Flat files - Connect_CDC - connect_cdc_mimix_share - Latest

You have the ability to use a flat file as the target for data storage purposes. The target is a flat, sequential file, not a random access file. A flat file has no interrelationship between its data records. Data rows are appended one after the other sequentially, similar to a journal.

Writing to a flat file is advantageous in some situations:

• A flat file can function as an intermediate target or a final target, thus providing a target for either temporary or permanent data storage.

• A flat file is ideal for storing data or records that are not used frequently.

The following features are currently available in Connect CDC:

• Only Replication requests are supported.

• Source tables mapped to target files are unidirectional.

• One source table may be mapped to only one target file on a File System server.

• The source table may be mapped to files on different FileSystems.

• The File System is a flat/sequential file that uses the Comma Separated Value (CSV) format. It is saved with a comma as the delimiter, although the delimiter can be changed. If the delimiter is changed, the file is still saved as a .csv file.

• Datatypes, except long character, long and short binary types, are written as a string.

• If a source table contains a mapped column of type Unicode, the target file properties defaults to a UTF-8 character set.

Connect CDC maintains the following guidelines when writing the target file:

• The double quote character (") quotes each value. This includes the null string representation.

• Empty strings have two quote characters with no character inside.

• New line character(s), outside of quoted values delineate a row. The new line character is dependent on the operating system’s default new line character.

• The first line of a file contains a list of delimited field names.

The table below specifies the datatypes supported for use with flat files.