BULK INSERT data using TUPLE into SET columns

This BULK INSERT statement uses the TRANSFORM...TUPLE clause to insert data in CSV format to IDSET and STRINGSET columns.

Before you begin

Run SHOW TABLE <tablename> to learn the structure of a destination table

BULK INSERT statement

BULK INSERT INTO tuple-set-target(
    _id,
    time_col,
    stringset_col,
    ideset_col,
    stringsetq_col,
    idesetq_col
  )
  MAP (
    0 STRING,
    1 TIMESTAMP,
    2 STRINGSET,
    3 IDSET
  )
  TRANSFORM(
    @0,
    @1,
    @2,
    @3,
    TUPLE(@1,@2),
    TUPLE(@1,@3)
  )
  FROM x'
    A,2013-07-15T01:18:46Z,stringset1, 1
    B,2014-07-15T01:18:46Z,stringset2, 2'
  WITH
    BATCHSIZE 10000
    format 'CSV'
    input 'INLINE';

Arguments

Argument	Description
`BULK INSERT INTO`	Define table and columns to be referenced by MAP and TRANSFORM clauses
`MAP`	Mapping starts at `0` for `_id` column followed by an integer corresponding to each subsequent column
`TRANSFORM`	Each integer corresponds to the mapping, with `TUPLE` functions applied to paired columns
`FROM x`	`x` indicates a multi-line string literal follows
`WITH`	States the number of rows to batch, data source `FORMAT` and the `INPUT` is included within the statement

Next step

SELECT FROM tuple-set-target