README.md

FBP-ETL

Componentizing ReadJDBC.java (3)

"Componentizing" ReadJDBC.java (continued)

In the previous step, we said in connection with the code in https://github.com/jpaulm/fbp-etl/blob/master/src/main/java/com/jpaulmorrison/Step12/code/components/ReadJDBC.java that

• there are no table column names or IP object (Book) field names hard-wired in the code
• what is (currently) hard-wired is the correspondence between VARCHAR and String, DECIMAL and BigDecimal, etc.

We are going to try to relax these constraints.

First, we are going to add a JSON file of correspondences between table columns and Book fields. This will be passed to ReadJDBC in a separate input port. The IP contents will look like this:

[{"colName": "id", 
  "objField": "id2"},
  {"colName": "title", 
  "objField": "title2"},
  {"colName": "author", 
  "objField": "author2"},
  {"colName": "price", 
  "objField": "price2"},
  {"colName": "qty", 
  "objField": "qty2"}
]

For convenience, we will put this in an IIP - however, as far as ReadJDBC is concerned, it could be in an IIP, or on a file - it doesn't care.

So here is our network, with the field correspondences added in:

The ReadJDBC component will now generate the getxxx method for each column based on the target field type - ReadJDBC will produce a message if there is a mismatch.

This completes the development stage for this component. If error messages, or other run-time exceptions, occur, please raise an issue in Issues.

We now have the "front end" of a generalized FBP-based ETL system.