使用数据流运行器在beam.pipeline内部运行没有输入的函数

import apache_beam as beam

def my_function(element):
    # 你的函数逻辑
    return element

with beam.Pipeline() as p:
    result = (
        p
        | 'Create' >> beam.Create(['dummy'])
        | 'Apply Function' >> beam.Map(my_function)
    )

import apache_beam as beam

class MyDoFn(beam.DoFn):
    def process(self, element):
        # 你的函数逻辑
        yield element

with beam.Pipeline() as p:
    result = (
        p
        | 'Create' >> beam.Create(['dummy'])
        | 'Apply Function' >> beam.ParDo(MyDoFn())
    )

import apache_beam as beam

def my_function(elements):
    # 你的函数逻辑
    return elements

with beam.Pipeline() as p:
    result = (
        p
        | 'Create' >> beam.Create(['dummy'])
        | 'Apply Function' >> beam.CombineGlobally(my_function)
    )