integrate database writing procedures for logging purposes

src/delta_barth/analysis/forecast.py

@@ -3,16 +3,19 @@ from __future__ import annotations
 import datetime
 import math
 from collections.abc import Mapping, Set
+from dataclasses import asdict
 from datetime import datetime as Datetime
 from typing import TYPE_CHECKING, Final, TypeAlias, cast
 
 import numpy as np
 import pandas as pd
 import scipy.stats
+import sqlalchemy as sql
 from sklearn.metrics import mean_absolute_error, r2_score
 from sklearn.model_selection import KFold, RandomizedSearchCV
 from xgboost import XGBRegressor
 
+from delta_barth import databases
 from delta_barth.analysis import parse
 from delta_barth.api.requests import (
     SalesPrognosisResponse,
@@ -29,7 +32,8 @@ from delta_barth.constants import (
     SALES_MIN_NUM_DATAPOINTS,
 )
 from delta_barth.errors import STATUS_HANDLER, wrap_result
-from delta_barth.logging import logger_pipelines as logger
+from delta_barth.logging import logger_db, logger_pipelines
+from delta_barth.management import SESSION
 from delta_barth.types import (
     BestParametersXGBRegressor,
     DualDict,
@@ -77,6 +81,21 @@ def _parse_df_to_results(
     return SalesPrognosisResults(daten=tuple(df_formatted))  # type: ignore
 
 
+def _write_sales_forecast_stats(
+    stats: SalesForecastStatistics,
+) -> None:
+    stats_db = asdict(stats)
+    _ = stats_db.pop("xgb_params")
+    xgb_params = stats.xgb_params
+
+    with SESSION.db_engine.begin() as conn:
+        res = conn.execute(sql.insert(databases.sf_stats).values(stats_db))
+        sf_id = cast(int, res.inserted_primary_key[0])  # type: ignore
+        if xgb_params is not None:
+            xgb_params["forecast_id"] = sf_id
+            conn.execute(sql.insert(databases.sf_XGB).values(xgb_params))
+
+
 @wrap_result()
 def _parse_api_resp_to_df_wrapped(
     resp: SalesPrognosisResponse,
@@ -91,23 +110,11 @@ def _parse_df_to_results_wrapped(
     return _parse_df_to_results(data)
 
 
-# ------------------------------------------------------------------------------
-# Input:
-#   DataFrame df mit Columns f_umsatz_fakt, firmen, art, v_warengrp
-#   kunde (muss enthalten sein in df['firmen']['firma_refid'])
-
-# Output:
-#   Integer umsetzung (Prognose möglich): 0 ja, 1 nein (zu wenig Daten verfügbar),
-# 2 nein (Daten nicht für Prognose geeignet)
-#   DataFrame test: Jahr, Monat, Vorhersage
-# -------------------------------------------------------------------------------
-
-
-# Prognose Umsatz je Firma
-
-
-# TODO: check usage of separate exception and handle it in API function
-# TODO set min number of data points as constant, not parameter
+@wrap_result()
+def _write_sales_forecast_stats_wrapped(
+    stats: SalesForecastStatistics,
+) -> None:
+    return _write_sales_forecast_stats(stats)
 
 
 def _preprocess_sales(
@@ -341,7 +348,7 @@ def _export_on_fail(
     return SalesPrognosisResultsExport(response=response, status=status)
 
 
-def pipeline_sales(
+def pipeline_sales_forecast(
     session: Session,
     company_id: int | None = None,
     start_date: Datetime | None = None,
@@ -352,8 +359,8 @@ def pipeline_sales(
         start_date=start_date,
     )
     if status != STATUS_HANDLER.SUCCESS:
-        logger.error(
-            "Error during sales prognosis data retrieval, Status: %s",
+        logger_pipelines.error(
+            "Error during sales forecast data retrieval, Status: %s",
             status,
             stack_info=True,
         )
@@ -365,8 +372,8 @@ def pipeline_sales(
         target_features=FEATURES_SALES_PROGNOSIS,
     )
     if pipe.status != STATUS_HANDLER.SUCCESS:
-        logger.error(
-            "Error during sales prognosis preprocessing, Status: %s",
+        logger_pipelines.error(
+            "Error during sales forecast preprocessing, Status: %s",
             pipe.status,
             stack_info=True,
         )
@@ -377,9 +384,16 @@ def pipeline_sales(
         min_num_data_points=SALES_MIN_NUM_DATAPOINTS,
         base_num_data_points_months=SALES_BASE_NUM_DATAPOINTS_MONTHS,
     )
+    if pipe.statistics is not None:
+        res = _write_sales_forecast_stats_wrapped(pipe.statistics)
+        if res.status != STATUS_HANDLER.SUCCESS:
+            logger_db.error(
+                "[DB] Error during write process of sales forecast statistics: %s",
+                res.status,
+            )
     if pipe.status != STATUS_HANDLER.SUCCESS:
-        logger.error(
-            "Error during sales prognosis main processing, Status: %s",
+        logger_pipelines.error(
+            "Error during sales forecast main processing, Status: %s",
             pipe.status,
             stack_info=True,
         )
@@ -390,8 +404,8 @@ def pipeline_sales(
         feature_map=DualDict(),
     )
     if pipe.status != STATUS_HANDLER.SUCCESS:
-        logger.error(
-            "Error during sales prognosis postprocessing, Status: %s",
+        logger_pipelines.error(
+            "Error during sales forecast postprocessing, Status: %s",
             pipe.status,
             stack_info=True,
         )

src/delta_barth/logging.py

@@ -34,8 +34,10 @@ logger_session = logging.getLogger("delta_barth.session")
 logger_session.setLevel(logging.DEBUG)
 logger_wrapped_results = logging.getLogger("delta_barth.wrapped_results")
 logger_wrapped_results.setLevel(logging.DEBUG)
-logger_pipelines = logging.getLogger("delta_barth.logger_pipelines")
+logger_pipelines = logging.getLogger("delta_barth.pipelines")
 logger_pipelines.setLevel(logging.DEBUG)
+logger_db = logging.getLogger("delta_barth.databases")
+logger_db.setLevel(logging.DEBUG)
 
 
 def setup_logging(

src/delta_barth/pipelines.py

@@ -11,7 +11,9 @@ def pipeline_sales_forecast(
     company_id: int | None,
     start_date: Datetime | None,
 ) -> JsonExportResponse:
-    result = forecast.pipeline_sales(SESSION, company_id=company_id, start_date=start_date)
+    result = forecast.pipeline_sales_forecast(
+        SESSION, company_id=company_id, start_date=start_date
+    )
     export = JsonExportResponse(result.model_dump_json())
 
     return export

src/delta_barth/types.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import enum
+import pprint
 import typing as t
 from collections.abc import Sequence
 from dataclasses import dataclass, field
@@ -29,6 +30,10 @@ class Status(BaseModel):
     message: SkipValidation[str] = ""
     api_server_error: SkipValidation[DelBarApiError | None] = None
 
+    def __str__(self) -> str:
+        py_repr = self.model_dump()
+        return pprint.pformat(py_repr, indent=4, sort_dicts=False)
+
 
 class ResponseType(BaseModel):
     pass