当窗口/分区使用正向填充时，将条件添加到pyspark sql中的last()函数

+---+-----+-----+
| id| name|value|
+---+-----+-----+
|  1| John| null|
|  1| Mike|   10|
|  1| Chris| null|
|  2|  Amy| null|
|  2| Lily|   20|
+---+-----+-----+

from pyspark.sql import Window
from pyspark.sql.functions import last, col

# 定义窗口
window_spec = Window.partitionBy("id").orderBy("name").rowsBetween(Window.unboundedPreceding, 0)

# 使用last函数进行正向填充并添加条件
df = df.withColumn("filled_value", last(col("value"), ignorenulls=True).over(window_spec))

# 筛选结果，只选择value不为空的行
df = df.filter(col("value").isNotNull())

df.show()

+---+-----+-----+------------+
| id| name|value|filled_value|
+---+-----+-----+------------+
|  1| John| null|        null|
|  1| Mike|   10|          10|
|  1| Chris| null|          10|
|  2|  Amy| null|        null|
|  2| Lily|   20|          20|
+---+-----+-----+------------+