Skip to content

Add support for PostgreSQL's ORDER BY ... USING <operator> clause #2246

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
LucaCappelletti94 wants to merge 3 commits into
base: main
Choose a base branch
from
+287 −96
Open

Add support for PostgreSQL's ORDER BY ... USING <operator> clause #2246

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fffbe64
Add support for PostgreSQL's ORDER BY ... USING <operator> clause
LucaCappelletti94 Feb 26, 2026
0093a3e
Refactor ORDER BY ... USING operator handling to use ObjectName inste…
LucaCappelletti94 Feb 26, 2026
3b2da72
Reformatted code
LucaCappelletti94 Feb 26, 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
13 changes: 12 additions & 1 deletion src/ast/query.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2860,6 +2860,8 @@ impl fmt::Display for OrderBy {
pub struct OrderByExpr {
/// The expression to order by.
pub expr: Expr,
/// Optional PostgreSQL `USING <operator>` clause.
pub using_operator: Option<ObjectName>,
/// Ordering options such as `ASC`/`DESC` and `NULLS` behavior.
pub options: OrderByOptions,
/// Optional `WITH FILL` clause (ClickHouse extension) which specifies how to fill gaps.
Expand All @@ -2870,6 +2872,7 @@ impl From<Ident> for OrderByExpr {
fn from(ident: Ident) -> Self {
OrderByExpr {
expr: Expr::Identifier(ident),
using_operator: None,
options: OrderByOptions::default(),
with_fill: None,
}
Expand All @@ -2878,7 +2881,15 @@ impl From<Ident> for OrderByExpr {

impl fmt::Display for OrderByExpr {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "{}{}", self.expr, self.options)?;
write!(f, "{}", self.expr)?;
if let Some(using_operator) = &self.using_operator {
if using_operator.0.len() > 1 {
write!(f, " USING OPERATOR({using_operator})")?;
} else {
write!(f, " USING {using_operator}")?;
}
}
write!(f, "{}", self.options)?;
if let Some(ref with_fill) = self.with_fill {
write!(f, " {with_fill}")?
}
Expand Down
1 change: 1 addition & 0 deletions src/ast/spans.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2087,6 +2087,7 @@ impl Spanned for OrderByExpr {
fn span(&self) -> Span {
let OrderByExpr {
expr,
using_operator: _,
options: _,
with_fill,
} = self;
Expand Down
258 changes: 163 additions & 95 deletions src/parser/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -508,10 +508,10 @@ impl<'a> Parser<'a> {
Token::EOF => break,

// end of statement
Token::Word(word) => {
if expecting_statement_delimiter && word.keyword == Keyword::END {
break;
}
Token::Word(word)
if expecting_statement_delimiter && word.keyword == Keyword::END =>
{
break;
}
_ => {}
}
Expand Down Expand Up @@ -1298,41 +1298,40 @@ impl<'a> Parser<'a> {

let next_token = self.next_token();
match next_token.token {
t @ (Token::Word(_) | Token::SingleQuotedString(_)) => {
if self.peek_token_ref().token == Token::Period {
let mut id_parts: Vec<Ident> = vec![match t {
Token::Word(w) => w.into_ident(next_token.span),
Token::SingleQuotedString(s) => Ident::with_quote('\'', s),
_ => {
return Err(ParserError::ParserError(
"Internal parser error: unexpected token type".to_string(),
))
t @ (Token::Word(_) | Token::SingleQuotedString(_))
if self.peek_token_ref().token == Token::Period =>
{
let mut id_parts: Vec<Ident> = vec![match t {
Token::Word(w) => w.into_ident(next_token.span),
Token::SingleQuotedString(s) => Ident::with_quote('\'', s),
_ => {
return Err(ParserError::ParserError(
"Internal parser error: unexpected token type".to_string(),
))
}
}];

while self.consume_token(&Token::Period) {
let next_token = self.next_token();
match next_token.token {
Token::Word(w) => id_parts.push(w.into_ident(next_token.span)),
Token::SingleQuotedString(s) => {
// SQLite has single-quoted identifiers
id_parts.push(Ident::with_quote('\'', s))
}
}];

while self.consume_token(&Token::Period) {
let next_token = self.next_token();
match next_token.token {
Token::Word(w) => id_parts.push(w.into_ident(next_token.span)),
Token::SingleQuotedString(s) => {
// SQLite has single-quoted identifiers
id_parts.push(Ident::with_quote('\'', s))
}
Token::Placeholder(s) => {
// Snowflake uses $1, $2, etc. for positional column references
// in staged data queries like: SELECT t.$1 FROM @stage t
id_parts.push(Ident::new(s))
}
Token::Mul => {
return Ok(Expr::QualifiedWildcard(
ObjectName::from(id_parts),
AttachedToken(next_token),
));
}
_ => {
return self
.expected("an identifier or a '*' after '.'", next_token);
}
Token::Placeholder(s) => {
// Snowflake uses $1, $2, etc. for positional column references
// in staged data queries like: SELECT t.$1 FROM @stage t
id_parts.push(Ident::new(s))
}
Token::Mul => {
return Ok(Expr::QualifiedWildcard(
ObjectName::from(id_parts),
AttachedToken(next_token),
));
}
_ => {
return self.expected("an identifier or a '*' after '.'", next_token);
}
}
}
Expand Down Expand Up @@ -4990,10 +4989,10 @@ impl<'a> Parser<'a> {
loop {
match &self.peek_nth_token_ref(0).token {
Token::EOF => break,
Token::Word(w) => {
if w.quote_style.is_none() && terminal_keywords.contains(&w.keyword) {
break;
}
Token::Word(w)
if w.quote_style.is_none() && terminal_keywords.contains(&w.keyword) =>
{
break;
}
_ => {}
}
Expand Down Expand Up @@ -8173,70 +8172,60 @@ impl<'a> Parser<'a> {
Keyword::LINES,
Keyword::NULL,
]) {
Some(Keyword::FIELDS) => {
if self.parse_keywords(&[Keyword::TERMINATED, Keyword::BY]) {
Some(Keyword::FIELDS)
if self.parse_keywords(&[Keyword::TERMINATED, Keyword::BY]) =>
{
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::FieldsTerminatedBy,
char: self.parse_identifier()?,
});

if self.parse_keywords(&[Keyword::ESCAPED, Keyword::BY]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::FieldsTerminatedBy,
delimiter: HiveDelimiter::FieldsEscapedBy,
char: self.parse_identifier()?,
});

if self.parse_keywords(&[Keyword::ESCAPED, Keyword::BY]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::FieldsEscapedBy,
char: self.parse_identifier()?,
});
}
} else {
break;
}
}
Some(Keyword::COLLECTION) => {
Some(Keyword::COLLECTION)
if self.parse_keywords(&[
Keyword::ITEMS,
Keyword::TERMINATED,
Keyword::BY,
]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::CollectionItemsTerminatedBy,
char: self.parse_identifier()?,
});
} else {
break;
}
]) =>
{
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::CollectionItemsTerminatedBy,
char: self.parse_identifier()?,
});
}
Some(Keyword::MAP) => {
Some(Keyword::MAP)
if self.parse_keywords(&[
Keyword::KEYS,
Keyword::TERMINATED,
Keyword::BY,
]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::MapKeysTerminatedBy,
char: self.parse_identifier()?,
});
} else {
break;
}
]) =>
{
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::MapKeysTerminatedBy,
char: self.parse_identifier()?,
});
}
Some(Keyword::LINES) => {
if self.parse_keywords(&[Keyword::TERMINATED, Keyword::BY]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::LinesTerminatedBy,
char: self.parse_identifier()?,
});
} else {
break;
}
Some(Keyword::LINES)
if self.parse_keywords(&[Keyword::TERMINATED, Keyword::BY]) =>
{
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::LinesTerminatedBy,
char: self.parse_identifier()?,
});
}
Some(Keyword::NULL) => {
if self.parse_keywords(&[Keyword::DEFINED, Keyword::AS]) {
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::NullDefinedAs,
char: self.parse_identifier()?,
});
} else {
break;
}
Some(Keyword::NULL)
if self.parse_keywords(&[Keyword::DEFINED, Keyword::AS]) =>
{
row_delimiters.push(HiveRowDelimiter {
delimiter: HiveDelimiter::NullDefinedAs,
char: self.parse_identifier()?,
});
}
_ => {
break;
Expand Down Expand Up @@ -18083,7 +18072,32 @@ impl<'a> Parser<'a> {
None
};

let options = self.parse_order_by_options()?;
let using_operator = if !with_operator_class
&& dialect_of!(self is PostgreSqlDialect)
&& self.parse_keyword(Keyword::USING)
{
Some(self.parse_order_by_using_operator()?)
} else {
None
};

let options = if using_operator.is_some() {
if self
.peek_one_of_keywords(&[Keyword::ASC, Keyword::DESC])
.is_some()
{
return parser_err!(
"ASC/DESC cannot be used together with USING in ORDER BY".to_string(),
self.peek_token_ref().span.start
);
}
OrderByOptions {
asc: None,
nulls_first: self.parse_order_by_nulls_first_last(),
}
} else {
self.parse_order_by_options()?
};

let with_fill = if self.dialect.supports_with_fill()
&& self.parse_keywords(&[Keyword::WITH, Keyword::FILL])
Expand All @@ -18096,23 +18110,76 @@ impl<'a> Parser<'a> {
Ok((
OrderByExpr {
expr,
using_operator,
options,
with_fill,
},
operator_class,
))
}

fn parse_order_by_options(&mut self) -> Result<OrderByOptions, ParserError> {
let asc = self.parse_asc_desc();
fn parse_order_by_using_operator(&mut self) -> Result<ObjectName, ParserError> {
if self.parse_keyword(Keyword::OPERATOR) {
self.expect_token(&Token::LParen)?;
let operator_name = self.parse_operator_name()?;
let Some(last_part) = operator_name.0.last() else {
return self.expected_ref("an operator name", self.peek_token_ref());
};
let operator = last_part.to_string();
if !Self::is_valid_order_by_using_operator_symbol(&operator) {
return self.expected_ref("an operator name", self.peek_token_ref());
}
self.expect_token(&Token::RParen)?;
return Ok(operator_name);
}

let nulls_first = if self.parse_keywords(&[Keyword::NULLS, Keyword::FIRST]) {
let token = self.next_token();
let operator = token.token.to_string();
if Self::is_valid_order_by_using_operator_symbol(&operator) {
Ok(ObjectName::from(vec![Ident::new(operator)]))
} else {
self.expected_ref("an ordering operator after USING", &token)
}
}

fn is_valid_order_by_using_operator_symbol(symbol: &str) -> bool {
!symbol.is_empty()
&& symbol.chars().all(|c| {
matches!(
c,
'+' | '-'
| '*'
| '/'
| '<'
| '>'
| '='
| '~'
| '!'
| '@'
| '#'
| '%'
| '^'
| '&'
| '|'
| '`'
| '?'
)
})
}

fn parse_order_by_nulls_first_last(&mut self) -> Option<bool> {
if self.parse_keywords(&[Keyword::NULLS, Keyword::FIRST]) {
Some(true)
} else if self.parse_keywords(&[Keyword::NULLS, Keyword::LAST]) {
Some(false)
} else {
None
};
}
}

fn parse_order_by_options(&mut self) -> Result<OrderByOptions, ParserError> {
let asc = self.parse_asc_desc();
let nulls_first = self.parse_order_by_nulls_first_last();

Ok(OrderByOptions { asc, nulls_first })
}
Expand Down Expand Up @@ -20309,6 +20376,7 @@ mod tests {
asc: None,
nulls_first: None,
},
using_operator: None,
with_fill: None,
},
operator_class: None,
Expand Down
2 changes: 2 additions & 0 deletions tests/sqlparser_bigquery.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2721,6 +2721,7 @@ fn test_export_data() {
asc: None,
nulls_first: None,
},
using_operator: None,
with_fill: None,
},]),
interpolate: None,
Expand Down Expand Up @@ -2827,6 +2828,7 @@ fn test_export_data() {
asc: None,
nulls_first: None,
},
using_operator: None,
with_fill: None,
},]),
interpolate: None,
Expand Down
Loading