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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
use std::collections::HashMap;
use polars::prelude::*;
use crate::{ColumnNameBuilder, FromDataFrameRow, IterFromColumn};
pub trait DataframeRowsIterExt<'a> {
fn rows_iter<T>(&'a self) -> PolarsResult<Box<dyn Iterator<Item = PolarsResult<T>> + 'a>>
where
T: FromDataFrameRow<'a>;
fn rows_iter_with_columns<T>(
&'a self,
build_fn: impl FnOnce(&mut T::Builder) -> &mut T::Builder,
) -> PolarsResult<Box<dyn Iterator<Item = PolarsResult<T>> + 'a>>
where
T: FromDataFrameRow<'a>;
fn scalar_iter<T>(&'a self, column_name: &'a str) -> PolarsResult<impl Iterator<Item = PolarsResult<T>> + 'a>
where
T: IterFromColumn<'a> + 'a;
}
impl<'a> DataframeRowsIterExt<'a> for DataFrame {
/// Creates a row iterator for this DataFrame with static column names defined in row struct
/// ```rust
/// use polars::prelude::*;
/// use polars_rows_iter::*;
///
/// #[derive(Debug, FromDataFrameRow)]
/// #[derive(PartialEq)] // for assert_eq
/// struct MyRow<'a>
/// {
/// #[column("col_a")]
/// a: i32,
/// // the column name defaults to the field name if no explicit name given
/// col_b: &'a str,
/// col_c: String,
/// #[column("col_d")]
/// optional: Option<f64>
/// }
///
/// let df = df!(
/// "col_a" => [1i32, 2, 3, 4, 5],
/// "col_b" => ["a", "b", "c", "d", "e"],
/// "col_c" => ["A", "B", "C", "D", "E"],
/// "col_d" => [Some(1.0f64), None, None, Some(2.0), Some(3.0)]
/// ).unwrap();
///
/// let rows_iter = df.rows_iter::<MyRow>().unwrap(); // ready to use row iterator
/// // collect to vector for assert_eq
/// let rows_vec = rows_iter.collect::<PolarsResult<Vec<MyRow>>>().unwrap();
///
/// assert_eq!(
/// rows_vec,
/// [
/// MyRow { a: 1, col_b: "a", col_c: "A".to_string(), optional: Some(1.0) },
/// MyRow { a: 2, col_b: "b", col_c: "B".to_string(), optional: None },
/// MyRow { a: 3, col_b: "c", col_c: "C".to_string(), optional: None },
/// MyRow { a: 4, col_b: "d", col_c: "D".to_string(), optional: Some(2.0) },
/// MyRow { a: 5, col_b: "e", col_c: "E".to_string(), optional: Some(3.0) },
/// ]
/// );
/// ```
fn rows_iter<T>(&'a self) -> PolarsResult<Box<dyn Iterator<Item = PolarsResult<T>> + 'a>>
where
T: FromDataFrameRow<'a>,
{
T::from_dataframe(self, HashMap::new())
}
/// Creates a row iterator for this DataFrame with custom column names, which can be defined over the lambda function
/// for every struct field. If no custom column name for a field is given, the column name falls back to
/// the statically defined one.
///```rust
///use polars::prelude::*;
///use polars_rows_iter::*;
///
///const ID: &str = "id";
///
///#[derive(Debug, FromDataFrameRow)]
///#[derive(PartialEq)] // for assert_eq
///struct MyRow<'a> {
/// #[column(ID)]
/// id: i32,
/// value_b: &'a str,
/// value_c: String,
/// optional: Option<f64>,
///}
///
/// let df = df!(
/// "id" => [1i32, 2, 3, 4, 5],
/// "col_b" => ["a", "b", "c", "d", "e"],
/// "col_c" => ["A", "B", "C", "D", "E"],
/// "col_d" => [Some(1.0f64), None, None, Some(2.0), Some(3.0)]
/// ).unwrap();
///
/// let value_b_column_name = "col_b".to_string();
/// let value_c_column_name = "col_c";
///
/// let rows_iter = df.rows_iter_with_columns::<MyRow>(|columns| {
/// columns
/// .value_b(&value_b_column_name)
/// .value_c(value_c_column_name)
/// .optional("col_d")
/// }).unwrap();
///
/// // collect to vector for assert_eq
/// let rows_vec = rows_iter.collect::<PolarsResult<Vec<MyRow>>>().unwrap();
///
/// assert_eq!(
/// rows_vec,
/// [
/// MyRow { id: 1, value_b: "a", value_c: "A".to_string(), optional: Some(1.0) },
/// MyRow { id: 2, value_b: "b", value_c: "B".to_string(), optional: None },
/// MyRow { id: 3, value_b: "c", value_c: "C".to_string(), optional: None },
/// MyRow { id: 4, value_b: "d", value_c: "D".to_string(), optional: Some(2.0) },
/// MyRow { id: 5, value_b: "e", value_c: "E".to_string(), optional: Some(3.0) },
/// ]
/// );
///```
fn rows_iter_with_columns<T>(
&'a self,
build_fn: impl FnOnce(&mut T::Builder) -> &mut T::Builder,
) -> PolarsResult<Box<dyn Iterator<Item = PolarsResult<T>> + 'a>>
where
T: FromDataFrameRow<'a>,
{
let mut builder = T::create_builder();
build_fn(&mut builder);
let columns = builder.build();
T::from_dataframe(self, columns)
}
/// Creates an iterator for a single column in the DataFrame
///
/// This is a simpler alternative to `rows_iter` when you only need to iterate over one column.
/// The type parameter `T` specifies the Rust type to convert column values to.
///
/// ```rust
/// use polars::prelude::*;
/// use polars_rows_iter::*;
///
/// let df = df!(
/// "col_a" => [1i32, 2, 3, 4, 5],
/// "col_b" => ["a", "b", "c", "d", "e"],
/// "col_c" => [Some("A"), Some("B"), None, None, Some("E")],
/// ).unwrap();
///
/// // Iterate over a column with non-nullable values
/// let values_a = df.scalar_iter::<i32>("col_a")
/// .unwrap()
/// .collect::<PolarsResult<Vec<i32>>>()
/// .unwrap();
/// assert_eq!(values_a, [1, 2, 3, 4, 5]);
///
/// // Iterate over a column with borrowed string values
/// let values_b = df.scalar_iter::<&str>("col_b")
/// .unwrap()
/// .collect::<PolarsResult<Vec<&str>>>()
/// .unwrap();
/// assert_eq!(values_b, ["a", "b", "c", "d", "e"]);
///
/// // Iterate over a column with optional values
/// let values_c = df.scalar_iter::<Option<String>>("col_c")
/// .unwrap()
/// .collect::<PolarsResult<Vec<Option<String>>>>()
/// .unwrap();
/// assert_eq!(
/// values_c,
/// [Some("A".to_string()), Some("B".to_string()), None, None, Some("E".to_string())]
/// );
/// ```
fn scalar_iter<T>(&'a self, column_name: &'a str) -> PolarsResult<impl Iterator<Item = PolarsResult<T>> + 'a>
where
T: IterFromColumn<'a> + 'a,
{
let column = self.column(column_name)?;
let column_dtype = column.dtype();
let iter = <T as IterFromColumn<'a>>::create_iter(column)?;
let iter = iter.map(|v| <T as IterFromColumn<'a>>::get_value(v, column_name, column_dtype));
Ok(iter)
}
}
#[cfg(test)]
mod tests {
#![allow(dead_code)]
use polars::df;
use crate::*;
#[derive(FromDataFrameRow)]
struct TestStruct {
x1: i32,
x2: i32,
}
#[test]
fn rows_iter_should_return_error_when_given_column_not_available() {
let df = df!(
"y1" => [1i32, 2, 3],
"x2" => [1i32, 2, 3]
)
.unwrap();
let result = df.rows_iter::<TestStruct>();
assert!(result.is_err());
}
#[test]
fn builder_should_build_hashmap_with_correct_entries() {
let mut builder = TestStruct::create_builder();
builder.x1("column_1").x2("column_2");
let columns = builder.build();
assert_eq!("column_1", *columns.get("x1").unwrap());
assert_eq!("column_2", *columns.get("x2").unwrap());
}
#[test]
fn rows_iter_with_columns_should_return_error_when_given_column_not_available() {
let df = df!(
"x1" => [1i32, 2, 3],
"x2" => [1i32, 2, 3]
)
.unwrap();
let result = df.rows_iter_with_columns::<TestStruct>(|b| b.x1("y1"));
assert!(result.is_err());
}
#[test]
fn rows_iter_with_columns_should_return_valid_iter() {
let df = df!(
"x_1" => [1i32, 2, 3],
"x_2" => [1i32, 2, 3]
)
.unwrap();
let result = df.rows_iter_with_columns::<TestStruct>(|b| b.x1("x_1").x2("x_2"));
assert!(result.is_ok());
}
}