red-arrow 0.15.1 → 0.16.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/ext/arrow/converters.hpp +6 -6
- data/lib/arrow/array-builder.rb +101 -52
- data/lib/arrow/array.rb +28 -10
- data/lib/arrow/chunked-array.rb +2 -0
- data/lib/arrow/csv-loader.rb +5 -0
- data/lib/arrow/csv-read-options.rb +18 -0
- data/lib/arrow/data-type.rb +35 -2
- data/lib/arrow/decimal128-array-builder.rb +0 -2
- data/lib/arrow/field.rb +1 -1
- data/lib/arrow/generic-filterable.rb +43 -0
- data/lib/arrow/generic-takeable.rb +38 -0
- data/lib/arrow/list-data-type.rb +58 -8
- data/lib/arrow/loader.rb +9 -1
- data/lib/arrow/{binary-array-builder.rb → null-array.rb} +3 -15
- data/lib/arrow/record-batch.rb +0 -3
- data/lib/arrow/schema.rb +0 -2
- data/lib/arrow/struct-data-type.rb +0 -2
- data/lib/arrow/table-loader.rb +29 -6
- data/lib/arrow/table-saver.rb +29 -9
- data/lib/arrow/table.rb +14 -50
- data/lib/arrow/version.rb +1 -1
- data/red-arrow.gemspec +3 -1
- data/test/test-array-builder.rb +17 -0
- data/test/test-array.rb +102 -0
- data/test/test-chunked-array.rb +94 -0
- data/test/test-csv-loader.rb +2 -2
- data/test/test-data-type.rb +11 -0
- data/test/test-list-data-type.rb +27 -1
- data/test/test-null-array.rb +23 -0
- data/test/test-slicer.rb +74 -30
- data/test/test-table.rb +147 -14
- data/test/test-timestamp-array.rb +19 -0
- metadata +60 -55
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 64b14ef4120f4ab290e8161020902ec2a22631c519d5a133a63ce383610e8545
|
4
|
+
data.tar.gz: 2f5850520e2dc69568a454cee0d4246909d52f1d49851221b1b9efd3149bc15c
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 0e19a4da6182437a51f9dad6212436e70b00881674ee6cd29e2a40910fe711fdef4076c81d2778f3ff9e9dd3f45b573c43e90378244cf08d7550b504ad8b53af
|
7
|
+
data.tar.gz: 547eb8b31fd59d9c1d5fc1163bb25da70154b797d103d3a01f6fda70dddc0b3c2cdb5391b9006c4510be37b8b75988195f809aaead8c91e7fb68f762cc5313de
|
data/ext/arrow/converters.hpp
CHANGED
@@ -504,14 +504,14 @@ namespace red_arrow {
|
|
504
504
|
uint8_t compute_child_index(const arrow::UnionArray& array,
|
505
505
|
arrow::UnionType* type,
|
506
506
|
const char* tag) {
|
507
|
-
const auto
|
508
|
-
|
509
|
-
|
510
|
-
if (
|
511
|
-
return
|
507
|
+
const auto type_code = array.raw_type_codes()[index_];
|
508
|
+
if (type_code >= 0 && type_code <= arrow::UnionType::kMaxTypeCode) {
|
509
|
+
const auto child_id = type->child_ids()[type_code];
|
510
|
+
if (child_id >= 0) {
|
511
|
+
return child_id;
|
512
512
|
}
|
513
513
|
}
|
514
|
-
check_status(arrow::Status::Invalid("Unknown type ID: ",
|
514
|
+
check_status(arrow::Status::Invalid("Unknown type ID: ", type_code),
|
515
515
|
tag);
|
516
516
|
return 0;
|
517
517
|
}
|
data/lib/arrow/array-builder.rb
CHANGED
@@ -26,60 +26,13 @@ module Arrow
|
|
26
26
|
return builder.build(values)
|
27
27
|
end
|
28
28
|
|
29
|
-
|
30
|
-
builder_class_arguments = []
|
29
|
+
builder_info = nil
|
31
30
|
values.each do |value|
|
32
|
-
|
33
|
-
|
34
|
-
# Ignore
|
35
|
-
when true, false
|
36
|
-
return BooleanArray.new(values)
|
37
|
-
when String
|
38
|
-
return StringArray.new(values)
|
39
|
-
when Float
|
40
|
-
return DoubleArray.new(values)
|
41
|
-
when Integer
|
42
|
-
if value < 0
|
43
|
-
builder = IntArrayBuilder.new
|
44
|
-
return builder.build(values)
|
45
|
-
else
|
46
|
-
builder_class = UIntArrayBuilder
|
47
|
-
builder_class_arguments = []
|
48
|
-
end
|
49
|
-
when Time
|
50
|
-
data_type = value.data_type
|
51
|
-
case data_type.unit
|
52
|
-
when TimeUnit::SECOND
|
53
|
-
if builder.nil?
|
54
|
-
builder = Time32ArrayBuilder
|
55
|
-
builder_class_arguments = [data_type]
|
56
|
-
end
|
57
|
-
when TimeUnit::MILLI
|
58
|
-
if builder != Time64ArrayBuilder
|
59
|
-
builder = Time32ArrayBuilder
|
60
|
-
builder_class_arguments = [data_type]
|
61
|
-
end
|
62
|
-
when TimeUnit::MICRO
|
63
|
-
builder = Time64ArrayBuilder
|
64
|
-
builder_class_arguments = [data_type]
|
65
|
-
when TimeUnit::NANO
|
66
|
-
builder = Time64ArrayBuilder.new(data_type)
|
67
|
-
return builder.build(values)
|
68
|
-
end
|
69
|
-
when ::Time
|
70
|
-
data_type = TimestampDataType.new(:nano)
|
71
|
-
builder = TimestampArrayBuilder.new(data_type)
|
72
|
-
return builder.build(values)
|
73
|
-
when DateTime
|
74
|
-
return Date64Array.new(values)
|
75
|
-
when Date
|
76
|
-
return Date32Array.new(values)
|
77
|
-
else
|
78
|
-
return StringArray.new(values)
|
79
|
-
end
|
31
|
+
builder_info = detect_builder_info(value, builder_info)
|
32
|
+
break if builder_info and builder_info[:detected]
|
80
33
|
end
|
81
|
-
if
|
82
|
-
builder =
|
34
|
+
if builder_info
|
35
|
+
builder = builder_info[:builder]
|
83
36
|
builder.build(values)
|
84
37
|
else
|
85
38
|
Arrow::StringArray.new(values)
|
@@ -89,6 +42,102 @@ module Arrow
|
|
89
42
|
def buildable?(args)
|
90
43
|
args.size == method(:build).arity
|
91
44
|
end
|
45
|
+
|
46
|
+
private
|
47
|
+
def detect_builder_info(value, builder_info)
|
48
|
+
case value
|
49
|
+
when nil
|
50
|
+
builder_info
|
51
|
+
when true, false
|
52
|
+
{
|
53
|
+
builder: BooleanArrayBuilder.new,
|
54
|
+
detected: true,
|
55
|
+
}
|
56
|
+
when String
|
57
|
+
{
|
58
|
+
builder: StringArrayBuilder.new,
|
59
|
+
detected: true,
|
60
|
+
}
|
61
|
+
when Float
|
62
|
+
{
|
63
|
+
builder: DoubleArrayBuilder.new,
|
64
|
+
detected: true,
|
65
|
+
}
|
66
|
+
when Integer
|
67
|
+
if value < 0
|
68
|
+
{
|
69
|
+
builder: IntArrayBuilder.new,
|
70
|
+
detected: true,
|
71
|
+
}
|
72
|
+
else
|
73
|
+
{
|
74
|
+
builder: UIntArrayBuilder.new,
|
75
|
+
}
|
76
|
+
end
|
77
|
+
when Time
|
78
|
+
data_type = value.data_type
|
79
|
+
case data_type.unit
|
80
|
+
when TimeUnit::SECOND
|
81
|
+
builder_info || {
|
82
|
+
builder: Time32ArrayBuilder.new(data_type)
|
83
|
+
}
|
84
|
+
when TimeUnit::MILLI
|
85
|
+
if builder_info and builder_info[:builder].is_a?(Time64ArrayBuilder)
|
86
|
+
builder_info
|
87
|
+
else
|
88
|
+
{
|
89
|
+
builder: Time32ArrayBuilder.new(data_type),
|
90
|
+
}
|
91
|
+
end
|
92
|
+
when TimeUnit::MICRO
|
93
|
+
{
|
94
|
+
builder: Time64ArrayBuilder.new(data_type),
|
95
|
+
}
|
96
|
+
when TimeUnit::NANO
|
97
|
+
{
|
98
|
+
builder: Time64ArrayBuilder.new(data_type),
|
99
|
+
detected: true
|
100
|
+
}
|
101
|
+
end
|
102
|
+
when ::Time
|
103
|
+
data_type = TimestampDataType.new(:nano)
|
104
|
+
{
|
105
|
+
builder: TimestampArrayBuilder.new(data_type),
|
106
|
+
detected: true,
|
107
|
+
}
|
108
|
+
when DateTime
|
109
|
+
{
|
110
|
+
builder: Date64ArrayBuilder.new,
|
111
|
+
detected: true,
|
112
|
+
}
|
113
|
+
when Date
|
114
|
+
{
|
115
|
+
builder: Date32ArrayBuilder.new,
|
116
|
+
detected: true,
|
117
|
+
}
|
118
|
+
when ::Array
|
119
|
+
sub_builder_info = nil
|
120
|
+
value.each do |sub_value|
|
121
|
+
sub_builder_info = detect_builder_info(sub_value, sub_builder_info)
|
122
|
+
break if sub_builder_info and sub_builder_info[:detected]
|
123
|
+
end
|
124
|
+
if sub_builder_info and sub_builder_info[:detected]
|
125
|
+
sub_value_data_type = sub_builder_info[:builder].value_data_type
|
126
|
+
field = Field.new("item", sub_value_data_type)
|
127
|
+
{
|
128
|
+
builder: ListArrayBuilder.new(ListDataType.new(field)),
|
129
|
+
detected: true,
|
130
|
+
}
|
131
|
+
else
|
132
|
+
builder_info
|
133
|
+
end
|
134
|
+
else
|
135
|
+
{
|
136
|
+
builder: StringArrayBuilder.new,
|
137
|
+
detected: true,
|
138
|
+
}
|
139
|
+
end
|
140
|
+
end
|
92
141
|
end
|
93
142
|
|
94
143
|
def build(values)
|
data/lib/arrow/array.rb
CHANGED
@@ -18,20 +18,21 @@
|
|
18
18
|
module Arrow
|
19
19
|
class Array
|
20
20
|
include Enumerable
|
21
|
+
include GenericFilterable
|
22
|
+
include GenericTakeable
|
21
23
|
|
22
24
|
class << self
|
23
25
|
def new(*args)
|
26
|
+
_builder_class = builder_class
|
27
|
+
return super if _builder_class.nil?
|
28
|
+
return super unless _builder_class.buildable?(args)
|
29
|
+
_builder_class.build(*args)
|
30
|
+
end
|
31
|
+
|
32
|
+
def builder_class
|
24
33
|
builder_class_name = "#{name}Builder"
|
25
|
-
|
26
|
-
|
27
|
-
if builder_class.buildable?(args)
|
28
|
-
builder_class.build(*args)
|
29
|
-
else
|
30
|
-
super
|
31
|
-
end
|
32
|
-
else
|
33
|
-
super
|
34
|
-
end
|
34
|
+
return nil unless const_defined?(builder_class_name)
|
35
|
+
const_get(builder_class_name)
|
35
36
|
end
|
36
37
|
end
|
37
38
|
|
@@ -82,5 +83,22 @@ module Arrow
|
|
82
83
|
def to_a
|
83
84
|
values
|
84
85
|
end
|
86
|
+
|
87
|
+
alias_method :is_in_raw, :is_in
|
88
|
+
def is_in(values)
|
89
|
+
case values
|
90
|
+
when ::Array
|
91
|
+
if self.class.builder_class.buildable?([values])
|
92
|
+
values = self.class.new(values)
|
93
|
+
else
|
94
|
+
values = self.class.new(value_data_type, values)
|
95
|
+
end
|
96
|
+
is_in_raw(values)
|
97
|
+
when ChunkedArray
|
98
|
+
is_in_chunked_array(values)
|
99
|
+
else
|
100
|
+
is_in_raw(values)
|
101
|
+
end
|
102
|
+
end
|
85
103
|
end
|
86
104
|
end
|
data/lib/arrow/chunked-array.rb
CHANGED
data/lib/arrow/csv-loader.rb
CHANGED
@@ -30,6 +30,9 @@ module Arrow
|
|
30
30
|
def initialize(path_or_data, **options)
|
31
31
|
@path_or_data = path_or_data
|
32
32
|
@options = options
|
33
|
+
if @options.key?(:delimiter)
|
34
|
+
@options[:col_sep] = @options.delete(:delimiter)
|
35
|
+
end
|
33
36
|
@compression = @options.delete(:compression)
|
34
37
|
end
|
35
38
|
|
@@ -113,6 +116,8 @@ module Arrow
|
|
113
116
|
options.add_schema(value)
|
114
117
|
when :encoding
|
115
118
|
# process encoding on opening input
|
119
|
+
when :col_sep
|
120
|
+
options.delimiter = value
|
116
121
|
else
|
117
122
|
setter = "#{key}="
|
118
123
|
if options.respond_to?(setter)
|
@@ -21,5 +21,23 @@ module Arrow
|
|
21
21
|
def add_column_type(name, type)
|
22
22
|
add_column_type_raw(name, DataType.resolve(type))
|
23
23
|
end
|
24
|
+
|
25
|
+
alias_method :delimiter_raw, :delimiter
|
26
|
+
def delimiter
|
27
|
+
delimiter_raw.chr
|
28
|
+
end
|
29
|
+
|
30
|
+
alias_method :delimiter_raw=, :delimiter=
|
31
|
+
def delimiter=(delimiter)
|
32
|
+
case delimiter
|
33
|
+
when String
|
34
|
+
if delimiter.bytesize != 1
|
35
|
+
message = "delimiter must be 1 byte character: #{delimiter.inspect}"
|
36
|
+
raise ArgumentError, message
|
37
|
+
end
|
38
|
+
delimiter = delimiter.ord
|
39
|
+
end
|
40
|
+
self.delimiter_raw = delimiter
|
41
|
+
end
|
24
42
|
end
|
25
43
|
end
|
data/lib/arrow/data-type.rb
CHANGED
@@ -121,6 +121,26 @@ module Arrow
|
|
121
121
|
end
|
122
122
|
end
|
123
123
|
|
124
|
+
def sub_types
|
125
|
+
types = {}
|
126
|
+
gtype.children.each do |child|
|
127
|
+
sub_type = child.to_class
|
128
|
+
types[sub_type] = true
|
129
|
+
sub_type.sub_types.each do |sub_sub_type|
|
130
|
+
types[sub_sub_type] = true
|
131
|
+
end
|
132
|
+
end
|
133
|
+
types.keys
|
134
|
+
end
|
135
|
+
|
136
|
+
def try_convert(value)
|
137
|
+
begin
|
138
|
+
resolve(value)
|
139
|
+
rescue ArgumentError
|
140
|
+
nil
|
141
|
+
end
|
142
|
+
end
|
143
|
+
|
124
144
|
private
|
125
145
|
def resolve_class(data_type)
|
126
146
|
components = data_type.to_s.split("_").collect(&:capitalize)
|
@@ -137,11 +157,24 @@ module Arrow
|
|
137
157
|
available_types << components.collect(&:downcase).join("_").to_sym
|
138
158
|
end
|
139
159
|
message =
|
140
|
-
"unknown type:
|
160
|
+
"unknown type: <#{data_type.inspect}>: " +
|
141
161
|
"available types: #{available_types.inspect}"
|
142
162
|
raise ArgumentError, message
|
143
163
|
end
|
144
|
-
Arrow.const_get(data_type_class_name)
|
164
|
+
data_type_class = Arrow.const_get(data_type_class_name)
|
165
|
+
if data_type_class.gtype.abstract?
|
166
|
+
not_abstract_types = data_type_class.sub_types.find_all do |sub_type|
|
167
|
+
not sub_type.gtype.abstract?
|
168
|
+
end
|
169
|
+
not_abstract_types = not_abstract_types.sort_by do |type|
|
170
|
+
type.name
|
171
|
+
end
|
172
|
+
message =
|
173
|
+
"abstract type: <#{data_type.inspect}>: " +
|
174
|
+
"use one of not abstract type: #{not_abstract_types.inspect}"
|
175
|
+
raise ArgumentError, message
|
176
|
+
end
|
177
|
+
data_type_class
|
145
178
|
end
|
146
179
|
end
|
147
180
|
|
data/lib/arrow/field.rb
CHANGED
@@ -59,7 +59,7 @@ module Arrow
|
|
59
59
|
# There is a shortcut for convenience. If field description
|
60
60
|
# doesn't have `:data_type`, all keys except `:name` are
|
61
61
|
# processes as data type description. For example, the
|
62
|
-
# following field
|
62
|
+
# following field descriptions are the same:
|
63
63
|
#
|
64
64
|
# ```ruby
|
65
65
|
# {name: "visible", data_type: {type: :boolean}}
|
@@ -0,0 +1,43 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one
|
2
|
+
# or more contributor license agreements. See the NOTICE file
|
3
|
+
# distributed with this work for additional information
|
4
|
+
# regarding copyright ownership. The ASF licenses this file
|
5
|
+
# to you under the Apache License, Version 2.0 (the
|
6
|
+
# "License"); you may not use this file except in compliance
|
7
|
+
# with the License. You may obtain a copy of the License at
|
8
|
+
#
|
9
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
#
|
11
|
+
# Unless required by applicable law or agreed to in writing,
|
12
|
+
# software distributed under the License is distributed on an
|
13
|
+
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
|
14
|
+
# KIND, either express or implied. See the License for the
|
15
|
+
# specific language governing permissions and limitations
|
16
|
+
# under the License.
|
17
|
+
|
18
|
+
module Arrow
|
19
|
+
module GenericFilterable
|
20
|
+
class << self
|
21
|
+
def included(base)
|
22
|
+
base.alias_method :filter_raw, :filter
|
23
|
+
base.alias_method :filter, :filter_generic
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
def filter_generic(filter)
|
28
|
+
case filter
|
29
|
+
when ::Array
|
30
|
+
filter_raw(BooleanArray.new(filter))
|
31
|
+
when ChunkedArray
|
32
|
+
if respond_to?(:filter_chunked_array)
|
33
|
+
filter_chunked_array(filter)
|
34
|
+
else
|
35
|
+
# TODO: Implement this in C++
|
36
|
+
filter_raw(filter.pack)
|
37
|
+
end
|
38
|
+
else
|
39
|
+
filter_raw(filter)
|
40
|
+
end
|
41
|
+
end
|
42
|
+
end
|
43
|
+
end
|
@@ -0,0 +1,38 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one
|
2
|
+
# or more contributor license agreements. See the NOTICE file
|
3
|
+
# distributed with this work for additional information
|
4
|
+
# regarding copyright ownership. The ASF licenses this file
|
5
|
+
# to you under the Apache License, Version 2.0 (the
|
6
|
+
# "License"); you may not use this file except in compliance
|
7
|
+
# with the License. You may obtain a copy of the License at
|
8
|
+
#
|
9
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
#
|
11
|
+
# Unless required by applicable law or agreed to in writing,
|
12
|
+
# software distributed under the License is distributed on an
|
13
|
+
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
|
14
|
+
# KIND, either express or implied. See the License for the
|
15
|
+
# specific language governing permissions and limitations
|
16
|
+
# under the License.
|
17
|
+
|
18
|
+
module Arrow
|
19
|
+
module GenericTakeable
|
20
|
+
class << self
|
21
|
+
def included(base)
|
22
|
+
base.alias_method :take_raw, :take
|
23
|
+
base.alias_method :take, :take_generic
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
def take_generic(indices)
|
28
|
+
case indices
|
29
|
+
when ::Array
|
30
|
+
take_raw(IntArrayBuilder.build(indices))
|
31
|
+
when ChunkedArray
|
32
|
+
take_chunked_array(indices)
|
33
|
+
else
|
34
|
+
take_raw(indices)
|
35
|
+
end
|
36
|
+
end
|
37
|
+
end
|
38
|
+
end
|