diff --git a/native/spark-expr/src/comet_scalar_funcs.rs b/native/spark-expr/src/comet_scalar_funcs.rs index 9ecb11dc52..bddd439666 100644 --- a/native/spark-expr/src/comet_scalar_funcs.rs +++ b/native/spark-expr/src/comet_scalar_funcs.rs @@ -16,6 +16,7 @@ // under the License. use crate::hash_funcs::*; +use crate::json_funcs::JsonArrayLength; use crate::map_funcs::spark_map_sort; use crate::math_funcs::abs::abs; use crate::math_funcs::checked_arithmetic::{checked_add, checked_div, checked_mul, checked_sub}; @@ -216,6 +217,7 @@ fn all_scalar_functions() -> Vec> { Arc::new(ScalarUDF::new_from_impl(SparkMakeDate::default())), Arc::new(ScalarUDF::new_from_impl(SparkSecondsToTimestamp::default())), Arc::new(ScalarUDF::new_from_impl(SparkSizeFunc::default())), + Arc::new(ScalarUDF::new_from_impl(JsonArrayLength::default())), ] } diff --git a/native/spark-expr/src/json_funcs/json_array_length.rs b/native/spark-expr/src/json_funcs/json_array_length.rs new file mode 100644 index 0000000000..289918363d --- /dev/null +++ b/native/spark-expr/src/json_funcs/json_array_length.rs @@ -0,0 +1,137 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +use arrow::array::{Array, ArrayRef, Int32Builder}; +use arrow::datatypes::DataType; +use datafusion::common::cast::as_string_array; +use datafusion::common::{exec_err, Result, ScalarValue}; +use datafusion::logical_expr::{ + ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility, +}; + +use std::any::Any; +use std::sync::Arc; + +#[derive(Debug, PartialEq, Eq, Hash)] +pub struct JsonArrayLength { + signature: Signature, +} + +impl Default for JsonArrayLength { + fn default() -> Self { + Self::new() + } +} + +impl JsonArrayLength { + pub fn new() -> Self { + Self { + signature: Signature::variadic(vec![DataType::Utf8], Volatility::Immutable), + } + } +} + +impl ScalarUDFImpl for JsonArrayLength { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "json_array_length" + } + + fn signature(&self) -> &Signature { + &self.signature + } + + fn return_type(&self, _arg_types: &[DataType]) -> Result { + Ok(DataType::Int32) + } + + fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result { + spark_json_array_length(&args.args) + } +} + +fn spark_json_array_length(args: &[ColumnarValue]) -> Result { + if args.len() != 1 { + return exec_err!("json_array_length function takes exactly one argument"); + } + match &args[0] { + ColumnarValue::Array(array) => { + let result = spark_json_array_length_array(array)?; + Ok(ColumnarValue::Array(result)) + } + ColumnarValue::Scalar(scalar) => { + let result = spark_json_array_length_scalar(scalar)?; + Ok(ColumnarValue::Scalar(result)) + } + } +} + +fn spark_json_array_length_array(array: &ArrayRef) -> Result { + match array.data_type() { + DataType::Utf8 => { + let array = as_string_array(array)?; + let mut builder = Int32Builder::with_capacity(array.len()); + + for row_idx in 0..array.len() { + if array.is_null(row_idx) { + builder.append_null(); + } else { + let json_str = array.value(row_idx); + if let Some(json_array_length) = get_json_array_length(json_str) { + builder.append_value(json_array_length); + } else { + builder.append_null() + } + } + } + Ok(Arc::new(builder.finish())) + } + other => { + exec_err!("Unsupported data type {other:?} for function `json_array_length`") + } + } +} + +fn spark_json_array_length_scalar(scalar: &ScalarValue) -> Result { + match scalar { + ScalarValue::Utf8(value) => { + let length = value + .clone() + .and_then(|json_str| get_json_array_length(&json_str)); + Ok(ScalarValue::Int32(length)) + } + other => { + exec_err!("Unsupported data type {other:?} for function `json_array_length`") + } + } +} + +fn get_json_array_length(json_str: &str) -> Option { + match serde_json::from_str::(json_str) { + Ok(json_value) => { + if json_value.is_array() { + Some(json_value.as_array().unwrap().len() as i32) + } else { + None + } + } + Err(_) => None, + } +} diff --git a/native/spark-expr/src/json_funcs/mod.rs b/native/spark-expr/src/json_funcs/mod.rs index 9f025070d7..59e1e5dd58 100644 --- a/native/spark-expr/src/json_funcs/mod.rs +++ b/native/spark-expr/src/json_funcs/mod.rs @@ -16,7 +16,9 @@ // under the License. mod from_json; +mod json_array_length; mod to_json; pub use from_json::FromJson; +pub use json_array_length::JsonArrayLength; pub use to_json::ToJson; diff --git a/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala b/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala index d85a2c30cb..297d3cb02e 100644 --- a/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala +++ b/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala @@ -249,6 +249,9 @@ object QueryPlanSerde extends Logging with CometExprShim with CometTypeShim { private val conversionExpressions: Map[Class[_ <: Expression], CometExpressionSerde[_]] = Map( classOf[Cast] -> CometCast) + private val jsonExpressions: Map[Class[_ <: Expression], CometExpressionSerde[_]] = Map( + classOf[LengthOfJsonArray] -> CometScalarFunction("json_array_length")) + private[comet] val miscExpressions: Map[Class[_ <: Expression], CometExpressionSerde[_]] = Map( // TODO PromotePrecision classOf[Alias] -> CometAlias, @@ -273,7 +276,7 @@ object QueryPlanSerde extends Logging with CometExprShim with CometTypeShim { mathExpressions ++ hashExpressions ++ stringExpressions ++ conditionalExpressions ++ mapExpressions ++ predicateExpressions ++ structExpressions ++ bitwiseExpressions ++ miscExpressions ++ arrayExpressions ++ - temporalExpressions ++ conversionExpressions + temporalExpressions ++ conversionExpressions ++ jsonExpressions /** * Mapping of Spark aggregate expression class to Comet expression handler. diff --git a/spark/src/main/scala/org/apache/comet/serde/statics.scala b/spark/src/main/scala/org/apache/comet/serde/statics.scala index bff64e753a..9aa315764f 100644 --- a/spark/src/main/scala/org/apache/comet/serde/statics.scala +++ b/spark/src/main/scala/org/apache/comet/serde/statics.scala @@ -20,6 +20,7 @@ package org.apache.comet.serde import org.apache.spark.sql.catalyst.expressions.{Attribute, ExpressionImplUtils, Literal, UrlCodec} +import org.apache.spark.sql.catalyst.expressions.json.JsonExpressionUtils import org.apache.spark.sql.catalyst.expressions.objects.StaticInvoke import org.apache.spark.sql.catalyst.util.CharVarcharCodegenUtils @@ -38,7 +39,9 @@ object CometStaticInvoke extends CometExpressionSerde[StaticInvoke] { "read_side_padding"), ("isLuhnNumber", classOf[ExpressionImplUtils]) -> CometScalarFunction("luhn_check"), ("encode", UrlCodec.getClass) -> CometUrlEncodeStaticInvoke, - ("decode", UrlCodec.getClass) -> CometUrlDecodeStaticInvoke) + ("decode", UrlCodec.getClass) -> CometUrlDecodeStaticInvoke, + ("lengthOfJsonArray", classOf[JsonExpressionUtils]) -> CometScalarFunction( + "json_array_length")) override def convert( expr: StaticInvoke, diff --git a/spark/src/test/resources/sql-tests/expressions/json/json_array_length.sql b/spark/src/test/resources/sql-tests/expressions/json/json_array_length.sql new file mode 100644 index 0000000000..5d2e82f2cd --- /dev/null +++ b/spark/src/test/resources/sql-tests/expressions/json/json_array_length.sql @@ -0,0 +1,61 @@ +-- Licensed to the Apache Software Foundation (ASF) under one +-- or more contributor license agreements. See the NOTICE file +-- distributed with this work for additional information +-- regarding copyright ownership. The ASF licenses this file +-- to you under the Apache License, Version 2.0 (the +-- "License"); you may not use this file except in compliance +-- with the License. You may obtain a copy of the License at +-- +-- http://www.apache.org/licenses/LICENSE-2.0 +-- +-- Unless required by applicable law or agreed to in writing, +-- software distributed under the License is distributed on an +-- "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +-- KIND, either express or implied. See the License for the +-- specific language governing permissions and limitations +-- under the License. + +statement +CREATE TABLE test_json_array_length(j string) USING parquet + +statement +INSERT INTO test_json_array_length VALUES + ('[1,2,3,4]'), + ('[]'), + ('[1]'), + (NULL), + ('[1,2,3,{"f1":1,"f2":[5,6]},4]'), + ('[[1,2],[3,4],[5,6]]'), + ('[{"a":1},{"b":2},{"c":3}]'), + ('[1,2'), + ('[1,2,3,]'), + ('not a json'), + ('{"object": "not array"}'), + (''), + (' '), + ('[true, false, null]'), + ('["string1", "string2", "string3"]'), + ('[1, "mixed", true, null, {"key":"value"}]'), + ('[1,2,3,4,5,6,7,8,9,10]'), + ('["line1\nline2", "tab\tseparated", "quote\"here"]'), + ('{"outer": [1,2,3], "inner": [[1,2],[3,4]]}'), + ('{"arrays": {"first": [1,2], "second": [3,4,5]}}'), + ('[{"arr": [1,2,3]}, {"arr": [4,5]}]') + +query +SELECT json_array_length(j) FROM test_json_array_length + +query +SELECT json_array_length('[1,2,3,4]') + +query +SELECT json_array_length('not an array') + +query +SELECT json_array_length('{"key":"value"}') + +query +SELECT json_array_length(NULL) + +query +SELECT json_array_length('[]')