datafusion_functions/string/
replace.rsuse std::any::Any;
use std::sync::Arc;
use arrow::array::{ArrayRef, GenericStringArray, OffsetSizeTrait, StringArray};
use arrow::datatypes::DataType;
use crate::utils::{make_scalar_function, utf8_to_str_type};
use datafusion_common::cast::{as_generic_string_array, as_string_view_array};
use datafusion_common::{exec_err, Result};
use datafusion_expr::{ColumnarValue, Documentation, Volatility};
use datafusion_expr::{ScalarUDFImpl, Signature};
use datafusion_macros::user_doc;
#[user_doc(
doc_section(label = "String Functions"),
description = "Replaces all occurrences of a specified substring in a string with a new substring.",
syntax_example = "replace(str, substr, replacement)",
sql_example = r#"```sql
> select replace('ABabbaBA', 'ab', 'cd');
+-------------------------------------------------+
| replace(Utf8("ABabbaBA"),Utf8("ab"),Utf8("cd")) |
+-------------------------------------------------+
| ABcdbaBA |
+-------------------------------------------------+
```"#,
standard_argument(name = "str", prefix = "String"),
standard_argument(
name = "substr",
prefix = "Substring expression to replace in the input string. Substring"
),
standard_argument(name = "replacement", prefix = "Replacement substring")
)]
#[derive(Debug)]
pub struct ReplaceFunc {
signature: Signature,
}
impl Default for ReplaceFunc {
fn default() -> Self {
Self::new()
}
}
impl ReplaceFunc {
pub fn new() -> Self {
Self {
signature: Signature::string(3, Volatility::Immutable),
}
}
}
impl ScalarUDFImpl for ReplaceFunc {
fn as_any(&self) -> &dyn Any {
self
}
fn name(&self) -> &str {
"replace"
}
fn signature(&self) -> &Signature {
&self.signature
}
fn return_type(&self, arg_types: &[DataType]) -> Result<DataType> {
utf8_to_str_type(&arg_types[0], "replace")
}
fn invoke_batch(
&self,
args: &[ColumnarValue],
_number_rows: usize,
) -> Result<ColumnarValue> {
match args[0].data_type() {
DataType::Utf8 => make_scalar_function(replace::<i32>, vec![])(args),
DataType::LargeUtf8 => make_scalar_function(replace::<i64>, vec![])(args),
DataType::Utf8View => make_scalar_function(replace_view, vec![])(args),
other => {
exec_err!("Unsupported data type {other:?} for function replace")
}
}
}
fn documentation(&self) -> Option<&Documentation> {
self.doc()
}
}
fn replace_view(args: &[ArrayRef]) -> Result<ArrayRef> {
let string_array = as_string_view_array(&args[0])?;
let from_array = as_string_view_array(&args[1])?;
let to_array = as_string_view_array(&args[2])?;
let result = string_array
.iter()
.zip(from_array.iter())
.zip(to_array.iter())
.map(|((string, from), to)| match (string, from, to) {
(Some(string), Some(from), Some(to)) => Some(string.replace(from, to)),
_ => None,
})
.collect::<StringArray>();
Ok(Arc::new(result) as ArrayRef)
}
fn replace<T: OffsetSizeTrait>(args: &[ArrayRef]) -> Result<ArrayRef> {
let string_array = as_generic_string_array::<T>(&args[0])?;
let from_array = as_generic_string_array::<T>(&args[1])?;
let to_array = as_generic_string_array::<T>(&args[2])?;
let result = string_array
.iter()
.zip(from_array.iter())
.zip(to_array.iter())
.map(|((string, from), to)| match (string, from, to) {
(Some(string), Some(from), Some(to)) => Some(string.replace(from, to)),
_ => None,
})
.collect::<GenericStringArray<T>>();
Ok(Arc::new(result) as ArrayRef)
}
#[cfg(test)]
mod tests {
use super::*;
use crate::utils::test::test_function;
use arrow::array::Array;
use arrow::array::LargeStringArray;
use arrow::array::StringArray;
use arrow::datatypes::DataType::{LargeUtf8, Utf8};
use datafusion_common::ScalarValue;
#[test]
fn test_functions() -> Result<()> {
test_function!(
ReplaceFunc::new(),
vec![
ColumnarValue::Scalar(ScalarValue::Utf8(Some(String::from("aabbdqcbb")))),
ColumnarValue::Scalar(ScalarValue::Utf8(Some(String::from("bb")))),
ColumnarValue::Scalar(ScalarValue::Utf8(Some(String::from("ccc")))),
],
Ok(Some("aacccdqcccc")),
&str,
Utf8,
StringArray
);
test_function!(
ReplaceFunc::new(),
vec![
ColumnarValue::Scalar(ScalarValue::LargeUtf8(Some(String::from(
"aabbb"
)))),
ColumnarValue::Scalar(ScalarValue::LargeUtf8(Some(String::from("bbb")))),
ColumnarValue::Scalar(ScalarValue::LargeUtf8(Some(String::from("cc")))),
],
Ok(Some("aacc")),
&str,
LargeUtf8,
LargeStringArray
);
test_function!(
ReplaceFunc::new(),
vec![
ColumnarValue::Scalar(ScalarValue::Utf8View(Some(String::from(
"aabbbcw"
)))),
ColumnarValue::Scalar(ScalarValue::Utf8View(Some(String::from("bb")))),
ColumnarValue::Scalar(ScalarValue::Utf8View(Some(String::from("cc")))),
],
Ok(Some("aaccbcw")),
&str,
Utf8,
StringArray
);
Ok(())
}
}