Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix(python): fix panic when doing a scan_parquet with hive partioning #15381

Merged
merged 2 commits into from
Mar 29, 2024
Merged

fix(python): fix panic when doing a scan_parquet with hive partioning #15381

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
559258e
fix(python): fix panic when doing a scan_parquet with hive partioning
kszlim Mar 29, 2024
fd7558f
add test
ritchie46 Mar 29, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion crates/polars-plan/src/logical_plan/optimizer/cache_states.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -114,6 +114,7 @@ pub(super) fn set_cache_states(
lp_arena: &mut Arena<ALogicalPlan>,
expr_arena: &mut Arena<AExpr>,
scratch: &mut Vec<Node>,
hive_partition_eval: HiveEval<'_>,
verbose: bool,
) -> PolarsResult<()> {
let mut stack = Vec::with_capacity(4);
Expand Down Expand Up @@ -262,7 +263,7 @@ pub(super) fn set_cache_states(
// back to the cache node again
if !cache_schema_and_children.is_empty() {
let mut proj_pd = ProjectionPushDown::new();
let pred_pd = PredicatePushDown::new(Default::default()).block_at_cache(false);
let pred_pd = PredicatePushDown::new(hive_partition_eval).block_at_cache(false);
for (_cache_id, v) in cache_schema_and_children {
// # CHECK IF WE NEED TO REMOVE CACHES
// If we encounter multiple predicates we remove the cache nodes completely as we don't
Expand Down
9 changes: 8 additions & 1 deletion crates/polars-plan/src/logical_plan/optimizer/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,7 +187,14 @@ pub fn optimize(
lp_top = opt.optimize_loop(&mut rules, expr_arena, lp_arena, lp_top)?;

if members.has_joins_or_unions && members.has_cache {
cache_states::set_cache_states(lp_top, lp_arena, expr_arena, scratch, verbose)?;
cache_states::set_cache_states(
lp_top,
lp_arena,
expr_arena,
scratch,
hive_partition_eval,
verbose,
)?;
}

// This one should run (nearly) last as this modifies the projections
Expand Down
6 changes: 6 additions & 0 deletions py-polars/tests/unit/io/test_hive.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,6 +88,12 @@ def test_hive_partitioned_predicate_pushdown_skips_correct_number_of_files(
assert q.filter(pl.col("a").is_in([1, 4])).collect().shape == (2, 2)
assert "hive partitioning: skipped 3 files" in capfd.readouterr().err

# Ensure the CSE can work with hive partitions.
q = q.filter(pl.col("a").gt(2))
assert q.join(q, on="a", how="left").collect(comm_subplan_elim=True).to_dict(
as_series=False
) == {"d": [3, 4], "a": [3, 4], "d_right": [3, 4]}


@pytest.mark.skip(
reason="Broken by pyarrow 15 release: https://github.com/pola-rs/polars/issues/13892"
Expand Down
Loading