Search in sources :

Example 1 with NORMALIZE_NAMES_FOR_AVRO

use of org.apache.nifi.processors.standard.util.JdbcCommon.NORMALIZE_NAMES_FOR_AVRO in project nifi by apache.

the class QueryDatabaseTable method onTrigger.

@Override
public void onTrigger(final ProcessContext context, final ProcessSessionFactory sessionFactory) throws ProcessException {
    // Fetch the column/table info once
    if (!setupComplete.get()) {
        super.setup(context);
    }
    ProcessSession session = sessionFactory.createSession();
    final List<FlowFile> resultSetFlowFiles = new ArrayList<>();
    final ComponentLog logger = getLogger();
    final DBCPService dbcpService = context.getProperty(DBCP_SERVICE).asControllerService(DBCPService.class);
    final DatabaseAdapter dbAdapter = dbAdapters.get(context.getProperty(DB_TYPE).getValue());
    final String tableName = context.getProperty(TABLE_NAME).evaluateAttributeExpressions().getValue();
    final String columnNames = context.getProperty(COLUMN_NAMES).evaluateAttributeExpressions().getValue();
    final String maxValueColumnNames = context.getProperty(MAX_VALUE_COLUMN_NAMES).evaluateAttributeExpressions().getValue();
    final String customWhereClause = context.getProperty(WHERE_CLAUSE).evaluateAttributeExpressions().getValue();
    final Integer fetchSize = context.getProperty(FETCH_SIZE).evaluateAttributeExpressions().asInteger();
    final Integer maxRowsPerFlowFile = context.getProperty(MAX_ROWS_PER_FLOW_FILE).evaluateAttributeExpressions().asInteger();
    final Integer outputBatchSizeField = context.getProperty(OUTPUT_BATCH_SIZE).evaluateAttributeExpressions().asInteger();
    final int outputBatchSize = outputBatchSizeField == null ? 0 : outputBatchSizeField;
    final Integer maxFragments = context.getProperty(MAX_FRAGMENTS).isSet() ? context.getProperty(MAX_FRAGMENTS).evaluateAttributeExpressions().asInteger() : 0;
    final JdbcCommon.AvroConversionOptions options = JdbcCommon.AvroConversionOptions.builder().recordName(tableName).maxRows(maxRowsPerFlowFile).convertNames(context.getProperty(NORMALIZE_NAMES_FOR_AVRO).asBoolean()).useLogicalTypes(context.getProperty(USE_AVRO_LOGICAL_TYPES).asBoolean()).defaultPrecision(context.getProperty(DEFAULT_PRECISION).evaluateAttributeExpressions().asInteger()).defaultScale(context.getProperty(DEFAULT_SCALE).evaluateAttributeExpressions().asInteger()).build();
    final StateManager stateManager = context.getStateManager();
    final StateMap stateMap;
    try {
        stateMap = stateManager.getState(Scope.CLUSTER);
    } catch (final IOException ioe) {
        getLogger().error("Failed to retrieve observed maximum values from the State Manager. Will not perform " + "query until this is accomplished.", ioe);
        context.yield();
        return;
    }
    // Make a mutable copy of the current state property map. This will be updated by the result row callback, and eventually
    // set as the current state map (after the session has been committed)
    final Map<String, String> statePropertyMap = new HashMap<>(stateMap.toMap());
    // If an initial max value for column(s) has been specified using properties, and this column is not in the state manager, sync them to the state property map
    for (final Map.Entry<String, String> maxProp : maxValueProperties.entrySet()) {
        String maxPropKey = maxProp.getKey().toLowerCase();
        String fullyQualifiedMaxPropKey = getStateKey(tableName, maxPropKey);
        if (!statePropertyMap.containsKey(fullyQualifiedMaxPropKey)) {
            String newMaxPropValue;
            // but store the new initial max value under the fully-qualified key.
            if (statePropertyMap.containsKey(maxPropKey)) {
                newMaxPropValue = statePropertyMap.get(maxPropKey);
            } else {
                newMaxPropValue = maxProp.getValue();
            }
            statePropertyMap.put(fullyQualifiedMaxPropKey, newMaxPropValue);
        }
    }
    List<String> maxValueColumnNameList = StringUtils.isEmpty(maxValueColumnNames) ? null : Arrays.asList(maxValueColumnNames.split("\\s*,\\s*"));
    final String selectQuery = getQuery(dbAdapter, tableName, columnNames, maxValueColumnNameList, customWhereClause, statePropertyMap);
    final StopWatch stopWatch = new StopWatch(true);
    final String fragmentIdentifier = UUID.randomUUID().toString();
    try (final Connection con = dbcpService.getConnection();
        final Statement st = con.createStatement()) {
        if (fetchSize != null && fetchSize > 0) {
            try {
                st.setFetchSize(fetchSize);
            } catch (SQLException se) {
                // Not all drivers support this, just log the error (at debug level) and move on
                logger.debug("Cannot set fetch size to {} due to {}", new Object[] { fetchSize, se.getLocalizedMessage() }, se);
            }
        }
        String jdbcURL = "DBCPService";
        try {
            DatabaseMetaData databaseMetaData = con.getMetaData();
            if (databaseMetaData != null) {
                jdbcURL = databaseMetaData.getURL();
            }
        } catch (SQLException se) {
        // Ignore and use default JDBC URL. This shouldn't happen unless the driver doesn't implement getMetaData() properly
        }
        final Integer queryTimeout = context.getProperty(QUERY_TIMEOUT).evaluateAttributeExpressions().asTimePeriod(TimeUnit.SECONDS).intValue();
        // timeout in seconds
        st.setQueryTimeout(queryTimeout);
        try {
            logger.debug("Executing query {}", new Object[] { selectQuery });
            final ResultSet resultSet = st.executeQuery(selectQuery);
            int fragmentIndex = 0;
            while (true) {
                final AtomicLong nrOfRows = new AtomicLong(0L);
                FlowFile fileToProcess = session.create();
                try {
                    fileToProcess = session.write(fileToProcess, out -> {
                        // Max values will be updated in the state property map by the callback
                        final MaxValueResultSetRowCollector maxValCollector = new MaxValueResultSetRowCollector(tableName, statePropertyMap, dbAdapter);
                        try {
                            nrOfRows.set(JdbcCommon.convertToAvroStream(resultSet, out, options, maxValCollector));
                        } catch (SQLException | RuntimeException e) {
                            throw new ProcessException("Error during database query or conversion of records to Avro.", e);
                        }
                    });
                } catch (ProcessException e) {
                    // Add flowfile to results before rethrowing so it will be removed from session in outer catch
                    resultSetFlowFiles.add(fileToProcess);
                    throw e;
                }
                if (nrOfRows.get() > 0) {
                    // set attribute how many rows were selected
                    fileToProcess = session.putAttribute(fileToProcess, RESULT_ROW_COUNT, String.valueOf(nrOfRows.get()));
                    fileToProcess = session.putAttribute(fileToProcess, RESULT_TABLENAME, tableName);
                    fileToProcess = session.putAttribute(fileToProcess, CoreAttributes.MIME_TYPE.key(), JdbcCommon.MIME_TYPE_AVRO_BINARY);
                    if (maxRowsPerFlowFile > 0) {
                        fileToProcess = session.putAttribute(fileToProcess, "fragment.identifier", fragmentIdentifier);
                        fileToProcess = session.putAttribute(fileToProcess, "fragment.index", String.valueOf(fragmentIndex));
                    }
                    logger.info("{} contains {} Avro records; transferring to 'success'", new Object[] { fileToProcess, nrOfRows.get() });
                    session.getProvenanceReporter().receive(fileToProcess, jdbcURL, stopWatch.getElapsed(TimeUnit.MILLISECONDS));
                    resultSetFlowFiles.add(fileToProcess);
                    // If we've reached the batch size, send out the flow files
                    if (outputBatchSize > 0 && resultSetFlowFiles.size() >= outputBatchSize) {
                        session.transfer(resultSetFlowFiles, REL_SUCCESS);
                        session.commit();
                        resultSetFlowFiles.clear();
                    }
                } else {
                    // If there were no rows returned, don't send the flowfile
                    session.remove(fileToProcess);
                    context.yield();
                    break;
                }
                fragmentIndex++;
                if (maxFragments > 0 && fragmentIndex >= maxFragments) {
                    break;
                }
            }
            // Even though the maximum value and total count are known at this point, to maintain consistent behavior if Output Batch Size is set, do not store the attributes
            if (outputBatchSize == 0) {
                for (int i = 0; i < resultSetFlowFiles.size(); i++) {
                    // Add maximum values as attributes
                    for (Map.Entry<String, String> entry : statePropertyMap.entrySet()) {
                        // Get just the column name from the key
                        String key = entry.getKey();
                        String colName = key.substring(key.lastIndexOf(NAMESPACE_DELIMITER) + NAMESPACE_DELIMITER.length());
                        resultSetFlowFiles.set(i, session.putAttribute(resultSetFlowFiles.get(i), "maxvalue." + colName, entry.getValue()));
                    }
                    // set count on all FlowFiles
                    if (maxRowsPerFlowFile > 0) {
                        resultSetFlowFiles.set(i, session.putAttribute(resultSetFlowFiles.get(i), "fragment.count", Integer.toString(fragmentIndex)));
                    }
                }
            }
        } catch (final SQLException e) {
            throw e;
        }
        session.transfer(resultSetFlowFiles, REL_SUCCESS);
    } catch (final ProcessException | SQLException e) {
        logger.error("Unable to execute SQL select query {} due to {}", new Object[] { selectQuery, e });
        if (!resultSetFlowFiles.isEmpty()) {
            session.remove(resultSetFlowFiles);
        }
        context.yield();
    } finally {
        session.commit();
        try {
            // Update the state
            stateManager.setState(statePropertyMap, Scope.CLUSTER);
        } catch (IOException ioe) {
            getLogger().error("{} failed to update State Manager, maximum observed values will not be recorded", new Object[] { this, ioe });
        }
    }
}
Also used : ProcessSession(org.apache.nifi.processor.ProcessSession) StandardValidators(org.apache.nifi.processor.util.StandardValidators) Arrays(java.util.Arrays) Connection(java.sql.Connection) StringUtils(org.apache.commons.lang3.StringUtils) PropertyDescriptor(org.apache.nifi.components.PropertyDescriptor) WritesAttributes(org.apache.nifi.annotation.behavior.WritesAttributes) Scope(org.apache.nifi.components.state.Scope) ResultSet(java.sql.ResultSet) Map(java.util.Map) ParseException(java.text.ParseException) TriggerSerially(org.apache.nifi.annotation.behavior.TriggerSerially) FlowFile(org.apache.nifi.flowfile.FlowFile) NORMALIZE_NAMES_FOR_AVRO(org.apache.nifi.processors.standard.util.JdbcCommon.NORMALIZE_NAMES_FOR_AVRO) Set(java.util.Set) WritesAttribute(org.apache.nifi.annotation.behavior.WritesAttribute) UUID(java.util.UUID) StateMap(org.apache.nifi.components.state.StateMap) InputRequirement(org.apache.nifi.annotation.behavior.InputRequirement) Stateful(org.apache.nifi.annotation.behavior.Stateful) List(java.util.List) DynamicProperty(org.apache.nifi.annotation.behavior.DynamicProperty) JdbcCommon(org.apache.nifi.processors.standard.util.JdbcCommon) StopWatch(org.apache.nifi.util.StopWatch) Tags(org.apache.nifi.annotation.documentation.Tags) DBCPService(org.apache.nifi.dbcp.DBCPService) ResultSetMetaData(java.sql.ResultSetMetaData) IntStream(java.util.stream.IntStream) CapabilityDescription(org.apache.nifi.annotation.documentation.CapabilityDescription) USE_AVRO_LOGICAL_TYPES(org.apache.nifi.processors.standard.util.JdbcCommon.USE_AVRO_LOGICAL_TYPES) DatabaseMetaData(java.sql.DatabaseMetaData) HashMap(java.util.HashMap) ComponentLog(org.apache.nifi.logging.ComponentLog) ProcessException(org.apache.nifi.processor.exception.ProcessException) ArrayList(java.util.ArrayList) DEFAULT_SCALE(org.apache.nifi.processors.standard.util.JdbcCommon.DEFAULT_SCALE) HashSet(java.util.HashSet) SQLException(java.sql.SQLException) Relationship(org.apache.nifi.processor.Relationship) DEFAULT_PRECISION(org.apache.nifi.processors.standard.util.JdbcCommon.DEFAULT_PRECISION) Requirement(org.apache.nifi.annotation.behavior.InputRequirement.Requirement) DatabaseAdapter(org.apache.nifi.processors.standard.db.DatabaseAdapter) StateManager(org.apache.nifi.components.state.StateManager) ProcessContext(org.apache.nifi.processor.ProcessContext) ProcessSession(org.apache.nifi.processor.ProcessSession) IOException(java.io.IOException) SeeAlso(org.apache.nifi.annotation.documentation.SeeAlso) ProcessSessionFactory(org.apache.nifi.processor.ProcessSessionFactory) TimeUnit(java.util.concurrent.TimeUnit) AtomicLong(java.util.concurrent.atomic.AtomicLong) OnScheduled(org.apache.nifi.annotation.lifecycle.OnScheduled) Statement(java.sql.Statement) CoreAttributes(org.apache.nifi.flowfile.attributes.CoreAttributes) OnStopped(org.apache.nifi.annotation.lifecycle.OnStopped) Collections(java.util.Collections) HashMap(java.util.HashMap) SQLException(java.sql.SQLException) StateMap(org.apache.nifi.components.state.StateMap) ArrayList(java.util.ArrayList) StateManager(org.apache.nifi.components.state.StateManager) ResultSet(java.sql.ResultSet) FlowFile(org.apache.nifi.flowfile.FlowFile) Statement(java.sql.Statement) Connection(java.sql.Connection) IOException(java.io.IOException) DatabaseMetaData(java.sql.DatabaseMetaData) ComponentLog(org.apache.nifi.logging.ComponentLog) DatabaseAdapter(org.apache.nifi.processors.standard.db.DatabaseAdapter) StopWatch(org.apache.nifi.util.StopWatch) JdbcCommon(org.apache.nifi.processors.standard.util.JdbcCommon) AtomicLong(java.util.concurrent.atomic.AtomicLong) ProcessException(org.apache.nifi.processor.exception.ProcessException) DBCPService(org.apache.nifi.dbcp.DBCPService) Map(java.util.Map) StateMap(org.apache.nifi.components.state.StateMap) HashMap(java.util.HashMap)

Example 2 with NORMALIZE_NAMES_FOR_AVRO

use of org.apache.nifi.processors.standard.util.JdbcCommon.NORMALIZE_NAMES_FOR_AVRO in project nifi by apache.

the class ExecuteSQL method onTrigger.

@Override
public void onTrigger(final ProcessContext context, final ProcessSession session) throws ProcessException {
    FlowFile fileToProcess = null;
    if (context.hasIncomingConnection()) {
        fileToProcess = session.get();
        // we know that we should run only if we have a FlowFile.
        if (fileToProcess == null && context.hasNonLoopConnection()) {
            return;
        }
    }
    final ComponentLog logger = getLogger();
    final DBCPService dbcpService = context.getProperty(DBCP_SERVICE).asControllerService(DBCPService.class);
    final Integer queryTimeout = context.getProperty(QUERY_TIMEOUT).asTimePeriod(TimeUnit.SECONDS).intValue();
    final boolean convertNamesForAvro = context.getProperty(NORMALIZE_NAMES_FOR_AVRO).asBoolean();
    final Boolean useAvroLogicalTypes = context.getProperty(USE_AVRO_LOGICAL_TYPES).asBoolean();
    final Integer defaultPrecision = context.getProperty(DEFAULT_PRECISION).evaluateAttributeExpressions().asInteger();
    final Integer defaultScale = context.getProperty(DEFAULT_SCALE).evaluateAttributeExpressions().asInteger();
    final StopWatch stopWatch = new StopWatch(true);
    final String selectQuery;
    if (context.getProperty(SQL_SELECT_QUERY).isSet()) {
        selectQuery = context.getProperty(SQL_SELECT_QUERY).evaluateAttributeExpressions(fileToProcess).getValue();
    } else {
        // If the query is not set, then an incoming flow file is required, and expected to contain a valid SQL select query.
        // If there is no incoming connection, onTrigger will not be called as the processor will fail when scheduled.
        final StringBuilder queryContents = new StringBuilder();
        session.read(fileToProcess, in -> queryContents.append(IOUtils.toString(in, Charset.defaultCharset())));
        selectQuery = queryContents.toString();
    }
    int resultCount = 0;
    try (final Connection con = dbcpService.getConnection();
        final PreparedStatement st = con.prepareStatement(selectQuery)) {
        // timeout in seconds
        st.setQueryTimeout(queryTimeout);
        if (fileToProcess != null) {
            JdbcCommon.setParameters(st, fileToProcess.getAttributes());
        }
        logger.debug("Executing query {}", new Object[] { selectQuery });
        boolean results = st.execute();
        while (results) {
            FlowFile resultSetFF;
            if (fileToProcess == null) {
                resultSetFF = session.create();
            } else {
                resultSetFF = session.create(fileToProcess);
                resultSetFF = session.putAllAttributes(resultSetFF, fileToProcess.getAttributes());
            }
            final AtomicLong nrOfRows = new AtomicLong(0L);
            resultSetFF = session.write(resultSetFF, out -> {
                try {
                    final ResultSet resultSet = st.getResultSet();
                    final JdbcCommon.AvroConversionOptions options = JdbcCommon.AvroConversionOptions.builder().convertNames(convertNamesForAvro).useLogicalTypes(useAvroLogicalTypes).defaultPrecision(defaultPrecision).defaultScale(defaultScale).build();
                    nrOfRows.set(JdbcCommon.convertToAvroStream(resultSet, out, options, null));
                } catch (final SQLException e) {
                    throw new ProcessException(e);
                }
            });
            long duration = stopWatch.getElapsed(TimeUnit.MILLISECONDS);
            // set attribute how many rows were selected
            resultSetFF = session.putAttribute(resultSetFF, RESULT_ROW_COUNT, String.valueOf(nrOfRows.get()));
            resultSetFF = session.putAttribute(resultSetFF, RESULT_QUERY_DURATION, String.valueOf(duration));
            resultSetFF = session.putAttribute(resultSetFF, CoreAttributes.MIME_TYPE.key(), JdbcCommon.MIME_TYPE_AVRO_BINARY);
            logger.info("{} contains {} Avro records; transferring to 'success'", new Object[] { resultSetFF, nrOfRows.get() });
            session.getProvenanceReporter().modifyContent(resultSetFF, "Retrieved " + nrOfRows.get() + " rows", duration);
            session.transfer(resultSetFF, REL_SUCCESS);
            resultCount++;
            // are there anymore result sets?
            try {
                results = st.getMoreResults();
            } catch (SQLException ex) {
                results = false;
            }
        }
        // pass the original flow file down the line to trigger downstream processors
        if (fileToProcess != null) {
            if (resultCount > 0) {
                session.remove(fileToProcess);
            } else {
                fileToProcess = session.write(fileToProcess, JdbcCommon::createEmptyAvroStream);
                session.transfer(fileToProcess, REL_SUCCESS);
            }
        }
    } catch (final ProcessException | SQLException e) {
        // pass the original flow file down the line to trigger downstream processors
        if (fileToProcess == null) {
            // This can happen if any exceptions occur while setting up the connection, statement, etc.
            logger.error("Unable to execute SQL select query {} due to {}. No FlowFile to route to failure", new Object[] { selectQuery, e });
            context.yield();
        } else {
            if (context.hasIncomingConnection()) {
                logger.error("Unable to execute SQL select query {} for {} due to {}; routing to failure", new Object[] { selectQuery, fileToProcess, e });
                fileToProcess = session.penalize(fileToProcess);
            } else {
                logger.error("Unable to execute SQL select query {} due to {}; routing to failure", new Object[] { selectQuery, e });
                context.yield();
            }
            session.transfer(fileToProcess, REL_FAILURE);
        }
    }
}
Also used : StandardValidators(org.apache.nifi.processor.util.StandardValidators) Connection(java.sql.Connection) CapabilityDescription(org.apache.nifi.annotation.documentation.CapabilityDescription) USE_AVRO_LOGICAL_TYPES(org.apache.nifi.processors.standard.util.JdbcCommon.USE_AVRO_LOGICAL_TYPES) EventDriven(org.apache.nifi.annotation.behavior.EventDriven) ComponentLog(org.apache.nifi.logging.ComponentLog) PropertyDescriptor(org.apache.nifi.components.PropertyDescriptor) ProcessException(org.apache.nifi.processor.exception.ProcessException) ArrayList(java.util.ArrayList) DEFAULT_SCALE(org.apache.nifi.processors.standard.util.JdbcCommon.DEFAULT_SCALE) HashSet(java.util.HashSet) SQLException(java.sql.SQLException) Charset(java.nio.charset.Charset) WritesAttributes(org.apache.nifi.annotation.behavior.WritesAttributes) Relationship(org.apache.nifi.processor.Relationship) ResultSet(java.sql.ResultSet) DEFAULT_PRECISION(org.apache.nifi.processors.standard.util.JdbcCommon.DEFAULT_PRECISION) Requirement(org.apache.nifi.annotation.behavior.InputRequirement.Requirement) ReadsAttributes(org.apache.nifi.annotation.behavior.ReadsAttributes) FlowFile(org.apache.nifi.flowfile.FlowFile) NORMALIZE_NAMES_FOR_AVRO(org.apache.nifi.processors.standard.util.JdbcCommon.NORMALIZE_NAMES_FOR_AVRO) ProcessContext(org.apache.nifi.processor.ProcessContext) Set(java.util.Set) ProcessSession(org.apache.nifi.processor.ProcessSession) WritesAttribute(org.apache.nifi.annotation.behavior.WritesAttribute) PreparedStatement(java.sql.PreparedStatement) TimeUnit(java.util.concurrent.TimeUnit) AtomicLong(java.util.concurrent.atomic.AtomicLong) IOUtils(org.apache.commons.io.IOUtils) List(java.util.List) InputRequirement(org.apache.nifi.annotation.behavior.InputRequirement) OnScheduled(org.apache.nifi.annotation.lifecycle.OnScheduled) JdbcCommon(org.apache.nifi.processors.standard.util.JdbcCommon) StopWatch(org.apache.nifi.util.StopWatch) AbstractProcessor(org.apache.nifi.processor.AbstractProcessor) Tags(org.apache.nifi.annotation.documentation.Tags) DBCPService(org.apache.nifi.dbcp.DBCPService) CoreAttributes(org.apache.nifi.flowfile.attributes.CoreAttributes) Collections(java.util.Collections) ReadsAttribute(org.apache.nifi.annotation.behavior.ReadsAttribute) FlowFile(org.apache.nifi.flowfile.FlowFile) SQLException(java.sql.SQLException) Connection(java.sql.Connection) PreparedStatement(java.sql.PreparedStatement) ComponentLog(org.apache.nifi.logging.ComponentLog) StopWatch(org.apache.nifi.util.StopWatch) AtomicLong(java.util.concurrent.atomic.AtomicLong) ProcessException(org.apache.nifi.processor.exception.ProcessException) DBCPService(org.apache.nifi.dbcp.DBCPService) ResultSet(java.sql.ResultSet)

Aggregations

Connection (java.sql.Connection)2 ResultSet (java.sql.ResultSet)2 SQLException (java.sql.SQLException)2 ArrayList (java.util.ArrayList)2 Collections (java.util.Collections)2 HashSet (java.util.HashSet)2 List (java.util.List)2 Set (java.util.Set)2 TimeUnit (java.util.concurrent.TimeUnit)2 AtomicLong (java.util.concurrent.atomic.AtomicLong)2 InputRequirement (org.apache.nifi.annotation.behavior.InputRequirement)2 Requirement (org.apache.nifi.annotation.behavior.InputRequirement.Requirement)2 WritesAttribute (org.apache.nifi.annotation.behavior.WritesAttribute)2 WritesAttributes (org.apache.nifi.annotation.behavior.WritesAttributes)2 CapabilityDescription (org.apache.nifi.annotation.documentation.CapabilityDescription)2 Tags (org.apache.nifi.annotation.documentation.Tags)2 OnScheduled (org.apache.nifi.annotation.lifecycle.OnScheduled)2 PropertyDescriptor (org.apache.nifi.components.PropertyDescriptor)2 DBCPService (org.apache.nifi.dbcp.DBCPService)2 FlowFile (org.apache.nifi.flowfile.FlowFile)2