如何在PySpark中基于条件计算窗口聚合上的distinct？

from pyspark.sql import SparkSession
from pyspark.sql.window import Window
from pyspark.sql.functions import col, countDistinct

spark = SparkSession.builder.getOrCreate()

data = [(1, 'A'), (1, 'B'), (2, 'A'), (2, 'C'), (3, 'B')]
df = spark.createDataFrame(data, ['id', 'value'])

window_spec = Window.partitionBy('id').orderBy('value')

distinct_values = df.select('id', 'value', countDistinct('value').over(window_spec).alias('distinct_count'))

distinct_values.show()

+---+-----+--------------+
| id|value|distinct_count|
+---+-----+--------------+
|  1|    A|             1|
|  1|    B|             2|
|  2|    A|             1|
|  2|    C|             2|
|  3|    B|             1|
+---+-----+--------------+