在pyspark数据帧中生成序列，以便在空值之后找到值时递增

from pyspark.sql import SparkSession
from pyspark.sql.functions import monotonically_increasing_id

spark = SparkSession.builder.getOrCreate()

df = spark.read.format("csv").options(header=True).load("your_file.csv")

df = df.withColumn("sequence", monotonically_increasing_id())

from pyspark.sql.window import Window
from pyspark.sql.functions import when, col

# 创建窗口函数，用于排序和行号
window = Window.orderBy("sequence")

# 找到空值后的非空值并递增
df = df.withColumn("sequence_increment", when(col("values").isNull(), None).otherwise(row_number().over(window)))

from pyspark.sql import SparkSession
from pyspark.sql.functions import monotonically_increasing_id, when, col
from pyspark.sql.window import Window

# 创建Spark会话
spark = SparkSession.builder.getOrCreate()

# 加载数据到数据帧
df = spark.read.format("csv").options(header=True).load("your_file.csv")

# 添加一个递增列
df = df.withColumn("sequence", monotonically_increasing_id())

# 创建窗口函数，用于排序和行号
window = Window.orderBy("sequence")

# 找到空值后的非空值并递增
df = df.withColumn("sequence_increment", when(col("values").isNull(), None).otherwise(row_number().over(window)))

# 显示结果
df.show()