polars-df 0.13.0 → 0.15.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +30 -0
- data/Cargo.lock +1368 -319
- data/LICENSE.txt +1 -0
- data/README.md +1 -2
- data/ext/polars/Cargo.toml +15 -6
- data/ext/polars/src/batched_csv.rs +10 -13
- data/ext/polars/src/conversion/any_value.rs +37 -21
- data/ext/polars/src/conversion/chunked_array.rs +3 -3
- data/ext/polars/src/conversion/mod.rs +159 -46
- data/ext/polars/src/dataframe/construction.rs +4 -7
- data/ext/polars/src/dataframe/export.rs +9 -2
- data/ext/polars/src/dataframe/general.rs +22 -16
- data/ext/polars/src/dataframe/io.rs +78 -174
- data/ext/polars/src/dataframe/mod.rs +1 -0
- data/ext/polars/src/dataframe/serde.rs +15 -0
- data/ext/polars/src/error.rs +31 -48
- data/ext/polars/src/exceptions.rs +24 -0
- data/ext/polars/src/expr/binary.rs +4 -42
- data/ext/polars/src/expr/datetime.rs +16 -7
- data/ext/polars/src/expr/general.rs +14 -23
- data/ext/polars/src/expr/list.rs +18 -11
- data/ext/polars/src/expr/name.rs +3 -2
- data/ext/polars/src/expr/rolling.rs +6 -7
- data/ext/polars/src/expr/string.rs +17 -37
- data/ext/polars/src/file.rs +59 -22
- data/ext/polars/src/functions/business.rs +15 -0
- data/ext/polars/src/functions/io.rs +6 -6
- data/ext/polars/src/functions/lazy.rs +17 -8
- data/ext/polars/src/functions/mod.rs +1 -0
- data/ext/polars/src/functions/range.rs +4 -2
- data/ext/polars/src/interop/arrow/mod.rs +1 -0
- data/ext/polars/src/interop/arrow/to_ruby.rs +83 -0
- data/ext/polars/src/interop/mod.rs +1 -0
- data/ext/polars/src/lazyframe/general.rs +877 -0
- data/ext/polars/src/lazyframe/mod.rs +3 -825
- data/ext/polars/src/lazyframe/serde.rs +31 -0
- data/ext/polars/src/lib.rs +44 -13
- data/ext/polars/src/map/dataframe.rs +46 -14
- data/ext/polars/src/map/lazy.rs +65 -4
- data/ext/polars/src/map/mod.rs +17 -16
- data/ext/polars/src/map/series.rs +106 -64
- data/ext/polars/src/on_startup.rs +2 -2
- data/ext/polars/src/series/aggregation.rs +1 -5
- data/ext/polars/src/series/arithmetic.rs +10 -10
- data/ext/polars/src/series/construction.rs +52 -25
- data/ext/polars/src/series/export.rs +1 -1
- data/ext/polars/src/series/general.rs +643 -0
- data/ext/polars/src/series/import.rs +55 -0
- data/ext/polars/src/series/mod.rs +11 -638
- data/ext/polars/src/series/scatter.rs +2 -2
- data/ext/polars/src/utils.rs +0 -20
- data/lib/polars/batched_csv_reader.rb +0 -2
- data/lib/polars/binary_expr.rb +133 -9
- data/lib/polars/binary_name_space.rb +101 -6
- data/lib/polars/config.rb +4 -0
- data/lib/polars/data_frame.rb +285 -62
- data/lib/polars/data_type_group.rb +28 -0
- data/lib/polars/data_types.rb +2 -0
- data/lib/polars/date_time_expr.rb +244 -0
- data/lib/polars/date_time_name_space.rb +87 -0
- data/lib/polars/expr.rb +109 -8
- data/lib/polars/functions/as_datatype.rb +51 -2
- data/lib/polars/functions/col.rb +1 -1
- data/lib/polars/functions/eager.rb +1 -3
- data/lib/polars/functions/lazy.rb +88 -10
- data/lib/polars/functions/range/time_range.rb +21 -21
- data/lib/polars/io/csv.rb +14 -16
- data/lib/polars/io/database.rb +2 -2
- data/lib/polars/io/ipc.rb +14 -12
- data/lib/polars/io/ndjson.rb +10 -0
- data/lib/polars/io/parquet.rb +168 -111
- data/lib/polars/lazy_frame.rb +649 -15
- data/lib/polars/list_name_space.rb +169 -0
- data/lib/polars/selectors.rb +1144 -0
- data/lib/polars/series.rb +470 -40
- data/lib/polars/string_cache.rb +27 -1
- data/lib/polars/string_expr.rb +0 -1
- data/lib/polars/string_name_space.rb +73 -3
- data/lib/polars/struct_name_space.rb +31 -7
- data/lib/polars/utils/various.rb +5 -1
- data/lib/polars/utils.rb +45 -10
- data/lib/polars/version.rb +1 -1
- data/lib/polars.rb +2 -1
- metadata +14 -4
- data/lib/polars/functions.rb +0 -57
@@ -0,0 +1,1144 @@
|
|
1
|
+
module Polars
|
2
|
+
module Selectors
|
3
|
+
# @private
|
4
|
+
class SelectorProxy < Expr
|
5
|
+
attr_accessor :_attrs
|
6
|
+
attr_accessor :_repr_override
|
7
|
+
|
8
|
+
def initialize(
|
9
|
+
expr,
|
10
|
+
name:,
|
11
|
+
parameters: nil
|
12
|
+
)
|
13
|
+
self._rbexpr = expr._rbexpr
|
14
|
+
self._attrs = {
|
15
|
+
name: name,
|
16
|
+
params: parameters
|
17
|
+
}
|
18
|
+
end
|
19
|
+
|
20
|
+
def inspect
|
21
|
+
if !_attrs
|
22
|
+
as_expr.inspect
|
23
|
+
elsif _repr_override
|
24
|
+
_repr_override
|
25
|
+
else
|
26
|
+
selector_name = _attrs[:name]
|
27
|
+
params = _attrs[:params] || {}
|
28
|
+
set_ops = {"and" => "&", "or" => "|", "sub" => "-", "xor" => "^"}
|
29
|
+
if set_ops.include?(selector_name)
|
30
|
+
op = set_ops[selector_name]
|
31
|
+
"(#{params.values.map(&:inspect).join(" #{op} ")})"
|
32
|
+
else
|
33
|
+
str_params = params.map { |k, v| k.start_with?("*") ? v.inspect[1..-2] : "#{k}=#{v.inspect}" }.join(", ")
|
34
|
+
"Polars.cs.#{selector_name}(#{str_params})"
|
35
|
+
end
|
36
|
+
end
|
37
|
+
end
|
38
|
+
|
39
|
+
def ~
|
40
|
+
if Utils.is_selector(self)
|
41
|
+
inverted = Selectors.all - self
|
42
|
+
inverted._repr_override = "~#{inspect}"
|
43
|
+
else
|
44
|
+
inverted = ~as_expr
|
45
|
+
end
|
46
|
+
inverted
|
47
|
+
end
|
48
|
+
|
49
|
+
def -(other)
|
50
|
+
if Utils.is_selector(other)
|
51
|
+
SelectorProxy.new(
|
52
|
+
meta._as_selector.meta._selector_sub(other),
|
53
|
+
parameters: {"self" => self, "other" => other},
|
54
|
+
name: "sub"
|
55
|
+
)
|
56
|
+
else
|
57
|
+
as_expr - other
|
58
|
+
end
|
59
|
+
end
|
60
|
+
|
61
|
+
def &(other)
|
62
|
+
if Utils.is_column(other)
|
63
|
+
raise Todo
|
64
|
+
end
|
65
|
+
if Utils.is_selector(other)
|
66
|
+
SelectorProxy.new(
|
67
|
+
meta._as_selector.meta._selector_and(other),
|
68
|
+
parameters: {"self" => self, "other" => other},
|
69
|
+
name: "and"
|
70
|
+
)
|
71
|
+
else
|
72
|
+
as_expr & other
|
73
|
+
end
|
74
|
+
end
|
75
|
+
|
76
|
+
def |(other)
|
77
|
+
if Utils.is_column(other)
|
78
|
+
raise Todo
|
79
|
+
end
|
80
|
+
if Utils.is_selector(other)
|
81
|
+
SelectorProxy.new(
|
82
|
+
meta._as_selector.meta._selector_and(other),
|
83
|
+
parameters: {"self" => self, "other" => other},
|
84
|
+
name: "or"
|
85
|
+
)
|
86
|
+
else
|
87
|
+
as_expr | other
|
88
|
+
end
|
89
|
+
end
|
90
|
+
|
91
|
+
def ^(other)
|
92
|
+
if Utils.is_column(other)
|
93
|
+
raise Todo
|
94
|
+
end
|
95
|
+
if Utils.is_selector(other)
|
96
|
+
SelectorProxy.new(
|
97
|
+
meta._as_selector.meta._selector_and(other),
|
98
|
+
parameters: {"self" => self, "other" => other},
|
99
|
+
name: "xor"
|
100
|
+
)
|
101
|
+
else
|
102
|
+
as_expr ^ other
|
103
|
+
end
|
104
|
+
end
|
105
|
+
|
106
|
+
def as_expr
|
107
|
+
Expr._from_rbexpr(_rbexpr)
|
108
|
+
end
|
109
|
+
end
|
110
|
+
|
111
|
+
# @private
|
112
|
+
def self._selector_proxy_(...)
|
113
|
+
SelectorProxy.new(...)
|
114
|
+
end
|
115
|
+
|
116
|
+
# @private
|
117
|
+
def self._re_string(string, escape: true)
|
118
|
+
if string.is_a?(::String)
|
119
|
+
rx = escape ? Utils.re_escape(string) : string
|
120
|
+
else
|
121
|
+
strings = []
|
122
|
+
string.each do |st|
|
123
|
+
if st.is_a?(Array)
|
124
|
+
strings.concat(st)
|
125
|
+
else
|
126
|
+
strings << st
|
127
|
+
end
|
128
|
+
end
|
129
|
+
rx = strings.map { |x| escape ? Utils.re_escape(x) : x }.join("|")
|
130
|
+
end
|
131
|
+
"(#{rx})"
|
132
|
+
end
|
133
|
+
|
134
|
+
# Select all columns.
|
135
|
+
#
|
136
|
+
# @return [SelectorProxy]
|
137
|
+
#
|
138
|
+
# @example
|
139
|
+
# df = Polars::DataFrame.new(
|
140
|
+
# {
|
141
|
+
# "dt" => [Date.new(1999, 12, 31), Date.new(2024, 1, 1)],
|
142
|
+
# "value" => [1_234_500, 5_000_555]
|
143
|
+
# },
|
144
|
+
# schema_overrides: {"value" => Polars::Int32}
|
145
|
+
# )
|
146
|
+
#
|
147
|
+
# @example Select all columns, casting them to string:
|
148
|
+
# df.select(Polars.cs.all.cast(Polars::String))
|
149
|
+
# # =>
|
150
|
+
# # shape: (2, 2)
|
151
|
+
# # ┌────────────┬─────────┐
|
152
|
+
# # │ dt ┆ value │
|
153
|
+
# # │ --- ┆ --- │
|
154
|
+
# # │ str ┆ str │
|
155
|
+
# # ╞════════════╪═════════╡
|
156
|
+
# # │ 1999-12-31 ┆ 1234500 │
|
157
|
+
# # │ 2024-01-01 ┆ 5000555 │
|
158
|
+
# # └────────────┴─────────┘
|
159
|
+
#
|
160
|
+
# @example Select all columns *except* for those matching the given dtypes:
|
161
|
+
# df.select(Polars.cs.all - Polars.cs.numeric)
|
162
|
+
# # =>
|
163
|
+
# # shape: (2, 1)
|
164
|
+
# # ┌────────────┐
|
165
|
+
# # │ dt │
|
166
|
+
# # │ --- │
|
167
|
+
# # │ date │
|
168
|
+
# # ╞════════════╡
|
169
|
+
# # │ 1999-12-31 │
|
170
|
+
# # │ 2024-01-01 │
|
171
|
+
# # └────────────┘
|
172
|
+
def self.all
|
173
|
+
_selector_proxy_(F.all, name: "all")
|
174
|
+
end
|
175
|
+
|
176
|
+
# Select all columns with alphabetic names (eg: only letters).
|
177
|
+
#
|
178
|
+
# @param ascii_only [Boolean]
|
179
|
+
# Indicate whether to consider only ASCII alphabetic characters, or the full
|
180
|
+
# Unicode range of valid letters (accented, idiographic, etc).
|
181
|
+
# @param ignore_spaces [Boolean]
|
182
|
+
# Indicate whether to ignore the presence of spaces in column names; if so,
|
183
|
+
# only the other (non-space) characters are considered.
|
184
|
+
#
|
185
|
+
# @return [SelectorProxy]
|
186
|
+
#
|
187
|
+
# @note
|
188
|
+
# Matching column names cannot contain *any* non-alphabetic characters. Note
|
189
|
+
# that the definition of "alphabetic" consists of all valid Unicode alphabetic
|
190
|
+
# characters (`\p{Alphabetic}`) by default; this can be changed by setting
|
191
|
+
# `ascii_only: true`.
|
192
|
+
#
|
193
|
+
# @example
|
194
|
+
# df = Polars::DataFrame.new(
|
195
|
+
# {
|
196
|
+
# "no1" => [100, 200, 300],
|
197
|
+
# "café" => ["espresso", "latte", "mocha"],
|
198
|
+
# "t or f" => [true, false, nil],
|
199
|
+
# "hmm" => ["aaa", "bbb", "ccc"],
|
200
|
+
# "都市" => ["東京", "大阪", "京都"]
|
201
|
+
# }
|
202
|
+
# )
|
203
|
+
#
|
204
|
+
# @example Select columns with alphabetic names; note that accented characters and kanji are recognised as alphabetic here:
|
205
|
+
# df.select(Polars.cs.alpha)
|
206
|
+
# # =>
|
207
|
+
# # shape: (3, 3)
|
208
|
+
# # ┌──────────┬─────┬──────┐
|
209
|
+
# # │ café ┆ hmm ┆ 都市 │
|
210
|
+
# # │ --- ┆ --- ┆ --- │
|
211
|
+
# # │ str ┆ str ┆ str │
|
212
|
+
# # ╞══════════╪═════╪══════╡
|
213
|
+
# # │ espresso ┆ aaa ┆ 東京 │
|
214
|
+
# # │ latte ┆ bbb ┆ 大阪 │
|
215
|
+
# # │ mocha ┆ ccc ┆ 京都 │
|
216
|
+
# # └──────────┴─────┴──────┘
|
217
|
+
#
|
218
|
+
# @example Constrain the definition of "alphabetic" to ASCII characters only:
|
219
|
+
# df.select(Polars.cs.alpha(ascii_only: true))
|
220
|
+
# # =>
|
221
|
+
# # shape: (3, 1)
|
222
|
+
# # ┌─────┐
|
223
|
+
# # │ hmm │
|
224
|
+
# # │ --- │
|
225
|
+
# # │ str │
|
226
|
+
# # ╞═════╡
|
227
|
+
# # │ aaa │
|
228
|
+
# # │ bbb │
|
229
|
+
# # │ ccc │
|
230
|
+
# # └─────┘
|
231
|
+
#
|
232
|
+
# @example
|
233
|
+
# df.select(Polars.cs.alpha(ascii_only: true, ignore_spaces: true))
|
234
|
+
# # =>
|
235
|
+
# # shape: (3, 2)
|
236
|
+
# # ┌────────┬─────┐
|
237
|
+
# # │ t or f ┆ hmm │
|
238
|
+
# # │ --- ┆ --- │
|
239
|
+
# # │ bool ┆ str │
|
240
|
+
# # ╞════════╪═════╡
|
241
|
+
# # │ true ┆ aaa │
|
242
|
+
# # │ false ┆ bbb │
|
243
|
+
# # │ null ┆ ccc │
|
244
|
+
# # └────────┴─────┘
|
245
|
+
#
|
246
|
+
# @example Select all columns *except* for those with alphabetic names:
|
247
|
+
# df.select(~Polars.cs.alpha)
|
248
|
+
# # =>
|
249
|
+
# # shape: (3, 2)
|
250
|
+
# # ┌─────┬────────┐
|
251
|
+
# # │ no1 ┆ t or f │
|
252
|
+
# # │ --- ┆ --- │
|
253
|
+
# # │ i64 ┆ bool │
|
254
|
+
# # ╞═════╪════════╡
|
255
|
+
# # │ 100 ┆ true │
|
256
|
+
# # │ 200 ┆ false │
|
257
|
+
# # │ 300 ┆ null │
|
258
|
+
# # └─────┴────────┘
|
259
|
+
#
|
260
|
+
# @example
|
261
|
+
# df.select(~Polars.cs.alpha(ignore_spaces: true))
|
262
|
+
# # =>
|
263
|
+
# # shape: (3, 1)
|
264
|
+
# # ┌─────┐
|
265
|
+
# # │ no1 │
|
266
|
+
# # │ --- │
|
267
|
+
# # │ i64 │
|
268
|
+
# # ╞═════╡
|
269
|
+
# # │ 100 │
|
270
|
+
# # │ 200 │
|
271
|
+
# # │ 300 │
|
272
|
+
# # └─────┘
|
273
|
+
def self.alpha(ascii_only: false, ignore_spaces: false)
|
274
|
+
# note that we need to supply a pattern compatible with the *rust* regex crate
|
275
|
+
re_alpha = ascii_only ? "a-zA-Z" : "\\p{Alphabetic}"
|
276
|
+
re_space = ignore_spaces ? " " : ""
|
277
|
+
_selector_proxy_(
|
278
|
+
F.col("^[#{re_alpha}#{re_space}]+$"),
|
279
|
+
name: "alpha",
|
280
|
+
parameters: {"ascii_only" => ascii_only, "ignore_spaces" => ignore_spaces},
|
281
|
+
)
|
282
|
+
end
|
283
|
+
|
284
|
+
# TODO
|
285
|
+
# def alphanumeric
|
286
|
+
# end
|
287
|
+
|
288
|
+
# Select all binary columns.
|
289
|
+
#
|
290
|
+
# @return [SelectorProxy]
|
291
|
+
#
|
292
|
+
# @example
|
293
|
+
# df = Polars::DataFrame.new({"a" => ["hello".b], "b" => ["world"], "c" => ["!".b], "d" => [":)"]})
|
294
|
+
# # =>
|
295
|
+
# # shape: (1, 4)
|
296
|
+
# # ┌──────────┬───────┬────────┬─────┐
|
297
|
+
# # │ a ┆ b ┆ c ┆ d │
|
298
|
+
# # │ --- ┆ --- ┆ --- ┆ --- │
|
299
|
+
# # │ binary ┆ str ┆ binary ┆ str │
|
300
|
+
# # ╞══════════╪═══════╪════════╪═════╡
|
301
|
+
# # │ b"hello" ┆ world ┆ b"!" ┆ :) │
|
302
|
+
# # └──────────┴───────┴────────┴─────┘
|
303
|
+
#
|
304
|
+
# @example Select binary columns and export as a dict:
|
305
|
+
# df.select(Polars.cs.binary).to_h(as_series: false)
|
306
|
+
# # => {"a"=>["hello"], "c"=>["!"]}
|
307
|
+
#
|
308
|
+
# @example Select all columns *except* for those that are binary:
|
309
|
+
# df.select(~Polars.cs.binary).to_h(as_series: false)
|
310
|
+
# # => {"b"=>["world"], "d"=>[":)"]}
|
311
|
+
def self.binary
|
312
|
+
_selector_proxy_(F.col(Binary), name: "binary")
|
313
|
+
end
|
314
|
+
|
315
|
+
# Select all boolean columns.
|
316
|
+
#
|
317
|
+
# @return [SelectorProxy]
|
318
|
+
#
|
319
|
+
# @example
|
320
|
+
# df = Polars::DataFrame.new({"n" => 1..4}).with_columns(n_even: Polars.col("n") % 2 == 0)
|
321
|
+
# # =>
|
322
|
+
# # shape: (4, 2)
|
323
|
+
# # ┌─────┬────────┐
|
324
|
+
# # │ n ┆ n_even │
|
325
|
+
# # │ --- ┆ --- │
|
326
|
+
# # │ i64 ┆ bool │
|
327
|
+
# # ╞═════╪════════╡
|
328
|
+
# # │ 1 ┆ false │
|
329
|
+
# # │ 2 ┆ true │
|
330
|
+
# # │ 3 ┆ false │
|
331
|
+
# # │ 4 ┆ true │
|
332
|
+
# # └─────┴────────┘
|
333
|
+
#
|
334
|
+
# @example Select and invert boolean columns:
|
335
|
+
# df.with_columns(is_odd: Polars.cs.boolean.not_)
|
336
|
+
# # =>
|
337
|
+
# # shape: (4, 3)
|
338
|
+
# # ┌─────┬────────┬────────┐
|
339
|
+
# # │ n ┆ n_even ┆ is_odd │
|
340
|
+
# # │ --- ┆ --- ┆ --- │
|
341
|
+
# # │ i64 ┆ bool ┆ bool │
|
342
|
+
# # ╞═════╪════════╪════════╡
|
343
|
+
# # │ 1 ┆ false ┆ true │
|
344
|
+
# # │ 2 ┆ true ┆ false │
|
345
|
+
# # │ 3 ┆ false ┆ true │
|
346
|
+
# # │ 4 ┆ true ┆ false │
|
347
|
+
# # └─────┴────────┴────────┘
|
348
|
+
#
|
349
|
+
# @example Select all columns *except* for those that are boolean:
|
350
|
+
# df.select(~Polars.cs.boolean)
|
351
|
+
# # =>
|
352
|
+
# # shape: (4, 1)
|
353
|
+
# # ┌─────┐
|
354
|
+
# # │ n │
|
355
|
+
# # │ --- │
|
356
|
+
# # │ i64 │
|
357
|
+
# # ╞═════╡
|
358
|
+
# # │ 1 │
|
359
|
+
# # │ 2 │
|
360
|
+
# # │ 3 │
|
361
|
+
# # │ 4 │
|
362
|
+
# # └─────┘
|
363
|
+
def self.boolean
|
364
|
+
_selector_proxy_(F.col(Boolean), name: "boolean")
|
365
|
+
end
|
366
|
+
|
367
|
+
# TODO
|
368
|
+
# def by_dtype
|
369
|
+
# end
|
370
|
+
|
371
|
+
# TODO
|
372
|
+
# def by_index
|
373
|
+
# end
|
374
|
+
|
375
|
+
# TODO
|
376
|
+
# def by_name
|
377
|
+
# end
|
378
|
+
|
379
|
+
# Select all categorical columns.
|
380
|
+
#
|
381
|
+
# @return [SelectorProxy]
|
382
|
+
#
|
383
|
+
# @example
|
384
|
+
# df = Polars::DataFrame.new(
|
385
|
+
# {
|
386
|
+
# "foo" => ["xx", "yy"],
|
387
|
+
# "bar" => [123, 456],
|
388
|
+
# "baz" => [2.0, 5.5]
|
389
|
+
# },
|
390
|
+
# schema_overrides: {"foo" => Polars::Categorical}
|
391
|
+
# )
|
392
|
+
#
|
393
|
+
# @example Select all categorical columns:
|
394
|
+
# df.select(Polars.cs.categorical)
|
395
|
+
# # =>
|
396
|
+
# # shape: (2, 1)
|
397
|
+
# # ┌─────┐
|
398
|
+
# # │ foo │
|
399
|
+
# # │ --- │
|
400
|
+
# # │ cat │
|
401
|
+
# # ╞═════╡
|
402
|
+
# # │ xx │
|
403
|
+
# # │ yy │
|
404
|
+
# # └─────┘
|
405
|
+
#
|
406
|
+
# @example Select all columns *except* for those that are categorical:
|
407
|
+
# df.select(~Polars.cs.categorical)
|
408
|
+
# # =>
|
409
|
+
# # shape: (2, 2)
|
410
|
+
# # ┌─────┬─────┐
|
411
|
+
# # │ bar ┆ baz │
|
412
|
+
# # │ --- ┆ --- │
|
413
|
+
# # │ i64 ┆ f64 │
|
414
|
+
# # ╞═════╪═════╡
|
415
|
+
# # │ 123 ┆ 2.0 │
|
416
|
+
# # │ 456 ┆ 5.5 │
|
417
|
+
# # └─────┴─────┘
|
418
|
+
def self.categorical
|
419
|
+
_selector_proxy_(F.col(Categorical), name: "categorical")
|
420
|
+
end
|
421
|
+
|
422
|
+
# Select columns whose names contain the given literal substring(s).
|
423
|
+
#
|
424
|
+
# @param substring [Object]
|
425
|
+
# Substring(s) that matching column names should contain.
|
426
|
+
#
|
427
|
+
# @return [SelectorProxy]
|
428
|
+
#
|
429
|
+
# @example
|
430
|
+
# df = Polars::DataFrame.new(
|
431
|
+
# {
|
432
|
+
# "foo" => ["x", "y"],
|
433
|
+
# "bar" => [123, 456],
|
434
|
+
# "baz" => [2.0, 5.5],
|
435
|
+
# "zap" => [false, true]
|
436
|
+
# }
|
437
|
+
# )
|
438
|
+
#
|
439
|
+
# @example Select columns that contain the substring 'ba':
|
440
|
+
# df.select(Polars.cs.contains("ba"))
|
441
|
+
# # =>
|
442
|
+
# # shape: (2, 2)
|
443
|
+
# # ┌─────┬─────┐
|
444
|
+
# # │ bar ┆ baz │
|
445
|
+
# # │ --- ┆ --- │
|
446
|
+
# # │ i64 ┆ f64 │
|
447
|
+
# # ╞═════╪═════╡
|
448
|
+
# # │ 123 ┆ 2.0 │
|
449
|
+
# # │ 456 ┆ 5.5 │
|
450
|
+
# # └─────┴─────┘
|
451
|
+
#
|
452
|
+
# @example Select columns that contain the substring 'ba' or the letter 'z':
|
453
|
+
# df.select(Polars.cs.contains("ba", "z"))
|
454
|
+
# # =>
|
455
|
+
# # shape: (2, 3)
|
456
|
+
# # ┌─────┬─────┬───────┐
|
457
|
+
# # │ bar ┆ baz ┆ zap │
|
458
|
+
# # │ --- ┆ --- ┆ --- │
|
459
|
+
# # │ i64 ┆ f64 ┆ bool │
|
460
|
+
# # ╞═════╪═════╪═══════╡
|
461
|
+
# # │ 123 ┆ 2.0 ┆ false │
|
462
|
+
# # │ 456 ┆ 5.5 ┆ true │
|
463
|
+
# # └─────┴─────┴───────┘
|
464
|
+
#
|
465
|
+
# @example Select all columns *except* for those that contain the substring 'ba':
|
466
|
+
# df.select(~Polars.cs.contains("ba"))
|
467
|
+
# # =>
|
468
|
+
# # shape: (2, 2)
|
469
|
+
# # ┌─────┬───────┐
|
470
|
+
# # │ foo ┆ zap │
|
471
|
+
# # │ --- ┆ --- │
|
472
|
+
# # │ str ┆ bool │
|
473
|
+
# # ╞═════╪═══════╡
|
474
|
+
# # │ x ┆ false │
|
475
|
+
# # │ y ┆ true │
|
476
|
+
# # └─────┴───────┘
|
477
|
+
def self.contains(*substring)
|
478
|
+
escaped_substring = _re_string(substring)
|
479
|
+
raw_params = "^.*#{escaped_substring}.*$"
|
480
|
+
|
481
|
+
_selector_proxy_(
|
482
|
+
F.col(raw_params),
|
483
|
+
name: "contains",
|
484
|
+
parameters: {"*substring" => escaped_substring}
|
485
|
+
)
|
486
|
+
end
|
487
|
+
|
488
|
+
# Select all date columns.
|
489
|
+
#
|
490
|
+
# @return [SelectorProxy]
|
491
|
+
#
|
492
|
+
# @example
|
493
|
+
# df = Polars::DataFrame.new(
|
494
|
+
# {
|
495
|
+
# "dtm" => [DateTime.new(2001, 5, 7, 10, 25), DateTime.new(2031, 12, 31, 0, 30)],
|
496
|
+
# "dt" => [Date.new(1999, 12, 31), Date.new(2024, 8, 9)]
|
497
|
+
# }
|
498
|
+
# )
|
499
|
+
#
|
500
|
+
# @example Select all date columns:
|
501
|
+
# df.select(Polars.cs.date)
|
502
|
+
# # =>
|
503
|
+
# # shape: (2, 1)
|
504
|
+
# # ┌────────────┐
|
505
|
+
# # │ dt │
|
506
|
+
# # │ --- │
|
507
|
+
# # │ date │
|
508
|
+
# # ╞════════════╡
|
509
|
+
# # │ 1999-12-31 │
|
510
|
+
# # │ 2024-08-09 │
|
511
|
+
# # └────────────┘
|
512
|
+
#
|
513
|
+
# @example Select all columns *except* for those that are dates:
|
514
|
+
# df.select(~Polars.cs.date)
|
515
|
+
# # =>
|
516
|
+
# # shape: (2, 1)
|
517
|
+
# # ┌─────────────────────┐
|
518
|
+
# # │ dtm │
|
519
|
+
# # │ --- │
|
520
|
+
# # │ datetime[ns] │
|
521
|
+
# # ╞═════════════════════╡
|
522
|
+
# # │ 2001-05-07 10:25:00 │
|
523
|
+
# # │ 2031-12-31 00:30:00 │
|
524
|
+
# # └─────────────────────┘
|
525
|
+
def self.date
|
526
|
+
_selector_proxy_(F.col(Date), name: "date")
|
527
|
+
end
|
528
|
+
|
529
|
+
# TODO
|
530
|
+
# def datetime
|
531
|
+
# end
|
532
|
+
|
533
|
+
# Select all decimal columns.
|
534
|
+
#
|
535
|
+
# @return [SelectorProxy]
|
536
|
+
#
|
537
|
+
# @example
|
538
|
+
# df = Polars::DataFrame.new(
|
539
|
+
# {
|
540
|
+
# "foo" => ["x", "y"],
|
541
|
+
# "bar" => [BigDecimal("123"), BigDecimal("456")],
|
542
|
+
# "baz" => [BigDecimal("2.0005"), BigDecimal("-50.5555")],
|
543
|
+
# },
|
544
|
+
# schema_overrides: {"baz" => Polars::Decimal.new(10, 5)}
|
545
|
+
# )
|
546
|
+
#
|
547
|
+
# @example Select all decimal columns:
|
548
|
+
# df.select(Polars.cs.decimal)
|
549
|
+
# # =>
|
550
|
+
# # shape: (2, 2)
|
551
|
+
# # ┌──────────────┬───────────────┐
|
552
|
+
# # │ bar ┆ baz │
|
553
|
+
# # │ --- ┆ --- │
|
554
|
+
# # │ decimal[*,0] ┆ decimal[10,5] │
|
555
|
+
# # ╞══════════════╪═══════════════╡
|
556
|
+
# # │ 123 ┆ 2.00050 │
|
557
|
+
# # │ 456 ┆ -50.55550 │
|
558
|
+
# # └──────────────┴───────────────┘
|
559
|
+
#
|
560
|
+
# @example Select all columns *except* the decimal ones:
|
561
|
+
#
|
562
|
+
# df.select(~Polars.cs.decimal)
|
563
|
+
# # =>
|
564
|
+
# # shape: (2, 1)
|
565
|
+
# # ┌─────┐
|
566
|
+
# # │ foo │
|
567
|
+
# # │ --- │
|
568
|
+
# # │ str │
|
569
|
+
# # ╞═════╡
|
570
|
+
# # │ x │
|
571
|
+
# # │ y │
|
572
|
+
# # └─────┘
|
573
|
+
def self.decimal
|
574
|
+
# TODO: allow explicit selection by scale/precision?
|
575
|
+
_selector_proxy_(F.col(Decimal), name: "decimal")
|
576
|
+
end
|
577
|
+
|
578
|
+
# Select columns that end with the given substring(s).
|
579
|
+
#
|
580
|
+
# @param suffix [Object]
|
581
|
+
# Substring(s) that matching column names should end with.
|
582
|
+
#
|
583
|
+
# @return [SelectorProxy]
|
584
|
+
#
|
585
|
+
# @example
|
586
|
+
# df = Polars::DataFrame.new(
|
587
|
+
# {
|
588
|
+
# "foo" => ["x", "y"],
|
589
|
+
# "bar" => [123, 456],
|
590
|
+
# "baz" => [2.0, 5.5],
|
591
|
+
# "zap" => [false, true]
|
592
|
+
# }
|
593
|
+
# )
|
594
|
+
#
|
595
|
+
# @example Select columns that end with the substring 'z':
|
596
|
+
# df.select(Polars.cs.ends_with("z"))
|
597
|
+
# # =>
|
598
|
+
# # shape: (2, 1)
|
599
|
+
# # ┌─────┐
|
600
|
+
# # │ baz │
|
601
|
+
# # │ --- │
|
602
|
+
# # │ f64 │
|
603
|
+
# # ╞═════╡
|
604
|
+
# # │ 2.0 │
|
605
|
+
# # │ 5.5 │
|
606
|
+
# # └─────┘
|
607
|
+
#
|
608
|
+
# @example Select columns that end with *either* the letter 'z' or 'r':
|
609
|
+
# df.select(Polars.cs.ends_with("z", "r"))
|
610
|
+
# # =>
|
611
|
+
# # shape: (2, 2)
|
612
|
+
# # ┌─────┬─────┐
|
613
|
+
# # │ bar ┆ baz │
|
614
|
+
# # │ --- ┆ --- │
|
615
|
+
# # │ i64 ┆ f64 │
|
616
|
+
# # ╞═════╪═════╡
|
617
|
+
# # │ 123 ┆ 2.0 │
|
618
|
+
# # │ 456 ┆ 5.5 │
|
619
|
+
# # └─────┴─────┘
|
620
|
+
#
|
621
|
+
# @example Select all columns *except* for those that end with the substring 'z':
|
622
|
+
# df.select(~Polars.cs.ends_with("z"))
|
623
|
+
# # =>
|
624
|
+
# # shape: (2, 3)
|
625
|
+
# # ┌─────┬─────┬───────┐
|
626
|
+
# # │ foo ┆ bar ┆ zap │
|
627
|
+
# # │ --- ┆ --- ┆ --- │
|
628
|
+
# # │ str ┆ i64 ┆ bool │
|
629
|
+
# # ╞═════╪═════╪═══════╡
|
630
|
+
# # │ x ┆ 123 ┆ false │
|
631
|
+
# # │ y ┆ 456 ┆ true │
|
632
|
+
# # └─────┴─────┴───────┘
|
633
|
+
def self.ends_with(*suffix)
|
634
|
+
escaped_suffix = _re_string(suffix)
|
635
|
+
raw_params = "^.*#{escaped_suffix}$"
|
636
|
+
|
637
|
+
_selector_proxy_(
|
638
|
+
F.col(raw_params),
|
639
|
+
name: "ends_with",
|
640
|
+
parameters: {"*suffix" => escaped_suffix},
|
641
|
+
)
|
642
|
+
end
|
643
|
+
|
644
|
+
# Select the first column in the current scope.
|
645
|
+
#
|
646
|
+
# @return [SelectorProxy]
|
647
|
+
#
|
648
|
+
# @example
|
649
|
+
# df = Polars::DataFrame.new(
|
650
|
+
# {
|
651
|
+
# "foo" => ["x", "y"],
|
652
|
+
# "bar" => [123, 456],
|
653
|
+
# "baz" => [2.0, 5.5],
|
654
|
+
# "zap" => [0, 1]
|
655
|
+
# }
|
656
|
+
# )
|
657
|
+
#
|
658
|
+
# @example Select the first column:
|
659
|
+
# df.select(Polars.cs.first)
|
660
|
+
# # =>
|
661
|
+
# # shape: (2, 1)
|
662
|
+
# # ┌─────┐
|
663
|
+
# # │ foo │
|
664
|
+
# # │ --- │
|
665
|
+
# # │ str │
|
666
|
+
# # ╞═════╡
|
667
|
+
# # │ x │
|
668
|
+
# # │ y │
|
669
|
+
# # └─────┘
|
670
|
+
#
|
671
|
+
# @example Select everything *except* for the first column:
|
672
|
+
# df.select(~Polars.cs.first)
|
673
|
+
# # =>
|
674
|
+
# # shape: (2, 3)
|
675
|
+
# # ┌─────┬─────┬─────┐
|
676
|
+
# # │ bar ┆ baz ┆ zap │
|
677
|
+
# # │ --- ┆ --- ┆ --- │
|
678
|
+
# # │ i64 ┆ f64 ┆ i64 │
|
679
|
+
# # ╞═════╪═════╪═════╡
|
680
|
+
# # │ 123 ┆ 2.0 ┆ 0 │
|
681
|
+
# # │ 456 ┆ 5.5 ┆ 1 │
|
682
|
+
# # └─────┴─────┴─────┘
|
683
|
+
def self.first
|
684
|
+
_selector_proxy_(F.first, name: "first")
|
685
|
+
end
|
686
|
+
|
687
|
+
# Select all float columns.
|
688
|
+
#
|
689
|
+
# @return [SelectorProxy]
|
690
|
+
#
|
691
|
+
# @example
|
692
|
+
# df = Polars::DataFrame.new(
|
693
|
+
# {
|
694
|
+
# "foo" => ["x", "y"],
|
695
|
+
# "bar" => [123, 456],
|
696
|
+
# "baz" => [2.0, 5.5],
|
697
|
+
# "zap" => [0.0, 1.0]
|
698
|
+
# },
|
699
|
+
# schema_overrides: {"baz" => Polars::Float32, "zap" => Polars::Float64}
|
700
|
+
# )
|
701
|
+
#
|
702
|
+
# @example Select all float columns:
|
703
|
+
# df.select(Polars.cs.float)
|
704
|
+
# # =>
|
705
|
+
# # shape: (2, 2)
|
706
|
+
# # ┌─────┬─────┐
|
707
|
+
# # │ baz ┆ zap │
|
708
|
+
# # │ --- ┆ --- │
|
709
|
+
# # │ f32 ┆ f64 │
|
710
|
+
# # ╞═════╪═════╡
|
711
|
+
# # │ 2.0 ┆ 0.0 │
|
712
|
+
# # │ 5.5 ┆ 1.0 │
|
713
|
+
# # └─────┴─────┘
|
714
|
+
#
|
715
|
+
# @example Select all columns *except* for those that are float:
|
716
|
+
# df.select(~Polars.cs.float)
|
717
|
+
# # =>
|
718
|
+
# # shape: (2, 2)
|
719
|
+
# # ┌─────┬─────┐
|
720
|
+
# # │ foo ┆ bar │
|
721
|
+
# # │ --- ┆ --- │
|
722
|
+
# # │ str ┆ i64 │
|
723
|
+
# # ╞═════╪═════╡
|
724
|
+
# # │ x ┆ 123 │
|
725
|
+
# # │ y ┆ 456 │
|
726
|
+
# # └─────┴─────┘
|
727
|
+
def self.float
|
728
|
+
_selector_proxy_(F.col(FLOAT_DTYPES), name: "float")
|
729
|
+
end
|
730
|
+
|
731
|
+
# Select all integer columns.
|
732
|
+
#
|
733
|
+
# @return [SelectorProxy]
|
734
|
+
#
|
735
|
+
# @example
|
736
|
+
# df = Polars::DataFrame.new(
|
737
|
+
# {
|
738
|
+
# "foo" => ["x", "y"],
|
739
|
+
# "bar" => [123, 456],
|
740
|
+
# "baz" => [2.0, 5.5],
|
741
|
+
# "zap" => [0, 1]
|
742
|
+
# }
|
743
|
+
# )
|
744
|
+
#
|
745
|
+
# @example Select all integer columns:
|
746
|
+
# df.select(Polars.cs.integer)
|
747
|
+
# # =>
|
748
|
+
# # shape: (2, 2)
|
749
|
+
# # ┌─────┬─────┐
|
750
|
+
# # │ bar ┆ zap │
|
751
|
+
# # │ --- ┆ --- │
|
752
|
+
# # │ i64 ┆ i64 │
|
753
|
+
# # ╞═════╪═════╡
|
754
|
+
# # │ 123 ┆ 0 │
|
755
|
+
# # │ 456 ┆ 1 │
|
756
|
+
# # └─────┴─────┘
|
757
|
+
#
|
758
|
+
# @example Select all columns *except* for those that are integer:
|
759
|
+
# df.select(~Polars.cs.integer)
|
760
|
+
# # =>
|
761
|
+
# # shape: (2, 2)
|
762
|
+
# # ┌─────┬─────┐
|
763
|
+
# # │ foo ┆ baz │
|
764
|
+
# # │ --- ┆ --- │
|
765
|
+
# # │ str ┆ f64 │
|
766
|
+
# # ╞═════╪═════╡
|
767
|
+
# # │ x ┆ 2.0 │
|
768
|
+
# # │ y ┆ 5.5 │
|
769
|
+
# # └─────┴─────┘
|
770
|
+
def self.integer
|
771
|
+
_selector_proxy_(F.col(INTEGER_DTYPES), name: "integer")
|
772
|
+
end
|
773
|
+
|
774
|
+
# Select all signed integer columns.
|
775
|
+
#
|
776
|
+
# @return [SelectorProxy]
|
777
|
+
#
|
778
|
+
# @example
|
779
|
+
# df = Polars::DataFrame.new(
|
780
|
+
# {
|
781
|
+
# "foo" => [-123, -456],
|
782
|
+
# "bar" => [3456, 6789],
|
783
|
+
# "baz" => [7654, 4321],
|
784
|
+
# "zap" => ["ab", "cd"]
|
785
|
+
# },
|
786
|
+
# schema_overrides: {"bar" => Polars::UInt32, "baz" => Polars::UInt64}
|
787
|
+
# )
|
788
|
+
#
|
789
|
+
# @example Select all signed integer columns:
|
790
|
+
# df.select(Polars.cs.signed_integer)
|
791
|
+
# # =>
|
792
|
+
# # shape: (2, 1)
|
793
|
+
# # ┌──────┐
|
794
|
+
# # │ foo │
|
795
|
+
# # │ --- │
|
796
|
+
# # │ i64 │
|
797
|
+
# # ╞══════╡
|
798
|
+
# # │ -123 │
|
799
|
+
# # │ -456 │
|
800
|
+
# # └──────┘
|
801
|
+
#
|
802
|
+
# @example
|
803
|
+
# df.select(~Polars.cs.signed_integer)
|
804
|
+
# # =>
|
805
|
+
# # shape: (2, 3)
|
806
|
+
# # ┌──────┬──────┬─────┐
|
807
|
+
# # │ bar ┆ baz ┆ zap │
|
808
|
+
# # │ --- ┆ --- ┆ --- │
|
809
|
+
# # │ u32 ┆ u64 ┆ str │
|
810
|
+
# # ╞══════╪══════╪═════╡
|
811
|
+
# # │ 3456 ┆ 7654 ┆ ab │
|
812
|
+
# # │ 6789 ┆ 4321 ┆ cd │
|
813
|
+
# # └──────┴──────┴─────┘
|
814
|
+
#
|
815
|
+
# @example Select all integer columns (both signed and unsigned):
|
816
|
+
# df.select(Polars.cs.integer)
|
817
|
+
# # =>
|
818
|
+
# # shape: (2, 3)
|
819
|
+
# # ┌──────┬──────┬──────┐
|
820
|
+
# # │ foo ┆ bar ┆ baz │
|
821
|
+
# # │ --- ┆ --- ┆ --- │
|
822
|
+
# # │ i64 ┆ u32 ┆ u64 │
|
823
|
+
# # ╞══════╪══════╪══════╡
|
824
|
+
# # │ -123 ┆ 3456 ┆ 7654 │
|
825
|
+
# # │ -456 ┆ 6789 ┆ 4321 │
|
826
|
+
# # └──────┴──────┴──────┘
|
827
|
+
def self.signed_integer
|
828
|
+
_selector_proxy_(F.col(SIGNED_INTEGER_DTYPES), name: "signed_integer")
|
829
|
+
end
|
830
|
+
|
831
|
+
# Select all unsigned integer columns.
|
832
|
+
#
|
833
|
+
# @return [SelectorProxy]
|
834
|
+
#
|
835
|
+
# @example
|
836
|
+
# df = Polars::DataFrame.new(
|
837
|
+
# {
|
838
|
+
# "foo" => [-123, -456],
|
839
|
+
# "bar" => [3456, 6789],
|
840
|
+
# "baz" => [7654, 4321],
|
841
|
+
# "zap" => ["ab", "cd"]
|
842
|
+
# },
|
843
|
+
# schema_overrides: {"bar" => Polars::UInt32, "baz" => Polars::UInt64}
|
844
|
+
# )
|
845
|
+
#
|
846
|
+
# @example Select all unsigned integer columns:
|
847
|
+
# df.select(Polars.cs.unsigned_integer)
|
848
|
+
# # =>
|
849
|
+
# # shape: (2, 2)
|
850
|
+
# # ┌──────┬──────┐
|
851
|
+
# # │ bar ┆ baz │
|
852
|
+
# # │ --- ┆ --- │
|
853
|
+
# # │ u32 ┆ u64 │
|
854
|
+
# # ╞══════╪══════╡
|
855
|
+
# # │ 3456 ┆ 7654 │
|
856
|
+
# # │ 6789 ┆ 4321 │
|
857
|
+
# # └──────┴──────┘
|
858
|
+
#
|
859
|
+
# @example Select all columns *except* for those that are unsigned integers:
|
860
|
+
# df.select(~Polars.cs.unsigned_integer)
|
861
|
+
# # =>
|
862
|
+
# # shape: (2, 2)
|
863
|
+
# # ┌──────┬─────┐
|
864
|
+
# # │ foo ┆ zap │
|
865
|
+
# # │ --- ┆ --- │
|
866
|
+
# # │ i64 ┆ str │
|
867
|
+
# # ╞══════╪═════╡
|
868
|
+
# # │ -123 ┆ ab │
|
869
|
+
# # │ -456 ┆ cd │
|
870
|
+
# # └──────┴─────┘
|
871
|
+
#
|
872
|
+
# @example Select all integer columns (both signed and unsigned):
|
873
|
+
# df.select(Polars.cs.integer)
|
874
|
+
# # =>
|
875
|
+
# # shape: (2, 3)
|
876
|
+
# # ┌──────┬──────┬──────┐
|
877
|
+
# # │ foo ┆ bar ┆ baz │
|
878
|
+
# # │ --- ┆ --- ┆ --- │
|
879
|
+
# # │ i64 ┆ u32 ┆ u64 │
|
880
|
+
# # ╞══════╪══════╪══════╡
|
881
|
+
# # │ -123 ┆ 3456 ┆ 7654 │
|
882
|
+
# # │ -456 ┆ 6789 ┆ 4321 │
|
883
|
+
# # └──────┴──────┴──────┘
|
884
|
+
def self.unsigned_integer
|
885
|
+
_selector_proxy_(F.col(UNSIGNED_INTEGER_DTYPES), name: "unsigned_integer")
|
886
|
+
end
|
887
|
+
|
888
|
+
# Select the last column in the current scope.
|
889
|
+
#
|
890
|
+
# @return [SelectorProxy]
|
891
|
+
#
|
892
|
+
# @example
|
893
|
+
# df = Polars::DataFrame.new(
|
894
|
+
# {
|
895
|
+
# "foo" => ["x", "y"],
|
896
|
+
# "bar" => [123, 456],
|
897
|
+
# "baz" => [2.0, 5.5],
|
898
|
+
# "zap" => [0, 1]
|
899
|
+
# }
|
900
|
+
# )
|
901
|
+
#
|
902
|
+
# @example Select the last column:
|
903
|
+
# df.select(Polars.cs.last)
|
904
|
+
# # =>
|
905
|
+
# # shape: (2, 1)
|
906
|
+
# # ┌─────┐
|
907
|
+
# # │ zap │
|
908
|
+
# # │ --- │
|
909
|
+
# # │ i64 │
|
910
|
+
# # ╞═════╡
|
911
|
+
# # │ 0 │
|
912
|
+
# # │ 1 │
|
913
|
+
# # └─────┘
|
914
|
+
#
|
915
|
+
# @example Select everything *except* for the last column:
|
916
|
+
# df.select(~Polars.cs.last)
|
917
|
+
# # =>
|
918
|
+
# # shape: (2, 3)
|
919
|
+
# # ┌─────┬─────┬─────┐
|
920
|
+
# # │ foo ┆ bar ┆ baz │
|
921
|
+
# # │ --- ┆ --- ┆ --- │
|
922
|
+
# # │ str ┆ i64 ┆ f64 │
|
923
|
+
# # ╞═════╪═════╪═════╡
|
924
|
+
# # │ x ┆ 123 ┆ 2.0 │
|
925
|
+
# # │ y ┆ 456 ┆ 5.5 │
|
926
|
+
# # └─────┴─────┴─────┘
|
927
|
+
def self.last
|
928
|
+
_selector_proxy_(F.last, name: "last")
|
929
|
+
end
|
930
|
+
|
931
|
+
# Select all numeric columns.
|
932
|
+
#
|
933
|
+
# @return [SelectorProxy]
|
934
|
+
#
|
935
|
+
# @example
|
936
|
+
# df = Polars::DataFrame.new(
|
937
|
+
# {
|
938
|
+
# "foo" => ["x", "y"],
|
939
|
+
# "bar" => [123, 456],
|
940
|
+
# "baz" => [2.0, 5.5],
|
941
|
+
# "zap" => [0, 0]
|
942
|
+
# },
|
943
|
+
# schema_overrides: {"bar" => Polars::Int16, "baz" => Polars::Float32, "zap" => Polars::UInt8},
|
944
|
+
# )
|
945
|
+
#
|
946
|
+
# @example Match all numeric columns:
|
947
|
+
# df.select(Polars.cs.numeric)
|
948
|
+
# # =>
|
949
|
+
# # shape: (2, 3)
|
950
|
+
# # ┌─────┬─────┬─────┐
|
951
|
+
# # │ bar ┆ baz ┆ zap │
|
952
|
+
# # │ --- ┆ --- ┆ --- │
|
953
|
+
# # │ i16 ┆ f32 ┆ u8 │
|
954
|
+
# # ╞═════╪═════╪═════╡
|
955
|
+
# # │ 123 ┆ 2.0 ┆ 0 │
|
956
|
+
# # │ 456 ┆ 5.5 ┆ 0 │
|
957
|
+
# # └─────┴─────┴─────┘
|
958
|
+
#
|
959
|
+
# @example Match all columns *except* for those that are numeric:
|
960
|
+
# df.select(~Polars.cs.numeric)
|
961
|
+
# # =>
|
962
|
+
# # shape: (2, 1)
|
963
|
+
# # ┌─────┐
|
964
|
+
# # │ foo │
|
965
|
+
# # │ --- │
|
966
|
+
# # │ str │
|
967
|
+
# # ╞═════╡
|
968
|
+
# # │ x │
|
969
|
+
# # │ y │
|
970
|
+
# # └─────┘
|
971
|
+
def self.numeric
|
972
|
+
_selector_proxy_(F.col(NUMERIC_DTYPES), name: "numeric")
|
973
|
+
end
|
974
|
+
|
975
|
+
# Select columns that start with the given substring(s).
|
976
|
+
#
|
977
|
+
# @param prefix [Object]
|
978
|
+
# Substring(s) that matching column names should start with.
|
979
|
+
#
|
980
|
+
# @return [SelectorProxy]
|
981
|
+
#
|
982
|
+
# @example
|
983
|
+
# df = Polars::DataFrame.new(
|
984
|
+
# {
|
985
|
+
# "foo" => [1.0, 2.0],
|
986
|
+
# "bar" => [3.0, 4.0],
|
987
|
+
# "baz" => [5, 6],
|
988
|
+
# "zap" => [7, 8]
|
989
|
+
# }
|
990
|
+
# )
|
991
|
+
#
|
992
|
+
# @example Match columns starting with a 'b':
|
993
|
+
# df.select(Polars.cs.starts_with("b"))
|
994
|
+
# # =>
|
995
|
+
# # shape: (2, 2)
|
996
|
+
# # ┌─────┬─────┐
|
997
|
+
# # │ bar ┆ baz │
|
998
|
+
# # │ --- ┆ --- │
|
999
|
+
# # │ f64 ┆ i64 │
|
1000
|
+
# # ╞═════╪═════╡
|
1001
|
+
# # │ 3.0 ┆ 5 │
|
1002
|
+
# # │ 4.0 ┆ 6 │
|
1003
|
+
# # └─────┴─────┘
|
1004
|
+
#
|
1005
|
+
# @example Match columns starting with *either* the letter 'b' or 'z':
|
1006
|
+
# df.select(Polars.cs.starts_with("b", "z"))
|
1007
|
+
# # =>
|
1008
|
+
# # shape: (2, 3)
|
1009
|
+
# # ┌─────┬─────┬─────┐
|
1010
|
+
# # │ bar ┆ baz ┆ zap │
|
1011
|
+
# # │ --- ┆ --- ┆ --- │
|
1012
|
+
# # │ f64 ┆ i64 ┆ i64 │
|
1013
|
+
# # ╞═════╪═════╪═════╡
|
1014
|
+
# # │ 3.0 ┆ 5 ┆ 7 │
|
1015
|
+
# # │ 4.0 ┆ 6 ┆ 8 │
|
1016
|
+
# # └─────┴─────┴─────┘
|
1017
|
+
#
|
1018
|
+
# @example Match all columns *except* for those starting with 'b':
|
1019
|
+
# df.select(~Polars.cs.starts_with("b"))
|
1020
|
+
# # =>
|
1021
|
+
# # shape: (2, 2)
|
1022
|
+
# # ┌─────┬─────┐
|
1023
|
+
# # │ foo ┆ zap │
|
1024
|
+
# # │ --- ┆ --- │
|
1025
|
+
# # │ f64 ┆ i64 │
|
1026
|
+
# # ╞═════╪═════╡
|
1027
|
+
# # │ 1.0 ┆ 7 │
|
1028
|
+
# # │ 2.0 ┆ 8 │
|
1029
|
+
# # └─────┴─────┘
|
1030
|
+
def self.starts_with(*prefix)
|
1031
|
+
escaped_prefix = _re_string(prefix)
|
1032
|
+
raw_params = "^#{escaped_prefix}.*$"
|
1033
|
+
|
1034
|
+
_selector_proxy_(
|
1035
|
+
F.col(raw_params),
|
1036
|
+
name: "starts_with",
|
1037
|
+
parameters: {"*prefix" => prefix}
|
1038
|
+
)
|
1039
|
+
end
|
1040
|
+
|
1041
|
+
# Select all String (and, optionally, Categorical) string columns.
|
1042
|
+
#
|
1043
|
+
# @return [SelectorProxy]
|
1044
|
+
#
|
1045
|
+
# @example
|
1046
|
+
# df = Polars::DataFrame.new(
|
1047
|
+
# {
|
1048
|
+
# "w" => ["xx", "yy", "xx", "yy", "xx"],
|
1049
|
+
# "x" => [1, 2, 1, 4, -2],
|
1050
|
+
# "y" => [3.0, 4.5, 1.0, 2.5, -2.0],
|
1051
|
+
# "z" => ["a", "b", "a", "b", "b"]
|
1052
|
+
# },
|
1053
|
+
# ).with_columns(
|
1054
|
+
# z: Polars.col("z").cast(Polars::Categorical.new("lexical")),
|
1055
|
+
# )
|
1056
|
+
#
|
1057
|
+
# @example Group by all string columns, sum the numeric columns, then sort by the string cols:
|
1058
|
+
# >>> df.group_by(Polars.cs.string).agg(Polars.cs.numeric.sum).sort(Polars.cs.string)
|
1059
|
+
# shape: (2, 3)
|
1060
|
+
# ┌─────┬─────┬─────┐
|
1061
|
+
# │ w ┆ x ┆ y │
|
1062
|
+
# │ --- ┆ --- ┆ --- │
|
1063
|
+
# │ str ┆ i64 ┆ f64 │
|
1064
|
+
# ╞═════╪═════╪═════╡
|
1065
|
+
# │ xx ┆ 0 ┆ 2.0 │
|
1066
|
+
# │ yy ┆ 6 ┆ 7.0 │
|
1067
|
+
# └─────┴─────┴─────┘
|
1068
|
+
#
|
1069
|
+
# @example Group by all string *and* categorical columns:
|
1070
|
+
# df.group_by(Polars.cs.string(include_categorical: true)).agg(Polars.cs.numeric.sum).sort(
|
1071
|
+
# Polars.cs.string(include_categorical: true)
|
1072
|
+
# )
|
1073
|
+
# # =>
|
1074
|
+
# # shape: (3, 4)
|
1075
|
+
# # ┌─────┬─────┬─────┬──────┐
|
1076
|
+
# # │ w ┆ z ┆ x ┆ y │
|
1077
|
+
# # │ --- ┆ --- ┆ --- ┆ --- │
|
1078
|
+
# # │ str ┆ cat ┆ i64 ┆ f64 │
|
1079
|
+
# # ╞═════╪═════╪═════╪══════╡
|
1080
|
+
# # │ xx ┆ a ┆ 2 ┆ 4.0 │
|
1081
|
+
# # │ xx ┆ b ┆ -2 ┆ -2.0 │
|
1082
|
+
# # │ yy ┆ b ┆ 6 ┆ 7.0 │
|
1083
|
+
# # └─────┴─────┴─────┴──────┘
|
1084
|
+
def self.string(include_categorical: false)
|
1085
|
+
string_dtypes = [String]
|
1086
|
+
if include_categorical
|
1087
|
+
string_dtypes << Categorical
|
1088
|
+
end
|
1089
|
+
|
1090
|
+
_selector_proxy_(
|
1091
|
+
F.col(string_dtypes),
|
1092
|
+
name: "string",
|
1093
|
+
parameters: {"include_categorical" => include_categorical},
|
1094
|
+
)
|
1095
|
+
end
|
1096
|
+
|
1097
|
+
# Select all time columns.
|
1098
|
+
#
|
1099
|
+
# @return [SelectorProxy]
|
1100
|
+
#
|
1101
|
+
# @example
|
1102
|
+
# df = Polars::DataFrame.new(
|
1103
|
+
# {
|
1104
|
+
# "dtm" => [DateTime.new(2001, 5, 7, 10, 25), DateTime.new(2031, 12, 31, 0, 30)],
|
1105
|
+
# "dt" => [Date.new(1999, 12, 31), Date.new(2024, 8, 9)],
|
1106
|
+
# "tm" => [Time.utc(2001, 1, 1, 0, 0, 0), Time.utc(2001, 1, 1, 23, 59, 59)]
|
1107
|
+
# },
|
1108
|
+
# schema_overrides: {"tm" => Polars::Time}
|
1109
|
+
# )
|
1110
|
+
#
|
1111
|
+
# @example Select all time columns:
|
1112
|
+
# df.select(Polars.cs.time)
|
1113
|
+
# # =>
|
1114
|
+
# # shape: (2, 1)
|
1115
|
+
# # ┌──────────┐
|
1116
|
+
# # │ tm │
|
1117
|
+
# # │ --- │
|
1118
|
+
# # │ time │
|
1119
|
+
# # ╞══════════╡
|
1120
|
+
# # │ 00:00:00 │
|
1121
|
+
# # │ 23:59:59 │
|
1122
|
+
# # └──────────┘
|
1123
|
+
#
|
1124
|
+
# @example Select all columns *except* for those that are times:
|
1125
|
+
# df.select(~Polars.cs.time)
|
1126
|
+
# # =>
|
1127
|
+
# # shape: (2, 2)
|
1128
|
+
# # ┌─────────────────────┬────────────┐
|
1129
|
+
# # │ dtm ┆ dt │
|
1130
|
+
# # │ --- ┆ --- │
|
1131
|
+
# # │ datetime[ns] ┆ date │
|
1132
|
+
# # ╞═════════════════════╪════════════╡
|
1133
|
+
# # │ 2001-05-07 10:25:00 ┆ 1999-12-31 │
|
1134
|
+
# # │ 2031-12-31 00:30:00 ┆ 2024-08-09 │
|
1135
|
+
# # └─────────────────────┴────────────┘
|
1136
|
+
def self.time
|
1137
|
+
_selector_proxy_(F.col(Time), name: "time")
|
1138
|
+
end
|
1139
|
+
end
|
1140
|
+
|
1141
|
+
def self.cs
|
1142
|
+
Polars::Selectors
|
1143
|
+
end
|
1144
|
+
end
|