org.apache.hadoop.hbase.client

Interface HTableInterface

All Superinterfaces:

AutoCloseable, Closeable, Table

All Known Implementing Classes:

HTable

Deprecated.

use Table instead

@Deprecated
@InterfaceAudience.Private
@InterfaceStability.Stable
public interface HTableInterface
extends Table

Used to communicate with a single HBase table. Obtain an instance from an HConnection.

Since:

0.21.0

Method Summary

Methods

Modifier and Type	Method and Description
Boolean[]	exists(List<Get> gets) Deprecated. Use Table.existsAll(java.util.List) instead.
Result	getRowOrBefore(byte[] row, byte[] family) Deprecated. As of version 0.92 this method is deprecated without replacement. Since version 0.96+, you can use reversed scan. getRowOrBefore is used internally to find entries in hbase:meta and makes various assumptions about the table (which are true for hbase:meta but not in general) to be efficient.
byte[]	getTableName() Deprecated. Use Table.getName() instead
long	incrementColumnValue(byte[] row, byte[] family, byte[] qualifier, long amount, boolean writeToWAL) Deprecated. Use Table.incrementColumnValue(byte[], byte[], byte[], long, Durability)
void	setAutoFlush(boolean autoFlush) Deprecated. in 0.96. When called with setAutoFlush(false), this function also set clearBufferOnFail to true, which is unexpected but kept for historical reasons. Replace it with setAutoFlush(false, false) if this is exactly what you want, or by Table.setAutoFlushTo(boolean) for all other cases.
void	setAutoFlush(boolean autoFlush, boolean clearBufferOnFail) Deprecated. in 0.99 since setting clearBufferOnFail is deprecated. Use Table.setAutoFlushTo(boolean)} instead.

Methods inherited from interface org.apache.hadoop.hbase.client.Table

append, batch, batch, batchCallback, batchCallback, batchCoprocessorService, batchCoprocessorService, checkAndDelete, checkAndDelete, checkAndMutate, checkAndPut, checkAndPut, close, coprocessorService, coprocessorService, coprocessorService, delete, delete, exists, existsAll, flushCommits, get, get, getConfiguration, getName, getScanner, getScanner, getScanner, getTableDescriptor, getWriteBufferSize, increment, incrementColumnValue, incrementColumnValue, isAutoFlush, mutateRow, put, put, setAutoFlushTo, setWriteBufferSize

Method Detail

getTableName

@Deprecated
byte[] getTableName()

Deprecated. Use Table.getName() instead

Gets the name of this table.

Returns:

the table name.

incrementColumnValue

@Deprecated
long incrementColumnValue(byte[] row,
                                   byte[] family,
                                   byte[] qualifier,
                                   long amount,
                                   boolean writeToWAL)
                          throws IOException

Deprecated. Use Table.incrementColumnValue(byte[], byte[], byte[], long, Durability)

Throws:

IOException

exists

@Deprecated
Boolean[] exists(List<Get> gets)
                 throws IOException

Deprecated. Use Table.existsAll(java.util.List) instead.

Throws:

IOException

setAutoFlush

@Deprecated
void setAutoFlush(boolean autoFlush)

Deprecated. in 0.96. When called with setAutoFlush(false), this function also set clearBufferOnFail to true, which is unexpected but kept for historical reasons. Replace it with setAutoFlush(false, false) if this is exactly what you want, or by Table.setAutoFlushTo(boolean) for all other cases.

See setAutoFlush(boolean, boolean)

Parameters:

autoFlush - Whether or not to enable 'auto-flush'.

setAutoFlush

@Deprecated
void setAutoFlush(boolean autoFlush,
                           boolean clearBufferOnFail)

Deprecated. in 0.99 since setting clearBufferOnFail is deprecated. Use Table.setAutoFlushTo(boolean)} instead.

Turns 'auto-flush' on or off.

When enabled (default), Put operations don't get buffered/delayed and are immediately executed. Failed operations are not retried. This is slower but safer.

Turning off #autoFlush means that multiple Puts will be accepted before any RPC is actually sent to do the write operations. If the application dies before pending writes get flushed to HBase, data will be lost.

When you turn #autoFlush off, you should also consider the #clearBufferOnFail option. By default, asynchronous Put requests will be retried on failure until successful. However, this can pollute the writeBuffer and slow down batching performance. Additionally, you may want to issue a number of Put requests and call Table.flushCommits() as a barrier. In both use cases, consider setting clearBufferOnFail to true to erase the buffer after Table.flushCommits() has been called, regardless of success.

In other words, if you call #setAutoFlush(false); HBase will retry N time for each flushCommit, including the last one when closing the table. This is NOT recommended, most of the time you want to call #setAutoFlush(false, true).

Parameters:

autoFlush - Whether or not to enable 'auto-flush'.

clearBufferOnFail - Whether to keep Put failures in the writeBuffer. If autoFlush is true, then the value of this parameter is ignored and clearBufferOnFail is set to true. Setting clearBufferOnFail to false is deprecated since 0.96.

See Also:

Table.flushCommits()

getRowOrBefore

@Deprecated
Result getRowOrBefore(byte[] row,
                               byte[] family)
                      throws IOException

Deprecated. As of version 0.92 this method is deprecated without replacement. Since version 0.96+, you can use reversed scan. getRowOrBefore is used internally to find entries in hbase:meta and makes various assumptions about the table (which are true for hbase:meta but not in general) to be efficient.

Return the row that matches row exactly, or the one that immediately precedes it.

Parameters:

row - A row key.

family - Column family to include in the Result.

Throws:

IOException - if a remote or network exception occurs.

Since:

0.20.0