Union

Purpose: Combine rows from two or more sources into a single result set. All sources should have compatible column structures.

How to use:

• Connect multiple sources with similar column structures
• The node combines all rows from all inputs
• Columns are matched by name
• Missing columns in some inputs are filled with NULL values

Data transformation:

• All rows from all inputs are combined into one output
• Rows maintain their original data values
• The output contains the union of all columns from all inputs
• Rows are not deduplicated (same as SQL's UNION ALL)
• No particular ordering is guaranteed unless you add a Sort node after

Example: Combine slices from different tracks into a single view. Or merge events from multiple processes for analysis.

Column handling:

• If all inputs have the same columns, the output has those columns
• If inputs have different columns, the output has all columns, with NULL for missing values
• Column types should be compatible across inputs

SQL equivalent: SELECT * FROM source1 UNION ALL SELECT * FROM source2 UNION ALL SELECT * FROM source3