com.aliyun.odps.storage.read

Class TableReadSession

java.lang.Object

com.aliyun.odps.storage.read.TableReadSession

All Implemented Interfaces:

AutoCloseable

public class TableReadSession
extends Object
implements AutoCloseable

Implementation of TableReadSession for the MaxCompute Storage API.

This class represents a read session for a MaxCompute table, allowing clients to read data in a distributed manner using input splits. Each session is associated with a specific table and provides methods to get the session ID, Arrow schema, and input splits for parallel processing.

Example usage:

 TableIdentifier tableId = TableIdentifier.of("my_project", "my_table");
 try (TableReadSession session = client.createReadSessionBuilder(tableId).build()) {
     List<InputSplit> splits = session.getSplits();
     for (InputSplit split : splits) {
         try (ArrowReader reader = session.createArrowReaderBuilder(split).build()) {
             // Process data from reader
         }
     }
 }
 

Constructor Summary

Constructors

Constructor and Description
TableReadSession(StorageStub storageStub, TableIdentifier tableId, org.apache.arrow.memory.BufferAllocator allocator, CreateTableReadSessionResponse response, TableReadSessionBuilder builder) Constructs a new TableReadSession with the provided parameters.

Method Summary

Modifier and Type	Method and Description
void	close()
TableReaderBuilder	createReaderBuilder(InputSplit split) Creates a new builder for an Arrow reader for the specified input split.
org.apache.arrow.vector.types.pojo.Schema	getArrowSchema() Gets the Arrow schema for the data in this read session.
String	getId() Gets the unique identifier of this read session.
List<InputSplit>	getSplits() Gets the input splits for this read session.
TableSchema	getTableSchema()

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

TableReadSession

public TableReadSession(StorageStub storageStub,
                        TableIdentifier tableId,
                        org.apache.arrow.memory.BufferAllocator allocator,
                        CreateTableReadSessionResponse response,
                        TableReadSessionBuilder builder)

Constructs a new TableReadSession with the provided parameters.

Parameters:

storageStub - The storage stub for communicating with the MaxCompute service

tableId - The identifier of the table to read from

allocator - The buffer allocator for Arrow memory management

response - The response from the create table read session API call

Method Detail

getId

public String getId()

Gets the unique identifier of this read session.

Returns:

The session ID

getArrowSchema

public org.apache.arrow.vector.types.pojo.Schema getArrowSchema()

Gets the Arrow schema for the data in this read session.

This method converts the MaxCompute table schema to an Arrow schema, which is used for efficient data processing and columnar format support.

Returns:

The Arrow schema representation of the table data

getTableSchema

public TableSchema getTableSchema()

getSplits

public List<InputSplit> getSplits()

Gets the input splits for this read session.

Input splits are used to divide the data into chunks that can be processed in parallel. Each split represents a portion of the data that can be read independently using an Arrow reader.

Returns:

A list of input splits for parallel processing

createReaderBuilder

public TableReaderBuilder createReaderBuilder(InputSplit split)

Creates a new builder for an Arrow reader for the specified input split.

This method initializes an Arrow reader builder that can be configured with various options before creating the actual Arrow reader. The reader allows reading data from a specific input split in Arrow format.

Parameters:

split - The input split to read data from

Returns:

A new TableReaderBuilder instance to configure the Arrow reader

close

public void close()

Specified by:

close in interface AutoCloseable