我在DF2中有一个管道,该管道必须查看blob中的文件夹,然后将145个文件中的每个文件依次处理到数据库表中。将每个文件加载到表中之后,应触发存储过程,该存储过程将检查每个记录并将其插入,或将现有记录更新到主表中。
在线查看,我感觉好像已经尝试了建议的“ Get MetaData”,“ For Each”,“ LookUp”和“ Assign Variable”激活的每种组合,但是由于某种原因,我的“复制数据”仍然可以拾取所有文件同时运行145次。
最近在网上找到了一个博客,我关注该博客使用“分配变量”,因为它对多个文件位置很有用,但对我不起作用。我需要将CSV文件读取为表而不是二进制对象,因此我认为这是我的问题。
{
"name": "BulkLoadPipeline",
"properties": {
"activities": [
{
"name": "GetFileNames",
"type": "GetMetadata",
"policy": {
"timeout": "7.00:00:00",
"retry": 0,
"retryIntervalInSeconds": 30,
"secureOutput": false,
"secureInput": false
},
"typeProperties": {
"dataset": {
"referenceName": "DelimitedText1",
"type": "DatasetReference",
"parameters": {
"fileName": "@item()"
}
},
"fieldList": [
"childItems"
],
"storeSettings": {
"type": "AzureBlobStorageReadSetting"
},
"formatSettings": {
"type": "DelimitedTextReadSetting"
}
}
},
{
"name": "CopyDataRunDeltaCheck",
"type": "ForEach",
"dependsOn": [
{
"activity": "BuildList",
"dependencyConditions": [
"Succeeded"
]
}
],
"typeProperties": {
"items": {
"value": "@variables('fileList')",
"type": "Expression"
},
"isSequential": true,
"activities": [
{
"name": "WriteToTables",
"type": "Copy",
"policy": {
"timeout": "7.00:00:00",
"retry": 0,
"retryIntervalInSeconds": 30,
"secureOutput": false,
"secureInput": false
},
"typeProperties": {
"source": {
"type": "DelimitedTextSource",
"storeSettings": {
"type": "AzureBlobStorageReadSetting",
"wildcardFileName": "*.*"
},
"formatSettings": {
"type": "DelimitedTextReadSetting"
}
},
"sink": {
"type": "AzureSqlSink"
},
"enableStaging": false,
"translator": {
"type": "TabularTranslator",
"mappings": [
{
"source": {
"name": "myID",
"type": "String"
},
"sink": {
"name": "myID",
"type": "String"
}
},
{
"source": {
"name": "Col1",
"type": "String"
},
"sink": {
"name": "Col1",
"type": "String"
}
},
{
"source": {
"name": "Col2",
"type": "String"
},
"sink": {
"name": "Col2",
"type": "String"
}
},
{
"source": {
"name": "Col3",
"type": "String"
},
"sink": {
"name": "Col3",
"type": "String"
}
},
{
"source": {
"name": "Col4",
"type": "String"
},
"sink": {
"name": "Col4",
"type": "String"
}
},
{
"source": {
"name": "DW Date Created",
"type": "String"
},
"sink": {
"name": "DW_Date_Created",
"type": "String"
}
},
{
"source": {
"name": "DW Date Updated",
"type": "String"
},
"sink": {
"name": "DW_Date_Updated",
"type": "String"
}
}
]
}
},
"inputs": [
{
"referenceName": "DelimitedText1",
"type": "DatasetReference",
"parameters": {
"fileName": "@item()"
}
}
],
"outputs": [
{
"referenceName": "myTable",
"type": "DatasetReference"
}
]
},
{
"name": "CheckDeltas",
"type": "SqlServerStoredProcedure",
"dependsOn": [
{
"activity": "WriteToTables",
"dependencyConditions": [
"Succeeded"
]
}
],
"policy": {
"timeout": "7.00:00:00",
"retry": 0,
"retryIntervalInSeconds": 30,
"secureOutput": false,
"secureInput": false
},
"typeProperties": {
"storedProcedureName": "[TL].[uspMyCheck]"
},
"linkedServiceName": {
"referenceName": "myService",
"type": "LinkedServiceReference"
}
}
]
}
},
{
"name": "BuildList",
"type": "ForEach",
"dependsOn": [
{
"activity": "GetFileNames",
"dependencyConditions": [
"Succeeded"
]
}
],
"typeProperties": {
"items": {
"value": "@activity('GetFileNames').output.childItems",
"type": "Expression"
},
"isSequential": true,
"activities": [
{
"name": "Create list from variables",
"type": "AppendVariable",
"typeProperties": {
"variableName": "fileList",
"value": "@item().name"
}
}
]
}
}
],
"variables": {
"fileList": {
"type": "Array"
}
}
}
}
管线输出的“详细信息”屏幕显示了Blob中项目数量的管道循环,但是每次都针对列表中的每个文件一次运行“复制数据”和“存储过程”,而不是一次运行一次。
我觉得我已经接近答案了,但缺少一个重要的部分。任何帮助或建议都将不胜感激。