Package org.apache.hadoop.hbase.client

Class AsyncTableImpl

java.lang.Object

org.apache.hadoop.hbase.client.AsyncTableImpl

All Implemented Interfaces:

AsyncTable<ScanResultConsumer>

@Private
class AsyncTableImpl
extends Object
implements AsyncTable<ScanResultConsumer>

Just a wrapper of RawAsyncTableImpl. The difference is that users need to provide a thread pool when constructing this class, and the callback methods registered to the returned CompletableFuture will be executed in this thread pool. So usually it is safe for users to do anything they want in the callbacks without breaking the rpc framework.

Nested Class Summary

Nested classes/interfaces inherited from interface org.apache.hadoop.hbase.client.AsyncTable

AsyncTable.CheckAndMutateBuilder, AsyncTable.CheckAndMutateWithFilterBuilder, AsyncTable.CoprocessorCallback<R>, AsyncTable.CoprocessorServiceBuilder<S,R>, AsyncTable.PartialResultCoprocessorCallback<S,R>

Field Summary

Fields

Modifier and Type	Field	Description
private final ExecutorService	pool
private final RawAsyncTableImpl	rawTable

Constructor Summary

Constructors

Constructor	Description
AsyncTableImpl(RawAsyncTableImpl rawTable, ExecutorService pool)

Method Summary

Modifier and Type	Method	Description
CompletableFuture<Result>	append(Append append)	Appends values to one or more columns within a single row.
<T> List<CompletableFuture<T>>	batch(List<? extends Row> actions)	Method that does a batch call on Deletes, Gets, Puts, Increments, Appends and RowMutations.
AsyncTable.CheckAndMutateBuilder	checkAndMutate(byte[] row, byte[] family)	Atomically checks if a row/family/qualifier value matches the expected value.
AsyncTable.CheckAndMutateWithFilterBuilder	checkAndMutate(byte[] row, Filter filter)	Atomically checks if a row matches the specified filter.
List<CompletableFuture<CheckAndMutateResult>>	checkAndMutate(List<CheckAndMutate> checkAndMutates)	Batch version of checkAndMutate.
CompletableFuture<CheckAndMutateResult>	checkAndMutate(CheckAndMutate checkAndMutate)	checkAndMutate that atomically checks if a row matches the specified condition.
<S, R> CompletableFuture<R>	coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker, ServiceCaller<S,R> callable, byte[] row)	Execute the given coprocessor call on the region which contains the given row.
<S, R> AsyncTable.CoprocessorServiceBuilder<S,R>	coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker, ServiceCaller<S,R> callable, AsyncTable.CoprocessorCallback<R> callback)	Execute a coprocessor call on the regions which are covered by a range.
<S, R> AsyncTable.CoprocessorServiceBuilder<S,R>	coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker, ServiceCaller<S,R> callable, AsyncTable.PartialResultCoprocessorCallback<S,R> callback)	Similar to above.
List<CompletableFuture<Void>>	delete(List<Delete> deletes)	Deletes the specified cells/rows in bulk.
CompletableFuture<Void>	delete(Delete delete)	Deletes the specified cells/row.
List<CompletableFuture<Result>>	get(List<Get> gets)	Extracts certain cells from the given rows, in batch.
CompletableFuture<Result>	get(Get get)	Extracts certain cells from a given row.
org.apache.hadoop.conf.Configuration	getConfiguration()	Returns the Configuration object used by this instance.
CompletableFuture<TableDescriptor>	getDescriptor()	Gets the TableDescriptor for this table.
TableName	getName()	Gets the fully qualified table name instance of this table.
long	getOperationTimeout(TimeUnit unit)	Get timeout of each operation in Table instance.
long	getReadRpcTimeout(TimeUnit unit)	Get timeout of each rpc read request in this Table instance.
AsyncTableRegionLocator	getRegionLocator()	Gets the AsyncTableRegionLocator for this table.
Map<String,byte[]>	getRequestAttributes()	Get the map of request attributes
long	getRpcTimeout(TimeUnit unit)	Get timeout of each rpc request in this Table instance.
ResultScanner	getScanner(Scan scan)	Returns a scanner on the current table as specified by the Scan object.
long	getScanTimeout(TimeUnit unit)	Get the timeout of a single operation in a scan.
long	getWriteRpcTimeout(TimeUnit unit)	Get timeout of each rpc write request in this Table instance.
CompletableFuture<Result>	increment(Increment increment)	Increments one or more columns within a single row.
CompletableFuture<Result>	mutateRow(RowMutations mutation)	Performs multiple mutations atomically on a single row.
List<CompletableFuture<Void>>	put(List<Put> puts)	Puts some data in the table, in batch.
CompletableFuture<Void>	put(Put put)	Puts some data to the table.
void	scan(Scan scan, ScanResultConsumer consumer)	The scan API uses the observer pattern.
private void	scan0(Scan scan, ScanResultConsumer consumer)
CompletableFuture<List<Result>>	scanAll(Scan scan)	Return all the results that match the given scan object.
private <T> CompletableFuture<T>	wrap(CompletableFuture<T> future)

Methods inherited from class java.lang.Object

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

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

batchAll, checkAndMutateAll, deleteAll, exists, exists, existsAll, getAll, getScanner, getScanner, incrementColumnValue, incrementColumnValue, putAll

Field Details

rawTable

private final RawAsyncTableImpl rawTable

pool

private final ExecutorService pool

Constructor Details

AsyncTableImpl

AsyncTableImpl(RawAsyncTableImpl rawTable,
 ExecutorService pool)

Method Details

getName

public TableName getName()

Description copied from interface: AsyncTable

Gets the fully qualified table name instance of this table.

Specified by:

getName in interface AsyncTable<ScanResultConsumer>

getConfiguration

public org.apache.hadoop.conf.Configuration getConfiguration()

Description copied from interface: AsyncTable

Returns the Configuration object used by this instance.

The reference returned is not a copy, so any change made to it will affect this instance.

Specified by:

getConfiguration in interface AsyncTable<ScanResultConsumer>

getDescriptor

public CompletableFuture<TableDescriptor> getDescriptor()

Description copied from interface: AsyncTable

Gets the TableDescriptor for this table.

Specified by:

getDescriptor in interface AsyncTable<ScanResultConsumer>

getRegionLocator

public AsyncTableRegionLocator getRegionLocator()

Description copied from interface: AsyncTable

Gets the AsyncTableRegionLocator for this table.

Specified by:

getRegionLocator in interface AsyncTable<ScanResultConsumer>

getRpcTimeout

public long getRpcTimeout(TimeUnit unit)

Description copied from interface: AsyncTable

Get timeout of each rpc request in this Table instance. It will be overridden by a more specific rpc timeout config such as readRpcTimeout or writeRpcTimeout.

Specified by:

getRpcTimeout in interface AsyncTable<ScanResultConsumer>

Parameters:

unit - the unit of time the timeout to be represented in

Returns:

rpc timeout in the specified time unit

See Also:

• AsyncTable.getReadRpcTimeout(TimeUnit)
• AsyncTable.getWriteRpcTimeout(TimeUnit)

getReadRpcTimeout

public long getReadRpcTimeout(TimeUnit unit)

Description copied from interface: AsyncTable

Get timeout of each rpc read request in this Table instance.

Specified by:

getReadRpcTimeout in interface AsyncTable<ScanResultConsumer>

Parameters:

unit - the unit of time the timeout to be represented in

Returns:

read rpc timeout in the specified time unit

getWriteRpcTimeout

public long getWriteRpcTimeout(TimeUnit unit)

Description copied from interface: AsyncTable

Get timeout of each rpc write request in this Table instance.

Specified by:

getWriteRpcTimeout in interface AsyncTable<ScanResultConsumer>

Parameters:

unit - the unit of time the timeout to be represented in

Returns:

write rpc timeout in the specified time unit

getOperationTimeout

public long getOperationTimeout(TimeUnit unit)

Description copied from interface: AsyncTable

Get timeout of each operation in Table instance.

Specified by:

getOperationTimeout in interface AsyncTable<ScanResultConsumer>

Parameters:

unit - the unit of time the timeout to be represented in

Returns:

operation rpc timeout in the specified time unit

getScanTimeout

public long getScanTimeout(TimeUnit unit)

Description copied from interface: AsyncTable

Get the timeout of a single operation in a scan. It works like operation timeout for other operations.

Specified by:

getScanTimeout in interface AsyncTable<ScanResultConsumer>

Parameters:

unit - the unit of time the timeout to be represented in

Returns:

scan rpc timeout in the specified time unit

getRequestAttributes

public Map<String,byte[]> getRequestAttributes()

Description copied from interface: AsyncTable

Get the map of request attributes

Specified by:

getRequestAttributes in interface AsyncTable<ScanResultConsumer>

Returns:

a map of request attributes supplied by the client

wrap

private <T> CompletableFuture<T> wrap(CompletableFuture<T> future)

get

public CompletableFuture<Result> get(Get get)

Description copied from interface: AsyncTable

Extracts certain cells from a given row.

Specified by:

get in interface AsyncTable<ScanResultConsumer>

Parameters:

get - The object that specifies what data to fetch and from which row.

Returns:

The data coming from the specified row, if it exists. If the row specified doesn't exist, the Result instance returned won't contain any KeyValue, as indicated by Result.isEmpty(). The return value will be wrapped by a CompletableFuture.

put

public CompletableFuture<Void> put(Put put)

Description copied from interface: AsyncTable

Puts some data to the table.

Specified by:

put in interface AsyncTable<ScanResultConsumer>

Parameters:

put - The data to put.

Returns:

A CompletableFuture that always returns null when complete normally.

delete

public CompletableFuture<Void> delete(Delete delete)

Description copied from interface: AsyncTable

Deletes the specified cells/row.

Specified by:

delete in interface AsyncTable<ScanResultConsumer>

Parameters:

delete - The object that specifies what to delete.

Returns:

A CompletableFuture that always returns null when complete normally.

append

public CompletableFuture<Result> append(Append append)

Description copied from interface: AsyncTable

Appends values to one or more columns within a single row.

This operation does not appear atomic to readers. Appends are done under a single row lock, so write operations to a row are synchronized, but readers do not take row locks so get and scan operations can see this operation partially completed.

Specified by:

append in interface AsyncTable<ScanResultConsumer>

Parameters:

append - object that specifies the columns and amounts to be used for the increment operations

Returns:

values of columns after the append operation (maybe null). The return value will be wrapped by a CompletableFuture.

increment

public CompletableFuture<Result> increment(Increment increment)

Description copied from interface: AsyncTable

Increments one or more columns within a single row.

This operation does not appear atomic to readers. Increments are done under a single row lock, so write operations to a row are synchronized, but readers do not take row locks so get and scan operations can see this operation partially completed.

Specified by:

increment in interface AsyncTable<ScanResultConsumer>

Parameters:

increment - object that specifies the columns and amounts to be used for the increment operations

Returns:

values of columns after the increment. The return value will be wrapped by a CompletableFuture.

checkAndMutate

public AsyncTable.CheckAndMutateBuilder checkAndMutate(byte[] row,
 byte[] family)

Description copied from interface: AsyncTable

Atomically checks if a row/family/qualifier value matches the expected value. If it does, it adds the Put/Delete/RowMutations.

Use the returned AsyncTable.CheckAndMutateBuilder to construct your request and then execute it. This is a fluent style API, the code is like:

 table.checkAndMutate(row, family).qualifier(qualifier).ifNotExists().thenPut(put)
   .thenAccept(succ -> {
     if (succ) {
       System.out.println("Check and put succeeded");
     } else {
       System.out.println("Check and put failed");
     }
   });
 

Specified by:

checkAndMutate in interface AsyncTable<ScanResultConsumer>

checkAndMutate

public AsyncTable.CheckAndMutateWithFilterBuilder checkAndMutate(byte[] row,
 Filter filter)

Description copied from interface: AsyncTable

Atomically checks if a row matches the specified filter. If it does, it adds the Put/Delete/RowMutations.

Use the returned AsyncTable.CheckAndMutateWithFilterBuilder to construct your request and then execute it. This is a fluent style API, the code is like:

 table.checkAndMutate(row, filter).thenPut(put).thenAccept(succ -> {
   if (succ) {
     System.out.println("Check and put succeeded");
   } else {
     System.out.println("Check and put failed");
   }
 });
 

Specified by:

checkAndMutate in interface AsyncTable<ScanResultConsumer>

checkAndMutate

public CompletableFuture<CheckAndMutateResult> checkAndMutate(CheckAndMutate checkAndMutate)

Description copied from interface: AsyncTable

checkAndMutate that atomically checks if a row matches the specified condition. If it does, it performs the specified action.

Specified by:

checkAndMutate in interface AsyncTable<ScanResultConsumer>

Parameters:

checkAndMutate - The CheckAndMutate object.

Returns:

A CompletableFutures that represent the result for the CheckAndMutate.

checkAndMutate

public List<CompletableFuture<CheckAndMutateResult>> checkAndMutate(List<CheckAndMutate> checkAndMutates)

Description copied from interface: AsyncTable

Batch version of checkAndMutate. The specified CheckAndMutates are batched only in the sense that they are sent to a RS in one RPC, but each CheckAndMutate operation is still executed atomically (and thus, each may fail independently of others).

Specified by:

checkAndMutate in interface AsyncTable<ScanResultConsumer>

Parameters:

checkAndMutates - The list of CheckAndMutate.

Returns:

A list of CompletableFutures that represent the result for each CheckAndMutate.

mutateRow

public CompletableFuture<Result> mutateRow(RowMutations mutation)

Description copied from interface: AsyncTable

Performs multiple mutations atomically on a single row. Currently Put and Delete are supported.

Specified by:

mutateRow in interface AsyncTable<ScanResultConsumer>

Parameters:

mutation - object that specifies the set of mutations to perform atomically

Returns:

A CompletableFuture that returns results of Increment/Append operations

scanAll

public CompletableFuture<List<Result>> scanAll(Scan scan)

Description copied from interface: AsyncTable

Return all the results that match the given scan object.

Notice that usually you should use this method with a Scan object that has limit set. For example, if you want to get the closest row after a given row, you could do this:

 table.scanAll(new Scan().withStartRow(row, false).setLimit(1)).thenAccept(results -> {
   if (results.isEmpty()) {
     System.out.println("No row after " + Bytes.toStringBinary(row));
   } else {
     System.out.println("The closest row after " + Bytes.toStringBinary(row) + " is "
       + Bytes.toStringBinary(results.stream().findFirst().get().getRow()));
   }
 });
 

If your result set is very large, you should use other scan method to get a scanner or use callback to process the results. They will do chunking to prevent OOM. The scanAll method will fetch all the results and store them in a List and then return the list to you.

The scan metrics will be collected background if you enable it but you have no way to get it. Usually you can get scan metrics from ResultScanner, or through ScanResultConsumer.onScanMetricsCreated but this method only returns a list of results. So if you really care about scan metrics then you'd better use other scan methods which return a ResultScanner or let you pass in a ScanResultConsumer. There is no performance difference between these scan methods so do not worry.

Specified by:

scanAll in interface AsyncTable<ScanResultConsumer>

Parameters:

scan - A configured Scan object. So if you use this method to fetch a really large result set, it is likely to cause OOM.

Returns:

The results of this small scan operation. The return value will be wrapped by a CompletableFuture.

getScanner

public ResultScanner getScanner(Scan scan)

Description copied from interface: AsyncTable

Returns a scanner on the current table as specified by the Scan object.

Specified by:

getScanner in interface AsyncTable<ScanResultConsumer>

Parameters:

scan - A configured Scan object.

Returns:

A scanner.

scan0

private void scan0(Scan scan,
 ScanResultConsumer consumer)

scan

public void scan(Scan scan,
 ScanResultConsumer consumer)

Description copied from interface: AsyncTable

The scan API uses the observer pattern.

Specified by:

scan in interface AsyncTable<ScanResultConsumer>

Parameters:

scan - A configured Scan object.

consumer - the consumer used to receive results.

See Also:

• ScanResultConsumer
• AdvancedScanResultConsumer

get

public List<CompletableFuture<Result>> get(List<Get> gets)

Description copied from interface: AsyncTable

Extracts certain cells from the given rows, in batch.

Notice that you may not get all the results with this function, which means some of the returned CompletableFutures may succeed while some of the other returned CompletableFutures may fail.

Specified by:

get in interface AsyncTable<ScanResultConsumer>

Parameters:

gets - The objects that specify what data to fetch and from which rows.

Returns:

A list of CompletableFutures that represent the result for each get.

put

public List<CompletableFuture<Void>> put(List<Put> puts)

Description copied from interface: AsyncTable

Puts some data in the table, in batch.

Specified by:

put in interface AsyncTable<ScanResultConsumer>

Parameters:

puts - The list of mutations to apply.

Returns:

A list of CompletableFutures that represent the result for each put.

delete

public List<CompletableFuture<Void>> delete(List<Delete> deletes)

Description copied from interface: AsyncTable

Deletes the specified cells/rows in bulk.

Specified by:

delete in interface AsyncTable<ScanResultConsumer>

Parameters:

deletes - list of things to delete.

Returns:

A list of CompletableFutures that represent the result for each delete.

batch

public <T> List<CompletableFuture<T>> batch(List<? extends Row> actions)

Description copied from interface: AsyncTable

Method that does a batch call on Deletes, Gets, Puts, Increments, Appends and RowMutations. The ordering of execution of the actions is not defined. Meaning if you do a Put and a Get in the same AsyncTable.batch(java.util.List<? extends org.apache.hadoop.hbase.client.Row>) call, you will not necessarily be guaranteed that the Get returns what the Put had put.

Specified by:

batch in interface AsyncTable<ScanResultConsumer>

Parameters:

actions - list of Get, Put, Delete, Increment, Append, and RowMutations objects

Returns:

A list of CompletableFutures that represent the result for each action.

coprocessorService

public <S,
R> CompletableFuture<R> coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker,
 ServiceCaller<S,R> callable,
 byte[] row)

Description copied from interface: AsyncTable

Execute the given coprocessor call on the region which contains the given row.

The stubMaker is just a delegation to the newStub call. Usually it is only a one line lambda expression, like:

 channel -> xxxService.newStub(channel)
 

Specified by:

coprocessorService in interface AsyncTable<ScanResultConsumer>

Type Parameters:

S - the type of the asynchronous stub

R - the type of the return value

Parameters:

stubMaker - a delegation to the actual newStub call.

callable - a delegation to the actual protobuf rpc call. See the comment of ServiceCaller for more details.

row - The row key used to identify the remote region location

Returns:

the return value of the protobuf rpc call, wrapped by a CompletableFuture.

See Also:

• ServiceCaller

coprocessorService

public <S,
R>
AsyncTable.CoprocessorServiceBuilder<S,R> coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker,
 ServiceCaller<S,R> callable,
 AsyncTable.CoprocessorCallback<R> callback)

Description copied from interface: AsyncTable

Execute a coprocessor call on the regions which are covered by a range.

Use the returned AsyncTable.CoprocessorServiceBuilder construct your request and then execute it.

The stubMaker is just a delegation to the xxxService.newStub call. Usually it is only a one line lambda expression, like:

 channel -> xxxService.newStub(channel)
 

Specified by:

coprocessorService in interface AsyncTable<ScanResultConsumer>

Parameters:

stubMaker - a delegation to the actual newStub call.

callable - a delegation to the actual protobuf rpc call. See the comment of ServiceCaller for more details.

callback - callback to get the response. See the comment of AsyncTable.CoprocessorCallback for more details.

coprocessorService

public <S,
R>
AsyncTable.CoprocessorServiceBuilder<S,R> coprocessorService(Function<org.apache.hbase.thirdparty.com.google.protobuf.RpcChannel,S> stubMaker,
 ServiceCaller<S,R> callable,
 AsyncTable.PartialResultCoprocessorCallback<S,R> callback)

Description copied from interface: AsyncTable

Similar to above. Use when your coprocessor client+endpoint supports partial results. If the server does not offer partial results, it is still safe to use this, assuming you implement your AsyncTable.PartialResultCoprocessorCallback.getNextCallable(Object, RegionInfo) correctly.

Specified by:

coprocessorService in interface AsyncTable<ScanResultConsumer>