Incremental reading is a way to read only the new records since the last graph run. This way, you can avoid reading already processed records. Incremental reading allows you to read new records from a single file as well as new records from multiple files. If a file URL possibly matches new files, it can read records from new files.
Incremental reading can be set with the Incremental file and Incremental key attributes. The Incremental key is a string that holds the information about read records/files. This key is stored in the Incremental file attribute. This way, the component reads only the records or files that have not been marked in Incremental file.
Readers with incremental reading are:
The component DatabaseReader which reads data from databases performs this incremental reading in a different way.
Incremental Reading in Database Components
Unlike other incremental readers, in a database component, more database columns can be evaluated and used as key fields. Incremental key is a sequence of the following individual expression separated by a semicolon:
key01=MAX(EmployeeID);key02=FIRST(CustomerID)!20). The functions that can be selected are:
At the same time, when you define an Incremental key, you also need to add these key parts to the Query.
In the query, a part of the
where sentence will appear; e.g.
where db_field1 > #key01 and db_field2 < #key02.
This way, you can limit which records will be read next time.
It depends on the values of
Only the records that satisfy the condition specified by the query will be read. These key fields values are stored in the Incremental file. To define Incremental key, click this attribute row and, by clicking the Plus or Minus buttons in the Define incremental key dialog, add or remove key names and select db field names and function names. Each one of the last two is to be selected from a combo list of possible values.
Since CloverETL Designer 2.8.1, you can also define