feat: simplify regex wildcard pattern

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>

Author: waynexia

datafusion/optimizer/src/simplify_expressions/regex.rs

@@ -33,6 +33,8 @@ const MAX_REGEX_ALTERNATIONS_EXPANSION: usize = 4;
 /// - full anchored regex patterns (e.g. `^foo$`) to `= 'foo'`
 /// - partial anchored regex patterns (e.g. `^foo`) to `LIKE 'foo%'`
 /// - combinations (alternatives) of the above, will be concatenated with `OR` or `AND`
+/// - `EQ .*` is always true
+/// - `NE .*` means IS EMPTY
 ///
 /// Dev note: unit tests of this function are in `expr_simplifier.rs`, case `test_simplify_regex`.
 pub fn simplify_regex_expr(
@@ -43,6 +45,23 @@ pub fn simplify_regex_expr(
     let mode = OperatorMode::new(&op);
 
     if let Expr::Literal(ScalarValue::Utf8(Some(pattern))) = right.as_ref() {
+        // Handle the special case for ".*" pattern
+        if pattern == ".*" {
+            let new_expr = if mode.not {
+                // not empty
+                let empty_lit = Box::new(lit(""));
+                Expr::BinaryExpr(BinaryExpr {
+                    left,
+                    op: Operator::Eq,
+                    right: empty_lit,
+                })
+            } else {
+                // always true
+                lit(true)
+            };
+            return Ok(new_expr);
+        }
+
         match regex_syntax::Parser::new().parse(pattern) {
             Ok(hir) => {
                 let kind = hir.kind();

datafusion/optimizer/src/simplify_expressions/simplify_exprs.rs

@@ -765,4 +765,48 @@ mod tests {
 
         assert_optimized_plan_eq(plan, expected)
     }
+
+    #[test]
+    fn test_simplify_regex_special_cases() -> Result<()> {
+        let schema = Schema::new(vec![Field::new("a", DataType::Utf8, false)]);
+        let table_scan = table_scan(Some("test"), &schema, None)?.build()?;
+
+        // Test `= ".*"` transforms to true (except for empty strings)
+        let plan = LogicalPlanBuilder::from(table_scan.clone())
+            .filter(binary_expr(col("a"), Operator::RegexMatch, lit(".*")))?
+            .build()?;
+        let expected = "Filter: Boolean(true)\
+        \n  TableScan: test";
+
+        assert_optimized_plan_eq(plan, expected)?;
+
+        // Test `!= ".*"` transforms to checking if the column is empty
+        let plan = LogicalPlanBuilder::from(table_scan.clone())
+            .filter(binary_expr(col("a"), Operator::RegexNotMatch, lit(".*")))?
+            .build()?;
+        let expected = "Filter: test.a = Utf8(\"\")\
+        \n  TableScan: test";
+
+        assert_optimized_plan_eq(plan, expected)?;
+
+        // Test case-insensitive versions
+
+        // Test `=~ ".*"` (case-insensitive) transforms to true (except for empty strings)
+        let plan = LogicalPlanBuilder::from(table_scan.clone())
+            .filter(binary_expr(col("a"), Operator::RegexIMatch, lit(".*")))?
+            .build()?;
+        let expected = "Filter: Boolean(true)\
+        \n  TableScan: test";
+
+        assert_optimized_plan_eq(plan, expected)?;
+
+        // Test `!~ ".*"` (case-insensitive) transforms to checking if the column is empty
+        let plan = LogicalPlanBuilder::from(table_scan.clone())
+            .filter(binary_expr(col("a"), Operator::RegexNotIMatch, lit(".*")))?
+            .build()?;
+        let expected = "Filter: test.a = Utf8(\"\")\
+        \n  TableScan: test";
+
+        assert_optimized_plan_eq(plan, expected)
+    }
 }

datafusion/sqllogictest/test_files/simplify_expr.slt

@@ -15,8 +15,8 @@
 # specific language governing permissions and limitations
 # under the License.
 
-statement count 0
-create table t(a int) as values (1);
+statement ok
+create table t(a int, b string) as values (1, 'a');
 
 # test between simplification
 query TT
@@ -30,5 +30,23 @@ physical_plan
 02)--FilterExec: a@0 = 3
 03)----DataSourceExec: partitions=1, partition_sizes=[1]
 
-statement count 0
+# test regex exprs
+query TT
+explain select b from t where b ~ '.*'
+----
+logical_plan TableScan: t projection=[b]
+physical_plan DataSourceExec: partitions=1, partition_sizes=[1]
+
+query TT
+explain select b from t where b !~ '.*'
+----
+logical_plan
+01)Filter: t.b = Utf8("")
+02)--TableScan: t projection=[b]
+physical_plan
+01)CoalesceBatchesExec: target_batch_size=8192
+02)--FilterExec: b@0 = 
+03)----DataSourceExec: partitions=1, partition_sizes=[1]
+
+statement ok
 drop table t;