Passing Side Input in PCollection Partition - google-cloud-platform

I want to pass a sideInput in PCollection Partition and On basis of that, i need to Divide my PCollection is their anyway....
PCollectionList<TableRow> part = merged.apply(Partition.of(Pcollection Count Function Called, new PartitionFn<TableRow>(){
#Override
public int partitionFor(TableRow arg0, int arg1) {
return 0;
}
}));
Any Other Way through Which I Can Partition My PCollection
//Without Dynamic destination partitioning BigQuery table
merge.apply("write into target", BigQueryIO.writeTableRows()
.to(new SerializableFunction<ValueInSingleWindow<TableRow>, TableDestination>() {
#Override
public TableDestination apply(ValueInSingleWindow<TableRow> value) {
TableRow row = value.getValue();
TableReference reference = new TableReference();
reference.setProjectId("XYZ");
reference.setDatasetId("ABC");
System.out.println("date of row " + row.get("authorized_transaction_date_yyyymmdd").toString());
LOG.info("date of row "+
row.get("authorized_transaction_date_yyyymmdd").toString());
String str = row.get("authorized_transaction_date_yyyymmdd").toString();
str = str.substring(0, str.length() - 2) + "01";
System.out.println("str value " + str);
LOG.info("str value " + str);
reference.setTableId("TargetTable$" + str);
return new TableDestination(reference, null);
}
}).withFormatFunction(new SerializableFunction<TableRow, TableRow>() {
#Override
public TableRow apply(TableRow input) {
LOG.info("format function:"+input.toString());
return input;
}
})
.withSchema(schema1).withWriteDisposition(BigQueryIO.Write.WriteDisposition.WRITE_TRUNCATE)
.withCreateDisposition(BigQueryIO.Write.CreateDisposition.CREATE_IF_NEEDED));
Now I have to use Dynamic Destination Any Solution.Insted Of this and have to Do Partition.

Based on seeing TableRow in your code, I suspect that you want to write a PCollection to BigQuery, sending different elements to different BigQuery tables. BigQueryIO.write() already provides a method to do that, using BigQueryIO.write().to(DynamicDestinations). See Writing different values to different BigQuery tables in Apache Beam.

Related

File to DB load using Apache beam

I need to load a file into my database, but before that I have to verify data is present in the database based on some file data. For instance, suppose I have 5 records in a file then I have to check 5 times in the database for separate records.
So how can I get this value dynamically? We have to pass dynamic value instead of 2 in line (preparedStatement.setString(1, "2");)
Here we are creating a Dataflow pipeline which loads data into the database using Apache Beam. Now we create a pipeline object and create a pipeline. Using a PCollection we are storing into database.
Pipeline p = Pipeline.create(options);
p.apply("Reading Text", TextIO.read().from(options.getInputFile()))
.apply(ParDo.of(new FilterHeaderFn(csvHeader)))
.apply(ParDo.of(new GetRatePlanID()))
.apply("Format Result", MapElements.into(
TypeDescriptors.strings()).via(
(KV < String, Integer > ABC) - >
ABC.getKey() + "," + ABC.getValue()))
.apply("Write File", TextIO.write()
.to(options.getOutputFile())
.withoutSharding());
// Retrieving data from database
PCollection < String > data =
p.apply(JdbcIO. < String > read()
.withDataSourceConfiguration(JdbcIO.DataSourceConfiguration.create(
"com.mysql.cj.jdbc.Driver", "jdbc:mysql://localhost:3306/XYZ")
.withUsername("root")
.withPassword("root1234"))
.withQuery("select * from xyz where z = ?")
.withCoder(StringUtf8Coder.of())
.withStatementPreparator(new JdbcIO.StatementPreparator() {
private static final long serialVersionUID = 1 L;
#Override
public void setParameters(PreparedStatement preparedStatement) throws Exception {
preparedStatement.setString(1, "2");
}
})
.withRowMapper(new JdbcIO.RowMapper < String > () {
private static final long serialVersionUID = 1 L;
public String mapRow(ResultSet resultSet) throws Exception {
return "Symbol: " + resultSet.getInt(1) + "\nPrice: " + resultSet.getString(2) +
"\nCompany: " + resultSet.getInt(3);
}
}));
As suggested, the most efficient would probably be loading the whole file into a temporary table and then doing a query to update the requisite rows.
If that can't be done, you could instead read the table into Dataflow (i.e. "select * from xyz") and then do a join/CoGroupByKey to match records with those found in your file. If you expect the existing database to be very large compared to the files you're hoping to upload into it, you could have a DoFn that makes queries to your database directly using JDBC (possibly caching the connection in the DoFn's setUp method) rather than using JdbcIO.

How to use Isolationforest in weka?

I am trying to use isolationforest in weka ,but I cannot find a easy example which shows how to use it ,who can help me ?thanks in advance
import weka.classifiers.misc.IsolationForest;
public class Test2 {
public static void main(String[] args) {
IsolationForest isolationForest = new IsolationForest();
.....................................................
}
}
I strongly suggest you to study a little bit the implementation for IslationForest.
The following code work loading a CSV file with first column with Class (note: a single class value will produce only (1-anomaly score) if it's binary you will get the anomaly score too. Otherwise it just return an error). Note I skip the second column (that in my case is the uuid that is not needed for anomaly detection)
private static void findOutlier(File in, File out) throws Exception {
CSVLoader loader = new CSVLoader();
loader.setSource(new File(in.getAbsolutePath()));
Instances data = loader.getDataSet();
// setting class attribute if the data format does not provide this information
// For example, the XRFF format saves the class attribute information as well
if (data.classIndex() == -1)
data.setClassIndex(0);
String[] options = new String[2];
options[0] = "-R"; // "range"
options[1] = "2"; // first attribute
Remove remove = new Remove(); // new instance of filter
remove.setOptions(options); // set options
remove.setInputFormat(data); // inform filter about dataset **AFTER** setting options
Instances newData = Filter.useFilter(data, remove); // apply filter
IsolationForest randomForest = new IsolationForest();
randomForest.buildClassifier(newData);
// System.out.println(randomForest);
FileWriter fw = new FileWriter(out);
final Enumeration<Attribute> attributeEnumeration = data.enumerateAttributes();
for (Attribute e = attributeEnumeration.nextElement(); attributeEnumeration.hasMoreElements(); e = attributeEnumeration.nextElement()) {
fw.write(e.name());
fw.write(",");
}
fw.write("(1 - anomaly score),anomaly score\n");
for (int i = 0; i < data.size(); ++i) {
Instance inst = data.get(i);
final double[] distributionForInstance = randomForest.distributionForInstance(inst);
fw.write(inst + ", " + distributionForInstance[0] + "," + (1 - distributionForInstance[0]));
fw.write(",\n");
}
fw.flush();
}
The previous function will add at the CSV at last column the anomaly values. Please note I'm using a single class so for getting the corresponding anomaly I do 1 - distributionForInstance[0] otherwise you ca do simply distributionForInstance[1] .
A sample input.csv for getting (1-anomaly score):
Class,ignore, feature_0, feature_1, feature_2
A,1,21,31,31
A,2,41,61,81
A,3,61,37,34
A sample input.csv for getting (1-anomaly score) and anomaly score:
Class,ignore, feature_0, feature_1, feature_2
A,1,21,31,31
B,2,41,61,81
A,3,61,37,34

How to do multiple parallel readers for data export using Google Spanner?

External Backups/Snapshots for Google Cloud Spanner recommends to use queries with timestamp bounds to create snapshots for export. On the bottom of the Timestamp Bounds documentation it states:
Cloud Spanner continuously garbage collects deleted and overwritten data in the background to reclaim storage space. This process is known as version GC. By default, version GC reclaims versions after they are one hour old. Because of this, Cloud Spanner cannot perform reads at a read timestamp more than one hour in the past.
So any export would need to complete within an hour. A single reader (i.e. select * from table; using timestamp X) would not be able to export the entire table within an hour.
How can multiple parallel readers be implemented in spanner?
Note: It is mentioned in one of the comments that support for Apache Beam is coming, but it looks like that uses a single reader:
/** A simplest read function implementation. Parallelism support is coming. */
https://github.com/apache/beam/blob/master/sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/spanner/NaiveSpannerReadFn.java#L26
Is there a way to do the parallel reader that beam requires today using exising APIs? Or will Beam need to use something that isn't released yet on google spanner?
It is possible to read data in parallel from Cloud Spanner with the BatchClient class. Follow read_data_in_parallel for more information.
If you are looking to export data from Cloud Spanner, I'd recommend you to use Cloud Dataflow (see the integration details here) as it provides higher level abstractions and takes care data processing details, like scaling and failure handling.
Edit 2018-03-30 - The example project has been updated to use the BatchClient offered by Google Cloud Spanner
After the release of the BatchClient for reading/downloading large amounts of data, the example project below has been updated to use the new batch client instead of the standard database client. The basic idea behind the project is still the same: Copy data to/from Cloud Spanner and any other database using standard jdbc functionality. The following code snippet sets the jdbc connection in batch read mode:
if (source.isWrapperFor(ICloudSpannerConnection.class))
{
ICloudSpannerConnection con = source.unwrap(ICloudSpannerConnection.class);
// Make sure no transaction is running
if (!con.isBatchReadOnly())
{
if (con.getAutoCommit())
{
con.setAutoCommit(false);
}
else
{
con.commit();
}
con.setBatchReadOnly(true);
}
}
When the connection is in 'batch read only mode', the connection will use the BatchClient of Google Cloud Spanner instead of the standard database client. When one of the Statement#execute(String) or PreparedStatement#execute() methods are called (as these allow multiple result sets to be returned) the jdbc driver will create a partitioned query instead of a normal query. The results of this partitioned query will be a number of result sets (one per partition) that can be fetched by the Statement#getResultSet() and Statement#getMoreResults(int) methods.
Statement statement = source.createStatement();
boolean hasResults = statement.execute(select);
int workerNumber = 0;
while (hasResults)
{
ResultSet rs = statement.getResultSet();
PartitionWorker worker = new PartitionWorker("PartionWorker-" + workerNumber, config, rs, tableSpec, table, insertCols);
workers.add(worker);
hasResults = statement.getMoreResults(Statement.KEEP_CURRENT_RESULT);
workerNumber++;
}
The result sets that are returned by the Statement#execute(String) are not executed directly, but only after the first call to ResultSet#next(). Passing these result sets to separate worker threads ensures parallel download and copying of the data.
Original answer:
This project was initially created for conversion in the other direction (from a local database to Cloud Spanner), but as it uses JDBC for both source and destination it can also be used the other way around: Converting a Cloud Spanner database to a local PostgreSQL database. Large tables are converted in parallel using a thread pool.
The project uses this open source JDBC driver instead of the JDBC driver supplied by Google. The source Cloud Spanner JDBC connection is set to read-only mode and autocommit=false. This ensures that the connection automatically creates a read-only transaction using the current time as timestamp the first time you execute a query. All subsequent queries within the same (read-only) transaction will use the same timestamp giving you a consistent snapshot of your Google Cloud Spanner database.
It works as follows:
Set the source database to read-only transactional mode.
The convert(String catalog, String schema) method iterates over all
tables in the source database (Cloud Spanner)
For each table the number of records is determined, and depending on the size of the table, the table is copied using either the main thread of the application or by a worker pool.
The class UploadWorker is responsible for the parallel copying. Each worker is assigned a range of records from the table (for example rows 1 to 2,400). The range is selected by a select statement in this format: 'SELECT * FROM $TABLE ORDER BY $PK_COLUMNS LIMIT $BATCH_SIZE OFFSET $CURRENT_OFFSET'
Commit the read-only transaction on the source database after ALL tables have been converted.
Below is a code snippet of the most important parts.
public void convert(String catalog, String schema) throws SQLException
{
int batchSize = config.getBatchSize();
destination.setAutoCommit(false);
// Set the source connection to transaction mode (no autocommit) and read-only
source.setAutoCommit(false);
source.setReadOnly(true);
try (ResultSet tables = destination.getMetaData().getTables(catalog, schema, null, new String[] { "TABLE" }))
{
while (tables.next())
{
String tableSchema = tables.getString("TABLE_SCHEM");
if (!config.getDestinationDatabaseType().isSystemSchema(tableSchema))
{
String table = tables.getString("TABLE_NAME");
// Check whether the destination table is empty.
int destinationRecordCount = getDestinationRecordCount(table);
if (destinationRecordCount == 0 || config.getDataConvertMode() == ConvertMode.DropAndRecreate)
{
if (destinationRecordCount > 0)
{
deleteAll(table);
}
int sourceRecordCount = getSourceRecordCount(getTableSpec(catalog, tableSchema, table));
if (sourceRecordCount > batchSize)
{
convertTableWithWorkers(catalog, tableSchema, table);
}
else
{
convertTable(catalog, tableSchema, table);
}
}
else
{
if (config.getDataConvertMode() == ConvertMode.ThrowExceptionIfExists)
throw new IllegalStateException("Table " + table + " is not empty");
else if (config.getDataConvertMode() == ConvertMode.SkipExisting)
log.info("Skipping data copy for table " + table);
}
}
}
}
source.commit();
}
private void convertTableWithWorkers(String catalog, String schema, String table) throws SQLException
{
String tableSpec = getTableSpec(catalog, schema, table);
Columns insertCols = getColumns(catalog, schema, table, false);
Columns selectCols = getColumns(catalog, schema, table, true);
if (insertCols.primaryKeyCols.isEmpty())
{
log.warning("Table " + tableSpec + " does not have a primary key. No data will be copied.");
return;
}
log.info("About to copy data from table " + tableSpec);
int batchSize = config.getBatchSize();
int totalRecordCount = getSourceRecordCount(tableSpec);
int numberOfWorkers = calculateNumberOfWorkers(totalRecordCount);
int numberOfRecordsPerWorker = totalRecordCount / numberOfWorkers;
if (totalRecordCount % numberOfWorkers > 0)
numberOfRecordsPerWorker++;
int currentOffset = 0;
ExecutorService service = Executors.newFixedThreadPool(numberOfWorkers);
for (int workerNumber = 0; workerNumber < numberOfWorkers; workerNumber++)
{
int workerRecordCount = Math.min(numberOfRecordsPerWorker, totalRecordCount - currentOffset);
UploadWorker worker = new UploadWorker("UploadWorker-" + workerNumber, selectFormat, tableSpec, table,
insertCols, selectCols, currentOffset, workerRecordCount, batchSize, source,
config.getUrlDestination(), config.isUseJdbcBatching());
service.submit(worker);
currentOffset = currentOffset + numberOfRecordsPerWorker;
}
service.shutdown();
try
{
service.awaitTermination(config.getUploadWorkerMaxWaitInMinutes(), TimeUnit.MINUTES);
}
catch (InterruptedException e)
{
log.severe("Error while waiting for workers to finish: " + e.getMessage());
throw new RuntimeException(e);
}
}
public class UploadWorker implements Runnable
{
private static final Logger log = Logger.getLogger(UploadWorker.class.getName());
private final String name;
private String selectFormat;
private String sourceTable;
private String destinationTable;
private Columns insertCols;
private Columns selectCols;
private int beginOffset;
private int numberOfRecordsToCopy;
private int batchSize;
private Connection source;
private String urlDestination;
private boolean useJdbcBatching;
UploadWorker(String name, String selectFormat, String sourceTable, String destinationTable, Columns insertCols,
Columns selectCols, int beginOffset, int numberOfRecordsToCopy, int batchSize, Connection source,
String urlDestination, boolean useJdbcBatching)
{
this.name = name;
this.selectFormat = selectFormat;
this.sourceTable = sourceTable;
this.destinationTable = destinationTable;
this.insertCols = insertCols;
this.selectCols = selectCols;
this.beginOffset = beginOffset;
this.numberOfRecordsToCopy = numberOfRecordsToCopy;
this.batchSize = batchSize;
this.source = source;
this.urlDestination = urlDestination;
this.useJdbcBatching = useJdbcBatching;
}
#Override
public void run()
{
// Connection source = DriverManager.getConnection(urlSource);
try (Connection destination = DriverManager.getConnection(urlDestination))
{
log.info(name + ": " + sourceTable + ": Starting copying " + numberOfRecordsToCopy + " records");
destination.setAutoCommit(false);
String sql = "INSERT INTO " + destinationTable + " (" + insertCols.getColumnNames() + ") VALUES \n";
sql = sql + "(" + insertCols.getColumnParameters() + ")";
PreparedStatement statement = destination.prepareStatement(sql);
int lastRecord = beginOffset + numberOfRecordsToCopy;
int recordCount = 0;
int currentOffset = beginOffset;
while (true)
{
int limit = Math.min(batchSize, lastRecord - currentOffset);
String select = selectFormat.replace("$COLUMNS", selectCols.getColumnNames());
select = select.replace("$TABLE", sourceTable);
select = select.replace("$PRIMARY_KEY", selectCols.getPrimaryKeyColumns());
select = select.replace("$BATCH_SIZE", String.valueOf(limit));
select = select.replace("$OFFSET", String.valueOf(currentOffset));
try (ResultSet rs = source.createStatement().executeQuery(select))
{
while (rs.next())
{
int index = 1;
for (Integer type : insertCols.columnTypes)
{
Object object = rs.getObject(index);
statement.setObject(index, object, type);
index++;
}
if (useJdbcBatching)
statement.addBatch();
else
statement.executeUpdate();
recordCount++;
}
if (useJdbcBatching)
statement.executeBatch();
}
destination.commit();
log.info(name + ": " + sourceTable + ": Records copied so far: " + recordCount + " of "
+ numberOfRecordsToCopy);
currentOffset = currentOffset + batchSize;
if (recordCount >= numberOfRecordsToCopy)
break;
}
}
catch (SQLException e)
{
log.severe("Error during data copy: " + e.getMessage());
throw new RuntimeException(e);
}
log.info(name + ": Finished copying");
}
}

Univocity - parse each TSV file row to different Type of class object

I have a tsv file which has fixed rows but each row is mapped to different Java Class.
For example.
recordType recordValue1
recordType recordValue1 recordValue2
for First row I have follofing class:
public class FirstRow implements ItsvRecord {
#Parsed(index = 0)
private String recordType;
#Parsed(index = 1)
private String recordValue1;
public FirstRow() {
}
}
and for second row I have:
public class SecondRow implements ItsvRecord {
#Parsed(index = 0)
private String recordType;
#Parsed(index = 1)
private String recordValue1;
public SecondRow() {
}
}
I want to parse the TSV file directly to the respective objects but I am falling short of ideas.
Use an InputValueSwitch. This will match a value in a particular column of each row to determine what RowProcessor to use. Example:
Create two (or more) processors for each type of record you need to process:
final BeanListProcessor<FirstRow> firstProcessor = new BeanListProcessor<FirstRow>(FirstRow.class);
final BeanListProcessor<SecondRow> secondProcessor = new BeanListProcessor<SecondRow>(SecondRow.class);
Create an InputValueSwitch:
//0 means that the first column of each row has a value that
//identifies what is the type of record you are dealing with
InputValueSwitch valueSwitch = new InputValueSwitch(0);
//assigns the first processor to rows whose first column contain the 'firstRowType' value
valueSwitch.addSwitchForValue("firstRowType", firstProcessor);
//assigns the second processor to rows whose first column contain the 'secondRowType' value
valueSwitch.addSwitchForValue("secondRowType", secondProcessor);
Parse as usual:
TsvParserSettings settings = new TsvParserSettings(); //configure...
// your row processor is the switch
settings.setProcessor(valueSwitch);
TsvParser parser = new TsvParser(settings);
Reader input = new StringReader(""+
"firstRowType\trecordValue1\n" +
"secondRowType\trecordValue1\trecordValue2");
parser.parse(input);
Get the parsed objects from your processors:
List<FirstRow> firstTypeObjects = firstProcessor.getBeans();
List<SecondRow> secondTypeObjects = secondProcessor.getBeans();
The output will be*:
[FirstRow{recordType='firstRowType', recordValue1='recordValue1'}]
[SecondRow{recordType='secondRowType', recordValue1='recordValue1', recordValue2='recordValue2'}]
Assuming you have a sane toString() implemented in your classes
If you want to manage associations among the objects that are parsed:
If your FirstRow should contain the elements parsed for records of type SecondRow, simply override the rowProcessorSwitched method:
InputValueSwitch valueSwitch = new InputValueSwitch(0) {
#Override
public void rowProcessorSwitched(RowProcessor from, RowProcessor to) {
if (from == secondProcessor) {
List<FirstRow> firstRows = firstProcessor.getBeans();
FirstRow mostRecentRow = firstRows.get(firstRows.size() - 1);
mostRecentRow.addRowsOfOtherType(secondProcessor.getBeans());
secondProcessor.getBeans().clear();
}
}
};
The above assumes your FirstRow class has a addRowsOfOtherType method that takes a list of SecondRow as parameter.
And that's it!
You can even mix and match other types of RowProcessor. There's another example here that demonstrates this.
Hope this helps.

How to get all rows containing (or equaling) a particular ID from an HBase table?

I have a method which select the row whose rowkey contains the parameter passed into.
HTable table = new HTable(Bytes.toBytes(objectsTableName), connection);
public List<ObjectId> lookUp(String partialId) {
if (partialId.matches("[a-fA-F0-9]+")) {
// create a regular expression from partialId, which can
//match any rowkey that contains partialId as a substring,
//and then get all the row with the specified rowkey
} else {
throw new IllegalArgumentException(
"query must be done with hexadecimal values only");
}
}
I don't know how to finish code above.
I just know the following code can get the row with specified rowkey in Hbase.
String rowkey = "123";
Get get = new Get(Bytes.toBytes(rowkey));
Result result = table.get(get);
You can use RowFilter filter with RegexStringComparator to do that. Or, if it is just to fetch the rows which match a given substring you can use RowFilter with SubstringComparator. This is how you use HBase filters :
public static void main(String[] args) throws IOException {
Configuration conf = HBaseConfiguration.create();
HTable table = new HTable(conf, "demo");
Scan s = new Scan();
Filter f = new RowFilter(CompareOp.EQUAL, new SubstringComparator("abc"));
s.setFilter(f);
ResultScanner rs = table.getScanner(s);
for(Result r : rs){
System.out.println("RowKey : " + Bytes.toString(r.getRow()));
//rest of your logic
}
rs.close();
table.close();
}
The above piece of code will give you all the rows which contain abc as a part of their rowkeys.
HTH