Hey folks,

I found some time to finish the work I started on the Apache Calcite adapter.
It is merged into develop in the integrations/apache-calcite module.
The adapter is based on the scraper (plc4j/utils/scraper).

Configuration is done with two yml files

  *   Calcite model definition (see src/test/resources/mode.yml)
  *   Plc-scraper configuration (see src/test/resources/example.yml)

The calcite model exposes two Schemas: PLC4X and PLC4X-TABLES.
Both contain the same “tables”. In the PLC4X Schema these are Streamable tables.
This means (roughly speaking) that they can only be used in Queries starting 
with

“SELECT STREAM … FROM …”

In the PLC4X-TABLES Schema the tables are real SQL Tables with n rows, where n 
is the “limits” parameter in the model.yml definition.

For each job in the plc scraper there is one table in both schemas (one 
truncated to a real table, the other as Streaming table).
The layout is

  *   Timestamp
  *   Source
  *   Fields from the Scraper Configuration

For examples you can look at the Tests or ask myself.. documentation is not 
perfect… yet.

Best
Julian

Reply via email to