This is an automated email from the ASF dual-hosted git repository.
alamb pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-datafusion.git
The following commit(s) were added to refs/heads/master by this push:
new 5c5a0bb Fix SQL planner to support multibyte column names (#357)
5c5a0bb is described below
commit 5c5a0bb0af8130069071c193ca8c214022666a4e
Author: Agata Naomichi <[email protected]>
AuthorDate: Thu May 20 01:44:07 2021 +0900
Fix SQL planner to support multibyte column names (#357)
* Fix SQL planner to support multibyte column names
* Fix test cases for multibyte column name support
* Use starts_with
---
datafusion/src/sql/planner.rs | 25 +++++++++++++++++--------
1 file changed, 17 insertions(+), 8 deletions(-)
diff --git a/datafusion/src/sql/planner.rs b/datafusion/src/sql/planner.rs
index 036c66d..34c5901 100644
--- a/datafusion/src/sql/planner.rs
+++ b/datafusion/src/sql/planner.rs
@@ -857,7 +857,7 @@ impl<'a, S: ContextProvider> SqlToRel<'a, S> {
),
SQLExpr::Identifier(ref id) => {
- if &id.value[0..1] == "@" {
+ if id.value.starts_with('@') {
let var_names = vec![id.value.clone()];
Ok(Expr::ScalarVariable(var_names))
} else {
@@ -1520,7 +1520,7 @@ mod tests {
use functions::ScalarFunctionImplementation;
const PERSON_COLUMN_NAMES: &str =
- "id, first_name, last_name, age, state, salary, birth_date";
+ "id, first_name, last_name, age, state, salary, birth_date, 😀";
#[test]
fn select_no_relation() {
@@ -1559,7 +1559,7 @@ mod tests {
let sql = "SELECT *, age FROM person";
let err = logical_plan(sql).expect_err("query should have failed");
assert_eq!(
- "Plan(\"Projections require unique expression names but the
expression \\\"#age\\\" at position 3 and \\\"#age\\\" at position 7 have the
same name. Consider aliasing (\\\"AS\\\") one of them.\")",
+ "Plan(\"Projections require unique expression names but the
expression \\\"#age\\\" at position 3 and \\\"#age\\\" at position 8 have the
same name. Consider aliasing (\\\"AS\\\") one of them.\")",
format!("{:?}", err)
);
}
@@ -1568,7 +1568,7 @@ mod tests {
fn select_wildcard_with_repeated_column_but_is_aliased() {
quick_test(
"SELECT *, first_name AS fn from person",
- "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date, #first_name AS fn\
+ "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date, #😀, #first_name AS fn\
\n TableScan: person projection=None",
);
}
@@ -2044,9 +2044,9 @@ mod tests {
#[test]
fn select_wildcard_with_groupby() {
quick_test(
- "SELECT * FROM person GROUP BY id, first_name, last_name, age,
state, salary, birth_date",
- "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date\
- \n Aggregate: groupBy=[[#id, #first_name, #last_name, #age,
#state, #salary, #birth_date]], aggr=[[]]\
+ r#"SELECT * FROM person GROUP BY id, first_name, last_name, age,
state, salary, birth_date, "😀""#,
+ "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date, #😀\
+ \n Aggregate: groupBy=[[#id, #first_name, #last_name, #age,
#state, #salary, #birth_date, #😀]], aggr=[[]]\
\n TableScan: person projection=None",
);
quick_test(
@@ -2365,7 +2365,7 @@ mod tests {
fn test_wildcard() {
quick_test(
"SELECT * from person",
- "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date\
+ "Projection: #id, #first_name, #last_name, #age, #state, #salary,
#birth_date, #😀\
\n TableScan: person projection=None",
);
}
@@ -2679,6 +2679,14 @@ mod tests {
quick_test(sql, expected);
}
+ #[test]
+ fn select_multibyte_column() {
+ let sql = r#"SELECT "😀" FROM person"#;
+ let expected = "Projection: #😀\
+ \n TableScan: person projection=None";
+ quick_test(sql, expected);
+ }
+
fn logical_plan(sql: &str) -> Result<LogicalPlan> {
let planner = SqlToRel::new(&MockContextProvider {});
let result = DFParser::parse_sql(&sql);
@@ -2712,6 +2720,7 @@ mod tests {
DataType::Timestamp(TimeUnit::Nanosecond, None),
false,
),
+ Field::new("😀", DataType::Int32, false),
])),
"orders" => Some(Schema::new(vec