Add tests for StringView / character functions, fix regexp_like and…

… `regexp_match` to work with StringView (#11753) * Minor: Add tests for StringView / character functions * Fix regexp_like and regexp_match to work with StringVeiw * Update for ASCII and BTRIM * Add comment about why it is ok to return boolean with catchall match * Fix character_length * Add ticket references
apache · Aug 8, 2024 · 368df80 · 368df80
1 parent 193955e
commit 368df80
Show file tree

Hide file tree

Showing 4 changed files with 427 additions and 105 deletions.
diff --git a/datafusion/functions/src/regex/regexplike.rs b/datafusion/functions/src/regex/regexplike.rs
@@ -75,13 +75,10 @@ impl ScalarUDFImpl for RegexpLikeFunc {
         use DataType::*;
 
         Ok(match &arg_types[0] {
-            LargeUtf8 | Utf8 => Boolean,
             Null => Null,
-            other => {
-                return plan_err!(
-                    "The regexp_like function can only accept strings. Got {other}"
-                );
-            }
+            // Type coercion is done by DataFusion based on signature, so if we
+            // get here, the first argument is always a string
+            _ => Boolean,
         })
     }
     fn invoke(&self, args: &[ColumnarValue]) -> Result<ColumnarValue> {

diff --git a/datafusion/functions/src/regex/regexpmatch.rs b/datafusion/functions/src/regex/regexpmatch.rs
@@ -74,17 +74,9 @@ impl ScalarUDFImpl for RegexpMatchFunc {
     }
 
     fn return_type(&self, arg_types: &[DataType]) -> Result<DataType> {
-        use DataType::*;
-
         Ok(match &arg_types[0] {
-            LargeUtf8 => List(Arc::new(Field::new("item", LargeUtf8, true))),
-            Utf8 => List(Arc::new(Field::new("item", Utf8, true))),
-            Null => Null,
-            other => {
-                return plan_err!(
-                    "The regexp_match function can only accept strings. Got {other}"
-                );
-            }
+            DataType::Null => DataType::Null,
+            other => DataType::List(Arc::new(Field::new("item", other.clone(), true))),
         })
     }
     fn invoke(&self, args: &[ColumnarValue]) -> Result<ColumnarValue> {

diff --git a/datafusion/functions/src/unicode/character_length.rs b/datafusion/functions/src/unicode/character_length.rs
@@ -44,7 +44,7 @@ impl CharacterLengthFunc {
         Self {
             signature: Signature::uniform(
                 1,
-                vec![Utf8, LargeUtf8],
+                vec![Utf8, LargeUtf8, Utf8View],
                 Volatility::Immutable,
             ),
             aliases: vec![String::from("length"), String::from("char_length")],