This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion-sqlparser-rs.git
The following commit(s) were added to refs/heads/main by this push:
new e380494e Only set `hive_formats` on `CreateTable` if formats are
present (#2105)
e380494e is described below
commit e380494eb0440a31d83542a99d9e9df8de3923ca
Author: Michael Victor Zink <[email protected]>
AuthorDate: Wed Nov 26 01:52:23 2025 -0800
Only set `hive_formats` on `CreateTable` if formats are present (#2105)
---
src/dialect/snowflake.rs | 10 +++++-----
src/parser/mod.rs | 45 ++++++++++++++++++++++++++++-----------------
tests/sqlparser_common.rs | 11 +++++++++++
tests/sqlparser_duckdb.rs | 7 +------
tests/sqlparser_mssql.rs | 14 ++------------
tests/sqlparser_postgres.rs | 7 +------
6 files changed, 48 insertions(+), 46 deletions(-)
diff --git a/src/dialect/snowflake.rs b/src/dialect/snowflake.rs
index bb0d4f16..4cfaddce 100644
--- a/src/dialect/snowflake.rs
+++ b/src/dialect/snowflake.rs
@@ -393,7 +393,7 @@ impl Dialect for SnowflakeDialect {
fn is_column_alias(&self, kw: &Keyword, parser: &mut Parser) -> bool {
match kw {
- // The following keywords can be considered an alias as long as
+ // The following keywords can be considered an alias as long as
// they are not followed by other tokens that may change their
meaning
// e.g. `SELECT * EXCEPT (col1) FROM tbl`
Keyword::EXCEPT
@@ -408,7 +408,7 @@ impl Dialect for SnowflakeDialect {
Keyword::LIMIT | Keyword::OFFSET if peek_for_limit_options(parser)
=> false,
// `FETCH` can be considered an alias as long as it's not followed
by `FIRST`` or `NEXT`
- // which would give it a different meanings, for example:
+ // which would give it a different meanings, for example:
// `SELECT 1 FETCH FIRST 10 ROWS` - not an alias
// `SELECT 1 FETCH 10` - not an alias
Keyword::FETCH if parser.peek_one_of_keywords(&[Keyword::FIRST,
Keyword::NEXT]).is_some()
@@ -417,8 +417,8 @@ impl Dialect for SnowflakeDialect {
false
}
- // Reserved keywords by the Snowflake dialect, which seem to be
less strictive
- // than what is listed in `keywords::RESERVED_FOR_COLUMN_ALIAS`.
The following
+ // Reserved keywords by the Snowflake dialect, which seem to be
less strictive
+ // than what is listed in `keywords::RESERVED_FOR_COLUMN_ALIAS`.
The following
// keywords were tested with the this statement: `SELECT 1 <KW>`.
Keyword::FROM
| Keyword::GROUP
@@ -688,7 +688,7 @@ pub fn parse_create_table(
.iceberg(iceberg)
.global(global)
.dynamic(dynamic)
- .hive_formats(Some(Default::default()));
+ .hive_formats(None);
// Snowflake does not enforce order of the parameters in the statement.
The parser needs to
// parse the statement in a loop.
diff --git a/src/parser/mod.rs b/src/parser/mod.rs
index 759e2eef..759f5189 100644
--- a/src/parser/mod.rs
+++ b/src/parser/mod.rs
@@ -5831,15 +5831,19 @@ impl<'a> Parser<'a> {
let hive_distribution = self.parse_hive_distribution()?;
let hive_formats = self.parse_hive_formats()?;
- let file_format = if let Some(ff) = &hive_formats.storage {
- match ff {
- HiveIOFormat::FileFormat { format } => Some(*format),
- _ => None,
+ let file_format = if let Some(ref hf) = hive_formats {
+ if let Some(ref ff) = hf.storage {
+ match ff {
+ HiveIOFormat::FileFormat { format } => Some(*format),
+ _ => None,
+ }
+ } else {
+ None
}
} else {
None
};
- let location = hive_formats.location.clone();
+ let location = hive_formats.as_ref().and_then(|hf|
hf.location.clone());
let table_properties = self.parse_options(Keyword::TBLPROPERTIES)?;
let table_options = if !table_properties.is_empty() {
CreateTableOptions::TableProperties(table_properties)
@@ -5850,7 +5854,7 @@ impl<'a> Parser<'a> {
.columns(columns)
.constraints(constraints)
.hive_distribution(hive_distribution)
- .hive_formats(Some(hive_formats))
+ .hive_formats(hive_formats)
.table_options(table_options)
.or_replace(or_replace)
.if_not_exists(if_not_exists)
@@ -7579,8 +7583,8 @@ impl<'a> Parser<'a> {
}
}
- pub fn parse_hive_formats(&mut self) -> Result<HiveFormat, ParserError> {
- let mut hive_format = HiveFormat::default();
+ pub fn parse_hive_formats(&mut self) -> Result<Option<HiveFormat>,
ParserError> {
+ let mut hive_format: Option<HiveFormat> = None;
loop {
match self.parse_one_of_keywords(&[
Keyword::ROW,
@@ -7589,7 +7593,9 @@ impl<'a> Parser<'a> {
Keyword::WITH,
]) {
Some(Keyword::ROW) => {
- hive_format.row_format = Some(self.parse_row_format()?);
+ hive_format
+ .get_or_insert_with(HiveFormat::default)
+ .row_format = Some(self.parse_row_format()?);
}
Some(Keyword::STORED) => {
self.expect_keyword_is(Keyword::AS)?;
@@ -7597,24 +7603,29 @@ impl<'a> Parser<'a> {
let input_format = self.parse_expr()?;
self.expect_keyword_is(Keyword::OUTPUTFORMAT)?;
let output_format = self.parse_expr()?;
- hive_format.storage = Some(HiveIOFormat::IOF {
- input_format,
- output_format,
- });
+
hive_format.get_or_insert_with(HiveFormat::default).storage =
+ Some(HiveIOFormat::IOF {
+ input_format,
+ output_format,
+ });
} else {
let format = self.parse_file_format()?;
- hive_format.storage = Some(HiveIOFormat::FileFormat {
format });
+
hive_format.get_or_insert_with(HiveFormat::default).storage =
+ Some(HiveIOFormat::FileFormat { format });
}
}
Some(Keyword::LOCATION) => {
- hive_format.location = Some(self.parse_literal_string()?);
+
hive_format.get_or_insert_with(HiveFormat::default).location =
+ Some(self.parse_literal_string()?);
}
Some(Keyword::WITH) => {
self.prev_token();
let properties = self
.parse_options_with_keywords(&[Keyword::WITH,
Keyword::SERDEPROPERTIES])?;
if !properties.is_empty() {
- hive_format.serde_properties = Some(properties);
+ hive_format
+ .get_or_insert_with(HiveFormat::default)
+ .serde_properties = Some(properties);
} else {
break;
}
@@ -7829,7 +7840,7 @@ impl<'a> Parser<'a> {
.if_not_exists(if_not_exists)
.transient(transient)
.hive_distribution(hive_distribution)
- .hive_formats(Some(hive_formats))
+ .hive_formats(hive_formats)
.global(global)
.query(query)
.without_rowid(without_rowid)
diff --git a/tests/sqlparser_common.rs b/tests/sqlparser_common.rs
index 3649e8d3..91952b8c 100644
--- a/tests/sqlparser_common.rs
+++ b/tests/sqlparser_common.rs
@@ -4724,6 +4724,17 @@ fn parse_create_external_table_lowercase() {
assert_matches!(ast, Statement::CreateTable(CreateTable { .. }));
}
+#[test]
+fn parse_create_table_hive_formats_none_when_no_options() {
+ let sql = "CREATE TABLE simple_table (id INT, name VARCHAR(100))";
+ match verified_stmt(sql) {
+ Statement::CreateTable(CreateTable { hive_formats, .. }) => {
+ assert_eq!(hive_formats, None);
+ }
+ _ => unreachable!(),
+ }
+}
+
#[test]
fn parse_alter_table() {
let add_column = "ALTER TABLE tab ADD COLUMN foo TEXT;";
diff --git a/tests/sqlparser_duckdb.rs b/tests/sqlparser_duckdb.rs
index 0f805195..73a1afe2 100644
--- a/tests/sqlparser_duckdb.rs
+++ b/tests/sqlparser_duckdb.rs
@@ -739,12 +739,7 @@ fn test_duckdb_union_datatype() {
],
constraints: Default::default(),
hive_distribution: HiveDistributionStyle::NONE,
- hive_formats: Some(HiveFormat {
- row_format: Default::default(),
- serde_properties: Default::default(),
- storage: Default::default(),
- location: Default::default()
- }),
+ hive_formats: None,
file_format: Default::default(),
location: Default::default(),
query: Default::default(),
diff --git a/tests/sqlparser_mssql.rs b/tests/sqlparser_mssql.rs
index 99a298f8..37e8e962 100644
--- a/tests/sqlparser_mssql.rs
+++ b/tests/sqlparser_mssql.rs
@@ -1881,12 +1881,7 @@ fn parse_create_table_with_valid_options() {
],
constraints: vec![],
hive_distribution: HiveDistributionStyle::NONE,
- hive_formats: Some(HiveFormat {
- row_format: None,
- serde_properties: None,
- storage: None,
- location: None,
- },),
+ hive_formats: None,
file_format: None,
location: None,
query: None,
@@ -2053,12 +2048,7 @@ fn parse_create_table_with_identity_column() {
},],
constraints: vec![],
hive_distribution: HiveDistributionStyle::NONE,
- hive_formats: Some(HiveFormat {
- row_format: None,
- serde_properties: None,
- storage: None,
- location: None,
- },),
+ hive_formats: None,
file_format: None,
location: None,
query: None,
diff --git a/tests/sqlparser_postgres.rs b/tests/sqlparser_postgres.rs
index 858e2e80..7309a6ba 100644
--- a/tests/sqlparser_postgres.rs
+++ b/tests/sqlparser_postgres.rs
@@ -6016,12 +6016,7 @@ fn parse_trigger_related_functions() {
],
constraints: vec![],
hive_distribution: HiveDistributionStyle::NONE,
- hive_formats: Some(HiveFormat {
- row_format: None,
- serde_properties: None,
- storage: None,
- location: None
- }),
+ hive_formats: None,
file_format: None,
location: None,
query: None,
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
