"REST join" - map the rows to data from REST requests

[OndraZizka]

In order to make getting data from a REST API easier,

a result of a SELECT could be mapped to the data from REST API in this way:

-in data.csv -sql "SELECT someId, someValue FROM ..." -map:rest="http://.../{someId}/{someValue}"

Each row of the SELECT result would form an URL using the template from -restJoin,
and the result would be added to a new table.

---

Alternatively:

This would need quite a few parameters on:

• how to create the request (HTTP method, url, headers, ...)
• how to parse the response (format, items location, ...),
• how to to create the output (name, format, json format, ...)
  and so on.

Some of these parameters are already implemented for -in and -out.
Therefore, it may be smarter to keep CsvCruncher a bit simpler, RISC-like, and instead, implement the REST-mapping as an alternative of -sql:

-in singleImport.csv -out mappedData.csv -map:rest="http://..."

Assuming there would be just a single input.
This would basically be just a limitation on supported use cases.

---

The mapping could perhaps be done using the Java-based SQL functions, which would

1. query
2. store to a table column
3. parse
4. extract the values

but this seems quite cumbersome way to do it.