Revision history [back]

Here is the procedure for reading and parsing a JSON file from an ADLS Gen2 storage account type using Java:

1. Create a storage account instance that points to your ADLS Gen2 storage account.

   StorageAccount storageAccount = StorageAccount.fromConnectionString("<storage connection string>");
   

2. Create a DataLakeStorageFileSystemManagementClient instance that will be used to access the file system.

   DataLakeStorageFileSystemManagementClient fileSystemClient = new DataLakeStorageFileSystemManagementClientBuilder()
          .endpoint("<file system endpoint>")
          .credential(new DefaultAzureCredentialBuilder().build())
          .buildClient();
   

3. Use the fileSystemClient to download the JSON file from your storage account into a ByteBuffer object.

   ByteBuffer buffer = fileSystemClient.readFile("<file system name>", "<file path>");
   

4. Convert the ByteBuffer into a String.

   String json = new String(buffer.array());
   

5. Use a JSON parser library, such as Jackson or GSON, to parse the JSON string into Java objects.

Here is the procedure for reading and parsing a JSON file from an ADLS Gen2 storage account type using Java with Spark:

1. Create a SparkContext instance.

   SparkConf conf = new SparkConf()
          .set("spark.hadoop.fs.adl.oauth2.access.token.provider.type", "ClientCredential")
          .setMaster("local[*]")
          .setAppName("<your app name>");
   SparkContext sparkContext = new SparkContext(conf);
   

2. Use the SparkContext to create a RDD of JSON strings from your ADLS Gen2 storage account.

   JavaRDD<String> jsonRDD = sparkContext.textFile("<ADLS Gen2 file path>");
   

3. Use a JSON parsing library, such as Jackson or GSON, to parse the JSON strings into Java objects.

   ObjectMapper objectMapper = new ObjectMapper();
   JavaRDD<MyObject> myObjectRDD = jsonRDD.map(json -> objectMapper.readValue(json, MyObject.class));