1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
use crate::PhysicalExpr;
use arrow::datatypes::{DataType, Schema};
use arrow::record_batch::RecordBatch;
use datafusion_common::Result;
use datafusion_expr::BuiltinScalarFunction;
use datafusion_expr::ColumnarValue;
pub use datafusion_expr::NullColumnarValue;
use datafusion_expr::ScalarFunctionImplementation;
use std::any::Any;
use std::fmt::Debug;
use std::fmt::{self, Formatter};
use std::sync::Arc;
pub struct ScalarFunctionExpr {
fun: ScalarFunctionImplementation,
name: String,
args: Vec<Arc<dyn PhysicalExpr>>,
return_type: DataType,
}
impl Debug for ScalarFunctionExpr {
fn fmt(&self, f: &mut Formatter) -> fmt::Result {
f.debug_struct("ScalarFunctionExpr")
.field("fun", &"<FUNC>")
.field("name", &self.name)
.field("args", &self.args)
.field("return_type", &self.return_type)
.finish()
}
}
impl ScalarFunctionExpr {
pub fn new(
name: &str,
fun: ScalarFunctionImplementation,
args: Vec<Arc<dyn PhysicalExpr>>,
return_type: &DataType,
) -> Self {
Self {
fun,
name: name.to_owned(),
args,
return_type: return_type.clone(),
}
}
pub fn fun(&self) -> &ScalarFunctionImplementation {
&self.fun
}
pub fn name(&self) -> &str {
&self.name
}
pub fn args(&self) -> &[Arc<dyn PhysicalExpr>] {
&self.args
}
pub fn return_type(&self) -> &DataType {
&self.return_type
}
}
impl fmt::Display for ScalarFunctionExpr {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(
f,
"{}({})",
self.name,
self.args
.iter()
.map(|e| format!("{}", e))
.collect::<Vec<String>>()
.join(", ")
)
}
}
impl PhysicalExpr for ScalarFunctionExpr {
fn as_any(&self) -> &dyn Any {
self
}
fn data_type(&self, _input_schema: &Schema) -> Result<DataType> {
Ok(self.return_type.clone())
}
fn nullable(&self, _input_schema: &Schema) -> Result<bool> {
Ok(true)
}
fn evaluate(&self, batch: &RecordBatch) -> Result<ColumnarValue> {
let inputs = match (self.args.len(), self.name.parse::<BuiltinScalarFunction>()) {
(0, Ok(scalar_fun)) if scalar_fun.supports_zero_argument() => {
vec![NullColumnarValue::from(batch)]
}
_ => self
.args
.iter()
.map(|e| e.evaluate(batch))
.collect::<Result<Vec<_>>>()?,
};
let fun = self.fun.as_ref();
(fun)(&inputs)
}
}