pySpark15 - date 타입 핸들링

# Date parsing
spark = SparkSession.builder.appName("df_manage_date").getOrCreate()
df = spark.read.csv(
    "file:///home/jovyan/work/sample/date_parsing.csv", header=True, inferSchema=True)
df.show()

+----------+------+
|      date|number|
+----------+------+
|2022-01-27|  2723|
|2021-12-29|  1460|
|2022-01-22|  3411|
|2022-01-06|  1527|
|2022-04-21|  3978|
|2022-10-23|  3443|
|2021-12-23|  1641|
|2022-05-31|  1633|
|2021-12-29|  1072|
|2021-12-30|  2936|
|2022-05-04|  2494|
|2022-06-22|  2019|
|2022-04-23|  3804|
|2022-08-04|  1619|
|2022-01-26|  1306|
|2022-09-23|  3918|
|2022-05-27|  3209|
|2022-09-20|  2333|
|2022-07-05|  1861|
|2022-07-18|  3404|
+----------+------+

show year

df.select(f.year('date')).show()

show month

df.select(f.month('date')).show()

show day

df.select(f.dayofmonth('date').alias('day')).show()
df.select(f.dayofyear('date').alias('day')).show()

df = df.withColumn("year", f.year('date')).groupBy("year").mean("number").withColumnRenamed("avg(number)", "avg")

df.show()

df.select("year", f.format_number("avg", 2).alias("avg")).show()