_backtest_metadata_filename
 
 
@@ -52,75 +54,59 @@ def store_backtest_results(
     analysis_results: dict[str, dict[str, DataFrame]] | None = None,
 ) -> Path:
     """
-    Stores backtest results and analysis data
+    Stores backtest results and analysis data in a zip file, with metadata stored separately
+    for convenience.
     :param config: Configuration dictionary
     :param stats: Dataframe containing the backtesting statistics
     :param dtappendix: Datetime to use for the filename
     :param market_change_data: Dataframe containing market change data
     :param analysis_results: Dictionary containing analysis results
     """
-
-    # Path object, which can either be a filename or a directory.
-    # Filenames will be appended with a timestamp right before the suffix
-    # while for directories, <directory>/backtest-result-<datetime>.json will be used as filename
     recordfilename: Path = config["exportfilename"]
-    filename = _generate_filename(recordfilename, dtappendix, ".json")
+    zip_filename = _generate_filename(recordfilename, dtappendix, ".zip")
+    base_filename = _generate_filename(recordfilename, dtappendix, "")
+    json_filename = _generate_filename(recordfilename, dtappendix, ".json")
 
-    # Store metadata separately.
-    file_dump_json(get_backtest_metadata_filename(filename), stats["metadata"])
-    # Don't mutate the original stats dict.
+    # Store metadata separately with .json extension
+    file_dump_json(get_backtest_metadata_filename(json_filename), stats["metadata"])
+
+    # Store latest backtest info separately
+    latest_filename = Path.joinpath(zip_filename.parent, LAST_BT_RESULT_FN)
+    file_dump_json(latest_filename, {"latest_backtest": str(zip_filename.name)})
+
+    # Create zip file and add the files
+    with zipfile.ZipFile(zip_filename, "w", zipfile.ZIP_DEFLATED) as zipf:
+        # Store stats
         stats_copy = {
             "strategy": stats["strategy"],
             "strategy_comparison": stats["strategy_comparison"],
         }
+        stats_buf = StringIO()
+        dump_json_to_file(stats_buf, stats_copy)
+        zipf.writestr(json_filename, stats_buf.getvalue())
 
-    file_dump_json(filename, stats_copy)
-
-    latest_filename = Path.joinpath(filename.parent, LAST_BT_RESULT_FN)
-    file_dump_json(latest_filename, {"latest_backtest": str(filename.name)})
-
+        # Add market change data if present
         if market_change_data is not None:
-        filename_mc = _generate_filename(recordfilename, f"{dtappendix}_market_change", ".feather")
+            market_change_name = f"{base_filename.stem}_market_change.feather"
+            market_change_buf = BytesIO()
             market_change_data.reset_index().to_feather(
-            filename_mc, compression_level=9, compression="lz4"
+                market_change_buf, compression_level=9, compression="lz4"
             )
+            market_change_buf.seek(0)
+            zipf.writestr(market_change_name, market_change_buf.getvalue())
 
+        # Add analysis results if present and running in backtest mode
         if (
             config.get("export", "none") == "signals"
             and analysis_results is not None
             and config.get("runmode", RunMode.OTHER) == RunMode.BACKTEST
         ):
-        _store_backtest_analysis_data(
-            recordfilename, analysis_results["signals"], dtappendix, "signals"
-        )
-        _store_backtest_analysis_data(
-            recordfilename, analysis_results["rejected"], dtappendix, "rejected"
-        )
-        _store_backtest_analysis_data(
-            recordfilename, analysis_results["exited"], dtappendix, "exited"
-        )
+            for name in ["signals", "rejected", "exited"]:
+                if name in analysis_results:
+                    analysis_name = f"{base_filename.stem}_{name}.pkl"
+                    analysis_buf = BytesIO()
+                    file_dump_joblib(analysis_buf, analysis_results[name])
+                    analysis_buf.seek(0)
+                    zipf.writestr(analysis_name, analysis_buf.getvalue())
 
-    return filename
-
-
-def _store_backtest_analysis_data(
-    recordfilename: Path, data: dict[str, dict], dtappendix: str, name: str
-) -> Path:
-    """
-    Stores backtest trade candles for analysis
-    :param recordfilename: Path object, which can either be a filename or a directory.
-        Filenames will be appended with a timestamp right before the suffix
-        while for directories, <directory>/backtest-result-<datetime>_<name>.pkl will be used
-        as filename
-    :param candles: Dict containing the backtesting data for analysis
-    :param dtappendix: Datetime to use for the filename
-    :param name: Name to use for the file, e.g. signals, rejected
-    """
-    filename = _generate_filename(recordfilename, f"{dtappendix}_{name}", ".pkl")
-
-    logger.info(f'dumping joblib to "{filename}"')
-    with filename.open("wb") as fp:
-        file_dump_joblib(fp, data)
-    logger.debug(f'done joblib dump to "{filename}"')
-
-    return filename
+    return zip_filename