Rollup Summary — Solution & ExplanationProblems / Rollup Summary / EditorialApproachBuild the result in three layers and combine with UNION ALL:

Detail level: GROUP BY year, quarter — one row per (year, quarter).

Subtotal level: GROUP BY year — one row per year, with quarter = 'ALL'.

Grand total: No GROUP BY — one row with year = 'ALL', quarter = 'ALL'.

Key ConceptThis manual rollup pattern is the foundation of OLAP reporting. SQL's GROUP BY ROLLUP(year, quarter) can produce the same result automatically, but the UNION ALL approach is more explicit and portable.PySpark NoteUse unionByName (not union) to match columns by name rather than position, which is safer when column order varies.Hints

UNION ALL combines results from multiple SELECT statements with the same column structure.

Use CAST(year AS VARCHAR) so that the grand-total row ('ALL') has a compatible type.

Solve Rollup Summary yourself →