[PySpark] - Add partial support for window function

[mariotaddeucci]

Initial support for window functions by introducing a PySpark-like Window API and a WindowSpec class.

The implementation is partial: partitioning, basic ordering-by-names, and frame specification (rowsBetween / rangeBetween) are supported and converted to SQL window clauses, but extracting ordering direction from Column expressions is not yet implemented (see the TODO in WindowSpec._columns_as_str). Using Column expressions in orderBy currently raises a ContributionsAcceptedError.

Additionally, a set of functions was added for convenient use with .over(window):

• row_number
• rank
• dense_rank
• cume_dist
• percent_rank
• lag
• lead
• nth_value

[mariotaddeucci]

I don't know how but the next step is to implement proper handling of Column expressions in ordering (including asc/desc).

[Copilot]

Copilot was unable to review this pull request because the user who requested the review is ineligible. To be eligible to request a review, you need a paid Copilot license, or your organization must enable Copilot code review.

[aaron-ang]

Hi @binste I would love to see this land. Thanks for the foundation here.

If you're not actively working on it, happy to pick it up. I can:

1. rebase onto current main
2. resolve the conflict in functions.py (a few new aggregate funcs landed adjacent to this block)
3. finish the TODO in WindowSpec._columns_as_str so orderBy(col.desc()) / orderBy(col.asc()) works instead of raising ContributionsAcceptedError.

The plan is to introspect the underlying Expression produced by Column.desc() / Column.asc() and emit the matching DESC / ASC (and NULLS FIRST/LAST for *_nulls_* variants) into the SQL OVER() clause.

@evertlammerts: is this PR still on the roadmap? If a fresh take is welcome, I can put up a continuation PR crediting the original author.

[binste]

Hi @aaron-ang ! Maybe a mix-up, I think I was never involved in this PR (and am also not a duckdb maintainer) but fully agree that it would be great to see this land. Good luck and thanks for picking it up!

[aaron-ang]

hi @binste, apologies for the mixup. i meant to tag @mariotaddeucci.