polars-df 0.1.4 → 0.1.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +5 -0
- data/Cargo.lock +70 -9
- data/Cargo.toml +2 -0
- data/ext/polars/Cargo.toml +6 -1
- data/ext/polars/src/apply/dataframe.rs +292 -0
- data/ext/polars/src/apply/mod.rs +254 -0
- data/ext/polars/src/apply/series.rs +1173 -0
- data/ext/polars/src/conversion.rs +100 -5
- data/ext/polars/src/dataframe.rs +146 -1
- data/ext/polars/src/error.rs +8 -0
- data/ext/polars/src/lazy/apply.rs +34 -2
- data/ext/polars/src/lazy/dataframe.rs +72 -1
- data/ext/polars/src/lazy/dsl.rs +38 -0
- data/ext/polars/src/lib.rs +165 -1
- data/ext/polars/src/series.rs +296 -0
- data/ext/polars/src/utils.rs +25 -0
- data/lib/polars/convert.rb +100 -0
- data/lib/polars/data_frame.rb +1457 -56
- data/lib/polars/dynamic_group_by.rb +49 -0
- data/lib/polars/expr.rb +258 -9
- data/lib/polars/functions.rb +192 -3
- data/lib/polars/group_by.rb +43 -3
- data/lib/polars/io.rb +19 -3
- data/lib/polars/lazy_frame.rb +792 -22
- data/lib/polars/lazy_functions.rb +561 -27
- data/lib/polars/rolling_group_by.rb +35 -0
- data/lib/polars/series.rb +132 -10
- data/lib/polars/utils.rb +16 -0
- data/lib/polars/version.rb +1 -1
- data/lib/polars.rb +9 -1
- metadata +9 -3
@@ -0,0 +1,254 @@
|
|
1
|
+
pub mod dataframe;
|
2
|
+
pub mod series;
|
3
|
+
|
4
|
+
use magnus::{RHash, Value};
|
5
|
+
use polars::chunked_array::builder::get_list_builder;
|
6
|
+
use polars::prelude::*;
|
7
|
+
use polars_core::export::rayon::prelude::*;
|
8
|
+
use polars_core::utils::CustomIterTools;
|
9
|
+
use polars_core::POOL;
|
10
|
+
|
11
|
+
use crate::{ObjectValue, RbPolarsErr, RbResult, RbSeries, Wrap};
|
12
|
+
|
13
|
+
pub trait RbArrowPrimitiveType: PolarsNumericType {}
|
14
|
+
|
15
|
+
impl RbArrowPrimitiveType for UInt8Type {}
|
16
|
+
impl RbArrowPrimitiveType for UInt16Type {}
|
17
|
+
impl RbArrowPrimitiveType for UInt32Type {}
|
18
|
+
impl RbArrowPrimitiveType for UInt64Type {}
|
19
|
+
impl RbArrowPrimitiveType for Int8Type {}
|
20
|
+
impl RbArrowPrimitiveType for Int16Type {}
|
21
|
+
impl RbArrowPrimitiveType for Int32Type {}
|
22
|
+
impl RbArrowPrimitiveType for Int64Type {}
|
23
|
+
impl RbArrowPrimitiveType for Float32Type {}
|
24
|
+
impl RbArrowPrimitiveType for Float64Type {}
|
25
|
+
|
26
|
+
fn iterator_to_struct(
|
27
|
+
it: impl Iterator<Item = Option<Value>>,
|
28
|
+
init_null_count: usize,
|
29
|
+
first_value: AnyValue,
|
30
|
+
name: &str,
|
31
|
+
capacity: usize,
|
32
|
+
) -> RbResult<RbSeries> {
|
33
|
+
let (vals, flds) = match &first_value {
|
34
|
+
AnyValue::Struct(vals, flds) => (&**vals, *flds),
|
35
|
+
AnyValue::StructOwned(payload) => (&*payload.0, &*payload.1),
|
36
|
+
_ => {
|
37
|
+
return Err(crate::error::ComputeError::new_err(format!(
|
38
|
+
"expected struct got {:?}",
|
39
|
+
first_value
|
40
|
+
)))
|
41
|
+
}
|
42
|
+
};
|
43
|
+
|
44
|
+
let struct_width = vals.len();
|
45
|
+
|
46
|
+
// every item in the struct is kept as its own buffer of anyvalues
|
47
|
+
// so as struct with 2 items: {a, b}
|
48
|
+
// will have
|
49
|
+
// [
|
50
|
+
// [ a values ]
|
51
|
+
// [ b values ]
|
52
|
+
// ]
|
53
|
+
let mut items = Vec::with_capacity(vals.len());
|
54
|
+
for item in vals {
|
55
|
+
let mut buf = Vec::with_capacity(capacity);
|
56
|
+
for _ in 0..init_null_count {
|
57
|
+
buf.push(AnyValue::Null);
|
58
|
+
}
|
59
|
+
buf.push(item.clone());
|
60
|
+
items.push(buf);
|
61
|
+
}
|
62
|
+
|
63
|
+
for dict in it {
|
64
|
+
match dict {
|
65
|
+
None => {
|
66
|
+
for field_items in &mut items {
|
67
|
+
field_items.push(AnyValue::Null);
|
68
|
+
}
|
69
|
+
}
|
70
|
+
Some(dict) => {
|
71
|
+
let dict = dict.try_convert::<RHash>()?;
|
72
|
+
if dict.len() != struct_width {
|
73
|
+
return Err(crate::error::ComputeError::new_err(
|
74
|
+
format!("Cannot create struct type.\n> The struct dtype expects {} fields, but it got a dict with {} fields.", struct_width, dict.len())
|
75
|
+
));
|
76
|
+
}
|
77
|
+
// we ignore the keys of the rest of the dicts
|
78
|
+
// the first item determines the output name
|
79
|
+
todo!()
|
80
|
+
// for ((_, val), field_items) in dict.iter().zip(&mut items) {
|
81
|
+
// let item = val.try_convert::<Wrap<AnyValue>>()?;
|
82
|
+
// field_items.push(item.0)
|
83
|
+
// }
|
84
|
+
}
|
85
|
+
}
|
86
|
+
}
|
87
|
+
|
88
|
+
let fields = POOL.install(|| {
|
89
|
+
items
|
90
|
+
.par_iter()
|
91
|
+
.zip(flds)
|
92
|
+
.map(|(av, fld)| Series::new(fld.name(), av))
|
93
|
+
.collect::<Vec<_>>()
|
94
|
+
});
|
95
|
+
|
96
|
+
Ok(StructChunked::new(name, &fields)
|
97
|
+
.unwrap()
|
98
|
+
.into_series()
|
99
|
+
.into())
|
100
|
+
}
|
101
|
+
|
102
|
+
fn iterator_to_primitive<T>(
|
103
|
+
it: impl Iterator<Item = Option<T::Native>>,
|
104
|
+
init_null_count: usize,
|
105
|
+
first_value: Option<T::Native>,
|
106
|
+
name: &str,
|
107
|
+
capacity: usize,
|
108
|
+
) -> ChunkedArray<T>
|
109
|
+
where
|
110
|
+
T: RbArrowPrimitiveType,
|
111
|
+
{
|
112
|
+
// safety: we know the iterators len
|
113
|
+
let mut ca: ChunkedArray<T> = unsafe {
|
114
|
+
if init_null_count > 0 {
|
115
|
+
(0..init_null_count)
|
116
|
+
.map(|_| None)
|
117
|
+
.chain(std::iter::once(first_value))
|
118
|
+
.chain(it)
|
119
|
+
.trust_my_length(capacity)
|
120
|
+
.collect_trusted()
|
121
|
+
} else if first_value.is_some() {
|
122
|
+
std::iter::once(first_value)
|
123
|
+
.chain(it)
|
124
|
+
.trust_my_length(capacity)
|
125
|
+
.collect_trusted()
|
126
|
+
} else {
|
127
|
+
it.collect()
|
128
|
+
}
|
129
|
+
};
|
130
|
+
debug_assert_eq!(ca.len(), capacity);
|
131
|
+
ca.rename(name);
|
132
|
+
ca
|
133
|
+
}
|
134
|
+
|
135
|
+
fn iterator_to_bool(
|
136
|
+
it: impl Iterator<Item = Option<bool>>,
|
137
|
+
init_null_count: usize,
|
138
|
+
first_value: Option<bool>,
|
139
|
+
name: &str,
|
140
|
+
capacity: usize,
|
141
|
+
) -> ChunkedArray<BooleanType> {
|
142
|
+
// safety: we know the iterators len
|
143
|
+
let mut ca: BooleanChunked = unsafe {
|
144
|
+
if init_null_count > 0 {
|
145
|
+
(0..init_null_count)
|
146
|
+
.map(|_| None)
|
147
|
+
.chain(std::iter::once(first_value))
|
148
|
+
.chain(it)
|
149
|
+
.trust_my_length(capacity)
|
150
|
+
.collect_trusted()
|
151
|
+
} else if first_value.is_some() {
|
152
|
+
std::iter::once(first_value)
|
153
|
+
.chain(it)
|
154
|
+
.trust_my_length(capacity)
|
155
|
+
.collect_trusted()
|
156
|
+
} else {
|
157
|
+
it.collect()
|
158
|
+
}
|
159
|
+
};
|
160
|
+
debug_assert_eq!(ca.len(), capacity);
|
161
|
+
ca.rename(name);
|
162
|
+
ca
|
163
|
+
}
|
164
|
+
|
165
|
+
fn iterator_to_object(
|
166
|
+
it: impl Iterator<Item = Option<ObjectValue>>,
|
167
|
+
init_null_count: usize,
|
168
|
+
first_value: Option<ObjectValue>,
|
169
|
+
name: &str,
|
170
|
+
capacity: usize,
|
171
|
+
) -> ObjectChunked<ObjectValue> {
|
172
|
+
// safety: we know the iterators len
|
173
|
+
let mut ca: ObjectChunked<ObjectValue> = unsafe {
|
174
|
+
if init_null_count > 0 {
|
175
|
+
(0..init_null_count)
|
176
|
+
.map(|_| None)
|
177
|
+
.chain(std::iter::once(first_value))
|
178
|
+
.chain(it)
|
179
|
+
.trust_my_length(capacity)
|
180
|
+
.collect_trusted()
|
181
|
+
} else if first_value.is_some() {
|
182
|
+
std::iter::once(first_value)
|
183
|
+
.chain(it)
|
184
|
+
.trust_my_length(capacity)
|
185
|
+
.collect_trusted()
|
186
|
+
} else {
|
187
|
+
it.collect()
|
188
|
+
}
|
189
|
+
};
|
190
|
+
debug_assert_eq!(ca.len(), capacity);
|
191
|
+
ca.rename(name);
|
192
|
+
ca
|
193
|
+
}
|
194
|
+
|
195
|
+
fn iterator_to_utf8(
|
196
|
+
it: impl Iterator<Item = Option<String>>,
|
197
|
+
init_null_count: usize,
|
198
|
+
first_value: Option<&str>,
|
199
|
+
name: &str,
|
200
|
+
capacity: usize,
|
201
|
+
) -> Utf8Chunked {
|
202
|
+
let first_value = first_value.map(|v| v.to_string());
|
203
|
+
|
204
|
+
// safety: we know the iterators len
|
205
|
+
let mut ca: Utf8Chunked = unsafe {
|
206
|
+
if init_null_count > 0 {
|
207
|
+
(0..init_null_count)
|
208
|
+
.map(|_| None)
|
209
|
+
.chain(std::iter::once(first_value))
|
210
|
+
.chain(it)
|
211
|
+
.trust_my_length(capacity)
|
212
|
+
.collect_trusted()
|
213
|
+
} else if first_value.is_some() {
|
214
|
+
std::iter::once(first_value)
|
215
|
+
.chain(it)
|
216
|
+
.trust_my_length(capacity)
|
217
|
+
.collect_trusted()
|
218
|
+
} else {
|
219
|
+
it.collect()
|
220
|
+
}
|
221
|
+
};
|
222
|
+
debug_assert_eq!(ca.len(), capacity);
|
223
|
+
ca.rename(name);
|
224
|
+
ca
|
225
|
+
}
|
226
|
+
|
227
|
+
fn iterator_to_list(
|
228
|
+
dt: &DataType,
|
229
|
+
it: impl Iterator<Item = Option<Series>>,
|
230
|
+
init_null_count: usize,
|
231
|
+
first_value: Option<&Series>,
|
232
|
+
name: &str,
|
233
|
+
capacity: usize,
|
234
|
+
) -> RbResult<ListChunked> {
|
235
|
+
let mut builder =
|
236
|
+
get_list_builder(dt, capacity * 5, capacity, name).map_err(RbPolarsErr::from)?;
|
237
|
+
for _ in 0..init_null_count {
|
238
|
+
builder.append_null()
|
239
|
+
}
|
240
|
+
builder.append_opt_series(first_value);
|
241
|
+
for opt_val in it {
|
242
|
+
match opt_val {
|
243
|
+
None => builder.append_null(),
|
244
|
+
Some(s) => {
|
245
|
+
if s.len() == 0 && s.dtype() != dt {
|
246
|
+
builder.append_series(&Series::full_null("", 0, dt))
|
247
|
+
} else {
|
248
|
+
builder.append_series(&s)
|
249
|
+
}
|
250
|
+
}
|
251
|
+
}
|
252
|
+
}
|
253
|
+
Ok(builder.finish())
|
254
|
+
}
|