alpaylan opened a new issue, #2806:
URL: https://github.com/apache/datafusion-sqlparser-rs/issues/2806
Each element of `ROLLUP`, `CUBE` and `GROUPING SETS` treats a leading `(` as
its own sublist delimiter, which breaks parenthesized expressions:
```
SELECT a FROM t GROUP BY ROLLUP (a, (SELECT 1))
-> sql parser error: Expected: ), found: 1 at Line: 1, Column: 45
SELECT a FROM t GROUP BY CUBE (a, (SELECT 1))
SELECT a FROM t GROUP BY GROUPING SETS (a, (SELECT 1))
-> same error
SELECT a FROM t GROUP BY ROLLUP (a, ((b)))
-> displays as: SELECT a FROM t GROUP BY ROLLUP (a, (b))
-> reparses as ROLLUP (a, b): element [Nested(b)] becomes [b]
```
PostgreSQL 18 runs `SELECT a FROM t GROUP BY ROLLUP (a, (SELECT 1))`. Its
docs: "The individual elements of a CUBE or ROLLUP clause may be either
individual expressions, or sublists of elements in parentheses"
(https://www.postgresql.org/docs/current/queries-table-expressions.html#QUERIES-GROUPING-SETS).
The second case changes only the AST, since PostgreSQL treats `((b))` and
`(b)` the same.
`parse_group_by_expr` parses each element with `parse_tuple(true, true)`,
which consumes the `(` of `(SELECT 1)` and then reads `SELECT` as an
identifier. Display for `Rollup`/`Cube` writes a one-element set without its
own parentheses, so `[Nested(b)]` prints as `(b)`.
Tested with PostgreSqlDialect on main ce12f1f6 (also 0.63.0 and 0.58.0).
Found with a property-based fuzzer (round trip check); I used AI assistance
for triage and verified every case above.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
