Skip to content

Hive: Support DISTRIBUTE BY and SORT BY in window specs #2340

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

Open
wugeer wants to merge 4 commits into
base: main
Choose a base branch
from
+96 −10
Open

Hive: Support DISTRIBUTE BY and SORT BY in window specs #2340

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
59f27eb
fix merge conflict
wugeer Jun 18, 2026
c7fb3f6
remove `supports_window_spec_distribute_sort` func and redundant `def…
wugeer Jun 18, 2026
75d7457
remove redundant `default` attributes
wugeer Jun 18, 2026
4c89b05
Merge branch 'main' into fix_over_distibute_by_and_sort_by
wugeer Jun 22, 2026
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
54 changes: 52 additions & 2 deletions src/ast/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2269,12 +2269,56 @@ pub struct WindowSpec {
pub window_name: Option<Ident>,
/// `OVER (PARTITION BY ...)`
pub partition_by: Vec<Expr>,
/// The kind of partitioning clause used in the window specification.
pub partition_by_kind: WindowPartitionByKind,

@iffyio iffyio Jun 18, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

for partition and order by kind, can we instead wrap the field with the enum instead of introducing a new field? i.e.

partition_by: WindowPartitionByKind

// where 
enum WindowPartitionByKind {
     PartitionBy(Vec<Expr>)
     DistributeBy(Vec<Expr>)
}

@wugeer wugeer Jun 18, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a good idea, but the change is very disruptive. Wouldn't it be better to open a separate PR for it?

@iffyio iffyio Jun 23, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

not sure it would make sense as a separate PR given this one, since such a PR would be undoing the changes made by this one

/// `OVER (ORDER BY ...)`
pub order_by: Vec<OrderByExpr>,
/// The kind of ordering clause used in the window specification.
pub order_by_kind: WindowOrderByKind,
/// `OVER (window frame)`
pub window_frame: Option<WindowFrame>,
}

/// The kind of partitioning clause in a window specification.
#[derive(Debug, Clone, Copy, PartialEq, PartialOrd, Eq, Ord, Hash)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[cfg_attr(feature = "visitor", derive(Visit, VisitMut))]
pub enum WindowPartitionByKind {
/// `PARTITION BY`
Partition,
/// Hive `DISTRIBUTE BY`
Distribute,
}

impl fmt::Display for WindowPartitionByKind {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.write_str(match self {
Self::Partition => "PARTITION BY",
Self::Distribute => "DISTRIBUTE BY",
})
}
}

/// The kind of ordering clause in a window specification.
#[derive(Debug, Clone, Copy, PartialEq, PartialOrd, Eq, Ord, Hash)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[cfg_attr(feature = "visitor", derive(Visit, VisitMut))]
pub enum WindowOrderByKind {
/// `ORDER BY`
Order,
/// Hive `SORT BY`
Sort,
}

impl fmt::Display for WindowOrderByKind {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.write_str(match self {
Self::Order => "ORDER BY",
Self::Sort => "SORT BY",
})
}
}

impl fmt::Display for WindowSpec {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
let mut is_first = true;
Expand All @@ -2292,7 +2336,8 @@ impl fmt::Display for WindowSpec {
is_first = false;
write!(
f,
"PARTITION BY {}",
"{} {}",
self.partition_by_kind,
display_comma_separated(&self.partition_by)
)?;
}
Expand All @@ -2301,7 +2346,12 @@ impl fmt::Display for WindowSpec {
SpaceOrNewline.fmt(f)?;
}
is_first = false;
write!(f, "ORDER BY {}", display_comma_separated(&self.order_by))?;
write!(
f,
"{} {}",
self.order_by_kind,
display_comma_separated(&self.order_by)
)?;
}
if let Some(window_frame) = &self.window_frame {
if !is_first {
Expand Down
35 changes: 27 additions & 8 deletions src/parser/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20275,15 +20275,32 @@ impl<'a> Parser<'a> {
_ => None,
};

let partition_by = if self.parse_keywords(&[Keyword::PARTITION, Keyword::BY]) {
self.parse_comma_separated(Parser::parse_expr)?
} else {
vec![]
};
let order_by = if self.parse_keywords(&[Keyword::ORDER, Keyword::BY]) {
self.parse_comma_separated(Parser::parse_order_by_expr)?
let (partition_by_kind, partition_by) =
if self.parse_keywords(&[Keyword::PARTITION, Keyword::BY]) {
(
WindowPartitionByKind::Partition,
self.parse_comma_separated(Parser::parse_expr)?,
)
} else if self.parse_keywords(&[Keyword::DISTRIBUTE, Keyword::BY]) {
(
WindowPartitionByKind::Distribute,
self.parse_comma_separated(Parser::parse_expr)?,
)
} else {
(WindowPartitionByKind::Partition, vec![])
};
let (order_by_kind, order_by) = if self.parse_keywords(&[Keyword::ORDER, Keyword::BY]) {
(
WindowOrderByKind::Order,
self.parse_comma_separated(Parser::parse_order_by_expr)?,
)
} else if self.parse_keywords(&[Keyword::SORT, Keyword::BY]) {
(
WindowOrderByKind::Sort,
self.parse_comma_separated(Parser::parse_order_by_expr)?,
)
} else {
vec![]
(WindowOrderByKind::Order, vec![])
};

let window_frame = if !self.consume_token(&Token::RParen) {
Expand All @@ -20296,7 +20313,9 @@ impl<'a> Parser<'a> {
Ok(WindowSpec {
window_name,
partition_by,
partition_by_kind,
order_by,
order_by_kind,
window_frame,
})
}
Expand Down
8 changes: 8 additions & 0 deletions tests/sqlparser_common.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3095,6 +3095,7 @@ fn parse_select_qualify() {
over: Some(WindowType::WindowSpec(WindowSpec {
window_name: None,
partition_by: vec![Expr::Identifier(Ident::new("p"))],
partition_by_kind: WindowPartitionByKind::Partition,
order_by: vec![OrderByExpr {
expr: Expr::Identifier(Ident::new("o")),
options: OrderByOptions {
Expand All @@ -3103,6 +3104,7 @@ fn parse_select_qualify() {
},
with_fill: None,
}],
order_by_kind: WindowOrderByKind::Order,
window_frame: None,
})),
within_group: vec![]
Expand Down Expand Up @@ -5815,6 +5817,7 @@ fn parse_window_functions() {
over: Some(WindowType::WindowSpec(WindowSpec {
window_name: None,
partition_by: vec![],
partition_by_kind: WindowPartitionByKind::Partition,
order_by: vec![OrderByExpr {
expr: Expr::Identifier(Ident::new("dt")),
options: OrderByOptions {
Expand All @@ -5823,6 +5826,7 @@ fn parse_window_functions() {
},
with_fill: None,
}],
order_by_kind: WindowOrderByKind::Order,
window_frame: None,
})),
within_group: vec![],
Expand Down Expand Up @@ -6037,6 +6041,7 @@ fn test_parse_named_window() {
NamedWindowExpr::WindowSpec(WindowSpec {
window_name: None,
partition_by: vec![],
partition_by_kind: WindowPartitionByKind::Partition,
order_by: vec![OrderByExpr {
expr: Expr::Identifier(Ident {
value: "C12".to_string(),
Expand All @@ -6049,6 +6054,7 @@ fn test_parse_named_window() {
},
with_fill: None,
}],
order_by_kind: WindowOrderByKind::Order,
window_frame: None,
}),
),
Expand All @@ -6065,7 +6071,9 @@ fn test_parse_named_window() {
quote_style: None,
span: Span::empty(),
})],
partition_by_kind: WindowPartitionByKind::Partition,
order_by: vec![],
order_by_kind: WindowOrderByKind::Order,
window_frame: None,
}),
),
Expand Down
7 changes: 7 additions & 0 deletions tests/sqlparser_hive.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -569,6 +569,13 @@ fn parse_create_table_with_map_column_comment() {
);
}

#[test]
fn parse_row_number_window_function() {
hive().verified_stmt(
"SELECT row_number() OVER (DISTRIBUTE BY age SORT BY update_time DESC) AS row_num FROM sdl.xxx",
);
}

fn hive() -> TestedDialects {
TestedDialects::new(vec![Box::new(HiveDialect {})])
}
Expand Down
2 changes: 2 additions & 0 deletions tests/sqlparser_sqlite.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -439,7 +439,9 @@ fn parse_window_function_with_filter() {
over: Some(WindowType::WindowSpec(WindowSpec {
window_name: None,
partition_by: vec![],
partition_by_kind: WindowPartitionByKind::Partition,
order_by: vec![],
order_by_kind: WindowOrderByKind::Order,
window_frame: None,
})),
filter: Some(Box::new(Expr::Identifier(Ident::new("y")))),
Expand Down
Loading