etl_Property_capacities_weekdays.py eingefügt

Abfragemöglichkeit für die Wochentage eingefügt

etl/src/api/main.py

@@ -2,6 +2,7 @@ import data
 import polars as pl
 from data import etl_property_capacities as etl_pc
 from data import etl_property_capacities_monthly as etl_pcm
+from data import etl_property_capacities_weekdays as etl_pcw
 from data import etl_region_capacities as etl_rc
 from data import etl_region_capacities_comparison as etl_rcc
 from fastapi import FastAPI, Response
@@ -48,6 +49,11 @@ def property_capacities_data(id: int, scrapeDate: str):
 	capacities = etl_pcm.property_capacities_monthly(id, scrapeDate)
 	return capacities
 
+@app.get("/property/{id}/capacities/weekdays/{scrapeDate}")
+def property_capacities_data(id: int, scrapeDate: str):
+	capacities = etl_pcw.property_capacities_weekdays(id, scrapeDate)
+	return capacities
+
 @app.get("/property/{id}/base")
 def property_base_data(id: int):
 	return d.property_base_data(id).pl().to_dicts()

etl/src/data/etl_property_capacities_weekdays.py

@@ -0,0 +1,30 @@
+from io import StringIO
+
+import polars as pl
+
+import data
+
+d = data.load()
+
+def property_capacities_weekdays(id: int, scrapeDate: str):
+	extractions = d.extractions_propId_scrapeDate(id, scrapeDate).pl()
+	df_calendar = pl.DataFrame()
+	numWeeks = 0
+
+	for row in extractions.rows(named=True):
+		scrapeDate = row['created_at']
+		df_calendar = pl.read_json(StringIO(row['calendar']))
+		columnTitles = df_calendar.columns
+		df_calendar = df_calendar.transpose()
+		df_calendar = df_calendar.with_columns(pl.Series(name="dates", values=columnTitles))
+		df_calendar = df_calendar.with_columns((pl.col("dates").str.to_date()))
+		numWeeks = round((df_calendar.get_column("dates").max() - df_calendar.get_column("dates").min()).days / 7, 0)
+		df_calendar = df_calendar.with_columns(pl.col("dates").dt.weekday().alias("weekday"))
+		df_calendar = df_calendar.drop("dates")
+
+	df_calendar = df_calendar.group_by("weekday").agg(pl.col("column_0").sum())
+	df_calendar = df_calendar.with_columns((pl.col("column_0") / numWeeks).alias("weekday"))
+	result = {"scraping-date": scrapeDate, "weekday": df_calendar['weekday'].to_list(), 'capacities': df_calendar['column_0'].to_list()}
+	return result
+
+