AWS Glue，输出一个带分区的文件

Question

我有一个 Glue ETL 脚本，它采用分区的 Athena 表并将其输出到 CSV。该表按两个标准进行分区，单元和站点。当 Glue 作业运行时，它会为单元和站点分区的每个组合创建一个不同的 CSV 文件。相反，我只想要一个包含所有分区的输出文件，类似于 athena 表的结构

我已经对“datasource0.toDF().repartition(1)”进行了一些修改，但我不确定它如何与 AWS 提供的脚本交互。我已经用镶木地板做了这个，但是这个脚本的结构不同

请注意下面的脚本我已经删除了大部分标签映射

from awsglue.transforms import *
from awsglue.utils import getResolvedOptions
from pyspark.context import SparkContext
from awsglue.context import GlueContext
from awsglue.job import Job

## @params: [JOB_NAME]
args = getResolvedOptions(sys.argv, ['JOB_NAME'])

sc = SparkContext()
glueContext = GlueContext(sc)
spark = glueContext.spark_session
job = Job(glueContext)
job.init(args['JOB_NAME'], args)
## @type: DataSource
## @args: [database = "testdata-2018-2019", table_name = "testdata", transformation_ctx = "datasource0"]
## @return: datasource0
## @inputs: []
datasource0 = glueContext.create_dynamic_frame.from_catalog(database = "formatted-test-2018-2019", table_name = "testdata", transformation_ctx = "datasource0")
datasource0.toDF().repartition(1)
## @type: ApplyMapping
## @args: [mapping = [("time", "string", "time", "string"), ("unit", "string", "unit", "string")], transformation_ctx = "applymapping1"]
## @return: applymapping1
## @inputs: [frame = datasource0]
applymapping1 = ApplyMapping.apply(frame = datasource0, mappings = [("time", "string", "time", "string"), ("`data.pv`", "double", ("site", "string", "site", "string"), ("unit", "string", "unit", "string")], transformation_ctx = "applymapping1")
## @type: DataSink
## @args: [connection_type = "s3", connection_options = {"path": "s3://testbucket/ParsedCSV-Data"}, format = "csv", transformation_ctx = "datasink2"]
## @return: datasink2
## @inputs: [frame = applymapping1]
datasink2 = glueContext.write_dynamic_frame.from_options(frame = applymapping1, connection_type = "s3", connection_options = {"path": "s3://buckettest/ParsedCSV-Data"}, format = "csv", transformation_ctx = "datasink2").repartition(1)
job.commit()

我想修改上述脚本以仅输出一个包含分区列的 CSV 文件。我该怎么做？

Answer 1

您需要在编写 DynamicFrame 之前重新分区。

repartitioned1 = applymapping1.repartition(1)
datasink2 = glueContext.write_dynamic_frame.from_options(frame = repartitioned1, connection_type = "s3", connection_options = {"path": "s3://20182019testdata/ParsedCSV-Data"}, format = "csv", transformation_ctx = "datasink2")

关于在输出文件中包含分区列，我认为这是不可能的。作为一种解决方法，您可以将一列复制到具有不同名称的新列中。

df = applymapping1.toDF
repartitioned_with_new_column_df = df.withColumn("_column1", df["column1"]).repartition(1)
dyf = DynamicFrame.fromDF(repartitioned_with_new_column_df, glueContext, "enriched")
datasink2 = glueContext.write_dynamic_frame.from_options(frame = dyf, connection_type = "s3", connection_options = {"path": "s3://20182019testdata/ParsedCSV-Data", , "partitionKeys": ["_column1"]}, format = "csv", transformation_ctx = "datasink2")

Answer 2

作为aws support。您可以使用.coalesce(1)。像这样的东西：

dynamic_Frame=applymapping1.coalesce(1)
datasink2 = glueContext.write_dynamic_frame.from_options(frame = dynamic_Frame, connection_type = "s3", connection_options = 

它适用于我的情况。