在ADF数据流连接表达式中，如果要在表达式中使用单引号，可以使用两个单引号来表示一个单引号。以下是一个代码示例：

{
  "name": "CopyData",
  "type": "Copy",
  "inputs": [
    {
      "referenceName": "SourceDataset",
      "type": "DatasetReference"
    }
  ],
  "outputs": [
    {
      "referenceName": "DestinationDataset",
      "type": "DatasetReference"
    }
  ],
  "activities": [
    {
      "name": "CopyActivity",
      "type": "Copy",
      "inputs": [
        {
          "referenceName": "SourceDataset"
        }
      ],
      "outputs": [
        {
          "referenceName": "DestinationDataset"
        }
      ],
      "typeProperties": {
        "source": {
          "type": "BlobSource",
          "sourceRetryCount": 3,
          "sourceRetryWait": "00:00:05"
        },
        "sink": {
          "type": "BlobSink",
          "writeBatchSize": 10000,
          "writeBatchTimeout": "00:05:00"
        },
        "enableStaging": false,
        "translator": {
          "type": "TabularTranslator",
          "columnMappings": "column1: 'column1',column2: 'column2',column3: 'column3'"
        }
      }
    }
  ]
}

在这个示例中，columnMappings属性的值是column1: 'column1',column2: 'column2',column3: 'column3'。注意到每个列名和列值之间都有一个单引号包围，而每个单引号都使用两个单引号来表示。

这样，ADF就能正确解析并使用包含单引号的数据流连接表达式了。