rafafrdz commented on code in PR #17485:
URL: https://github.com/apache/datafusion/pull/17485#discussion_r2340230080
##########
datafusion/spark/src/function/url/parse_url.rs:
##########
@@ -47,23 +46,7 @@ impl Default for ParseUrl {
impl ParseUrl {
pub fn new() -> Self {
Self {
- signature: Signature::one_of(
- vec![
- TypeSignature::Uniform(
- 1,
- vec![DataType::Utf8View, DataType::Utf8,
DataType::LargeUtf8],
- ),
- TypeSignature::Uniform(
- 2,
- vec![DataType::Utf8View, DataType::Utf8,
DataType::LargeUtf8],
- ),
- TypeSignature::Uniform(
- 3,
- vec![DataType::Utf8View, DataType::Utf8,
DataType::LargeUtf8],
- ),
- ],
- Volatility::Immutable,
- ),
+ signature: Signature::user_defined(Volatility::Immutable),
Review Comment:
Sorry, I don’t quite follow… Why should dictionary types be used as
arguments for parse_url or try_parse_url? As far as I understand, the spec only
expects string types [parse_url doc
ref](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.parse_url.html)
##########
datafusion/spark/src/function/url/parse_url.rs:
##########
@@ -152,22 +135,39 @@ impl ScalarUDFImpl for ParseUrl {
arg_types.len()
);
}
+ // The return type should match the largest size datatype
match arg_types.len() {
- 2 | 3 => {
+ 2 | 3 if arg_types.iter().all(is_string_type) => {
if arg_types
.iter()
.any(|arg| matches!(arg, DataType::LargeUtf8))
{
Ok(DataType::LargeUtf8)
- } else if arg_types
- .iter()
- .any(|arg| matches!(arg, DataType::Utf8View))
- {
- Ok(DataType::Utf8View)
} else {
Ok(DataType::Utf8)
}
}
+ 2 | 3 => plan_err!(
+ "`{}` expects STRING arguments, got {:?}",
+ &self.name(),
+ arg_types
+ ),
+ _ => plan_err!(
+ "`{}` expects 2 or 3 arguments, got {}",
+ &self.name(),
+ arg_types.len()
+ ),
Review Comment:
done! 😄
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
