This project is up for adoption - Feel free to fork it.

Informix CDC Java client

This is an Informix CDC client that allows you to capture and unserialize changes on one or multiple tables of the database

It implements methods described by IBM in this sections: https://www.ibm.com/support/knowledgecenter/en/SSGU8G_12.1.0/com.ibm.cdc.doc/ids_cdc_057.htm

Features - Work in progress

This project is still a work in progress, some data types are not correctly unserialized.

Data type	Status	Comment
CHAR	Supported
SMALLINT	Supported
INTEGER	Supported
FLOAT	Supported
SMALLFLOAT	Supported
DECIMAL	Supported
SERIAL	Supported	Unit test fails and needs to be adapted
DATE	Not supported
MONEY	Supported
DATETIME	Not supported	Supported up to "YEAR TO SECONDS" precision
VARCHAR	Supported
NCHAR	Supported
NVARCHAR	Supported
INT8	Not supported
SERIAL8	Supported	Unit test fails and needs to be adapted
LVARCHAR	Supported
BOOLEAN	Supported
BIGINT	Not supported

A unit test is provided to test the various data types.

Use the provided docker-compose file to launch an Informix database instance, and use this command to create the CDC schema before running the test.

dbaccess -e informix < $INFORMIXDIR/etc/syscdcv1.sql

Usage

See Unit tests for detailed use.

// 1 - Create the CDCConnection
onnectionDetails connectionDetails = new CDCConnectionDetails(connectionString);
connectionDetails.setUsername(username);
connectionDetails.setPassword(password);
connectionDetails.setInformixServer(informixServer);
connectionDetails.setMaxRecordsPerReturn(maxRecordsPerReturn);
connectionDetails.setCdcTimeout(timeout);
connectionDetails.setInterfaceMajorVersion(1);
connectionDetails.setInterfaceMinorVersion(1);
connectionDetails.setDebugging(false);
cdcConnection = new CDCConnection(connectionDetails);
cdcConnection.connect();

// 2 - Enable capture on the table
cdcConnection.enableCapture(catalog + ":" + schema + "." + table, columnNames);

// 3 - Wait for changes
boolean exit = false;
do {
  log.debug("Reading...");
  cdcRecords = cdcConnection.readData();
  if (!cdcRecords.isEmpty()) {
    log.debug("Got " + cdcRecords.size() + " record(s) (" + cdcConnection.getConnectionDetails().getMaxRecordsPerReturn() + "max)");
  }
  for (CDCRecord cdcRecord : cdcRecords) {
    // Process record (see @4)
    processRecord(cdcRecord);
    // Exit if no change has been captured (timeout)
    if (cdcRecord.isTimeoutRecord()) { exit = true; }
  }
} while (!exit);

// 4 - Process records
private void processRecord(CDCRecord record) {
    // Ignore timeout records
    if (!record.isTimeoutRecord()) {
    	log.debug("{}", record);
    }
    // A single Metadata record is fired when starting capture, save this records for later, they'll be used for unserialization
    if (record.isMetadataRecord()) {
    	CDCMetadataRecord cdcMetadataRecord = (CDCMetadataRecord) record;
    	metadataRecords.put(cdcMetadataRecord.getUserData(), cdcMetadataRecord);
    }
    // An operationnal record is fired when  an INSERT/UPDATE/DELETE operation is done
    if (record.isOperationalRecord()) {
      CDCOperationRecord cdcOperationRecord = (CDCOperationRecord) record;
      CDCMetadataRecord cdcMetadataRecord = metadataRecords.get(cdcOperationRecord.getUserData());
      // Parse the payload data for the operation record to get the column values
      Map<String, Object> cdcOperationRecordPayload = CDCMessageFactory.parseCDCRecordPayload(cdcOperationRecord, cdcMetadataRecord);
      log.info("{}", cdcOperationRecordPayload);
    }
}

Implementation details

Implementation for the CDC client is loosely based on sources published by pushtechnology: https://www.programcreek.com/java-api-examples/index.php?source_dir=adapters-master/cdc/src/com/pushtechnology/diffusion/api/adapters/cdc/CDCConnection.java#

Implementation for the data unserializing is adapted from IBM's C sample: https://www.ibm.com/support/knowledgecenter/fr/SSGU8G_11.50.0/com.ibm.cdc.doc/ids_cdc_060.htm