2.1.0

• New CLI arguments syntax to support per-input and per-output options
• Retains the column names as they come, even from JSON - i.e. SELECT foo.bar.baz FROM myJson is possible.

Switched to Kotlin

1.14.0

Switch the language to Kotlin.

1.13.1 - IT fix

Only releasing sources because of a tests input files fix.

1.13.0 - Uses column types for JSON output (fixes numbers and boolean format)

See: Proper JSON output for BOOLEAN columns #30
Also, a bunch of new tests were added.

JSON nulls

Empty CSV cells are now represented as null in JSON.

1.11.1 Bug fixes

Fix i#29  Outputs from subgroups are not sorted properly #29

1.11.0 Allow running SQL against each table individually.

For the cases when the input files are combined and their schema is not consistent, CSV Cruncher still can be useful by running the SQL against each of these schema versions.

However, that prevents CSV Cruncher to treat all files as sources to the same table, therefore multiple tables are created, like, foo_1, foo_2 etc, one for each schema version (columns combination).

This release implements this feature.
It also needed a major refactoring.

1.10.0 group combined files by CSV structure, add --include --exclude

Fix filtering bug, fix counter column SELECT bug, fix skipping single…

… input file, improve logging, add --exclude LOAD* to all E2E tests.

1.9.0 - Fix #20 SQLSyntaxErrorException 'duplicate column name in derived table'

Fix #20

1.8.0 - Fix concat; Delete files at output path; add --ignoreFirstLines and --ignoreLinesMatching

Prevent HSQLDB bug that proves when `SELECT *` and --rowNumbers is us…

…ed. Also add a test for it.