visual_width 0.0.1
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 +7 -0
- data/.gitignore +17 -0
- data/.rspec +2 -0
- data/Changes +4 -0
- data/Gemfile +8 -0
- data/LICENSE.txt +22 -0
- data/README.md +42 -0
- data/Rakefile +14 -0
- data/lib/visual_width/data.rb +204 -0
- data/lib/visual_width/string_ext.rb +11 -0
- data/lib/visual_width/string_refine.rb +13 -0
- data/lib/visual_width/version.rb +3 -0
- data/lib/visual_width.rb +53 -0
- data/spec/visual_width/string_refine_spec.rb +19 -0
- data/spec/visual_width_spec.rb +90 -0
- data/tool/benchmark.rb +21 -0
- data/tool/east-asian-width.pl +113 -0
- data/visual_width.gemspec +23 -0
- metadata +92 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 05cf7a6f8d1cb732fa2f35fabbe3a495c727e4d3
|
4
|
+
data.tar.gz: 47a372ba9ede858b3cf1c1f697c6476027ecfa4f
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 3077e27f95bf06842b167e819396557fe3337c48bed7f520e1225da0f08918004120676ec78e7fa1188f1c610628c3dc557d67af14ddb86616bded777ce90cb6
|
7
|
+
data.tar.gz: 7150d1b9c1f9fe84109098848d98f1b0da45cf56a2125b011b974a4ff8270a8f251077b508633245d73a78e85dc1faeaba85d3307f906bb91078660698f428df
|
data/.gitignore
ADDED
data/.rspec
ADDED
data/Changes
ADDED
data/Gemfile
ADDED
data/LICENSE.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2013 Fuji, Goro
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,42 @@
|
|
1
|
+
# VisualWidth [](https://travis-ci.org/gfx/visual_width.rb)
|
2
|
+
|
3
|
+
This gem handles Unicode East Asian Width:
|
4
|
+
|
5
|
+
* http://www.unicode.org/reports/tr11/
|
6
|
+
|
7
|
+
## Installation
|
8
|
+
|
9
|
+
Add this line to your application's Gemfile:
|
10
|
+
|
11
|
+
gem 'visual_width'
|
12
|
+
|
13
|
+
And then execute:
|
14
|
+
|
15
|
+
$ bundle
|
16
|
+
|
17
|
+
Or install it yourself as:
|
18
|
+
|
19
|
+
$ gem install visual_width
|
20
|
+
|
21
|
+
## Usage
|
22
|
+
|
23
|
+
```ruby
|
24
|
+
require 'visual_width'
|
25
|
+
|
26
|
+
p VisualWidth.count("こんにちは") # => 10
|
27
|
+
p VisualWidth.count("abcdefghij") # => 10
|
28
|
+
|
29
|
+
p VisualWidth.truncate("恋すてふ 我が名はまだき 立ちにけり 人知れずこそ 思ひそめしか", 20) # => "恋すてふ 我が名は..."
|
30
|
+
```
|
31
|
+
|
32
|
+
Each method can take `east_asian: false` to tell it is not in an East Asian context, regarding ambiguous characters as half-width.
|
33
|
+
|
34
|
+
See [Ambiguous Characters](http://www.unicode.org/reports/tr11/#Ambiguous) in the report.
|
35
|
+
|
36
|
+
## Contributing
|
37
|
+
|
38
|
+
1. Fork it
|
39
|
+
2. Create your feature branch (`git checkout -b my-new-feature`)
|
40
|
+
3. Commit your changes (`git commit -am 'Add some feature'`)
|
41
|
+
4. Push to the branch (`git push origin my-new-feature`)
|
42
|
+
5. Create new Pull Request
|
data/Rakefile
ADDED
@@ -0,0 +1,14 @@
|
|
1
|
+
require "bundler/gem_tasks"
|
2
|
+
require 'rspec/core/rake_task'
|
3
|
+
|
4
|
+
RSpec::Core::RakeTask.new(:spec)
|
5
|
+
|
6
|
+
task :default => :spec
|
7
|
+
|
8
|
+
task :data do
|
9
|
+
sh "tool/east-asian-width.pl > lib/visual_width/data.rb"
|
10
|
+
end
|
11
|
+
|
12
|
+
task :bench do
|
13
|
+
sh "ruby -Ilib tool/benchmark.rb"
|
14
|
+
end
|
@@ -0,0 +1,204 @@
|
|
1
|
+
# THIS FILE IS AUTOMATICALLY GENERATED.
|
2
|
+
# DO NOT EDIT IT!
|
3
|
+
#
|
4
|
+
# Unicode version 6.3.0 of EastAsianWidth characters
|
5
|
+
#
|
6
|
+
# see http://www.unicode.org/reports/tr11/ for specification
|
7
|
+
# see http://www.unicode.org/Public/UNIDATA/EastAsianWidth.txt for data
|
8
|
+
|
9
|
+
module VisualWidth
|
10
|
+
Wide = '[' +
|
11
|
+
'\u{1100}-\u{115F}' +
|
12
|
+
'\u{2329}-\u{232A}' +
|
13
|
+
'\u{2E80}-\u{2FFF}' +
|
14
|
+
'\u{3001}-\u{303E}' +
|
15
|
+
'\u{3041}-\u{3247}' +
|
16
|
+
'\u{3250}-\u{4DBF}' +
|
17
|
+
'\u{4E00}-\u{A4CF}' +
|
18
|
+
'\u{A960}-\u{A97F}' +
|
19
|
+
'\u{AC00}-\u{D7AF}' +
|
20
|
+
'\u{F900}-\u{FAFF}' +
|
21
|
+
'\u{FE10}-\u{FE1F}' +
|
22
|
+
'\u{FE30}-\u{FE6F}' +
|
23
|
+
'\u{1B000}-\u{1CFFF}' +
|
24
|
+
'\u{1F200}-\u{1F2FF}' +
|
25
|
+
'\u{20000}-\u{E0000}' +
|
26
|
+
']'
|
27
|
+
|
28
|
+
Fullwide = '[' +
|
29
|
+
'\u{3000}' +
|
30
|
+
'\u{FF01}-\u{FF60}' +
|
31
|
+
'\u{FFE0}-\u{FFE7}' +
|
32
|
+
']'
|
33
|
+
|
34
|
+
Ambiguous = '[' +
|
35
|
+
'\u{00A1}' +
|
36
|
+
'\u{00A4}' +
|
37
|
+
'\u{00A7}-\u{00A8}' +
|
38
|
+
'\u{00AA}' +
|
39
|
+
'\u{00AD}-\u{00AE}' +
|
40
|
+
'\u{00B0}-\u{00B4}' +
|
41
|
+
'\u{00B6}-\u{00BA}' +
|
42
|
+
'\u{00BC}-\u{00BF}' +
|
43
|
+
'\u{00C6}' +
|
44
|
+
'\u{00D0}' +
|
45
|
+
'\u{00D7}-\u{00D8}' +
|
46
|
+
'\u{00DE}-\u{00E1}' +
|
47
|
+
'\u{00E6}' +
|
48
|
+
'\u{00E8}-\u{00EA}' +
|
49
|
+
'\u{00EC}-\u{00ED}' +
|
50
|
+
'\u{00F0}' +
|
51
|
+
'\u{00F2}-\u{00F3}' +
|
52
|
+
'\u{00F7}-\u{00FA}' +
|
53
|
+
'\u{00FC}' +
|
54
|
+
'\u{00FE}' +
|
55
|
+
'\u{0101}' +
|
56
|
+
'\u{0111}' +
|
57
|
+
'\u{0113}' +
|
58
|
+
'\u{011B}' +
|
59
|
+
'\u{0126}-\u{0127}' +
|
60
|
+
'\u{012B}' +
|
61
|
+
'\u{0131}-\u{0133}' +
|
62
|
+
'\u{0138}' +
|
63
|
+
'\u{013F}-\u{0142}' +
|
64
|
+
'\u{0144}' +
|
65
|
+
'\u{0148}-\u{014B}' +
|
66
|
+
'\u{014D}' +
|
67
|
+
'\u{0152}-\u{0153}' +
|
68
|
+
'\u{0166}-\u{0167}' +
|
69
|
+
'\u{016B}' +
|
70
|
+
'\u{01CE}' +
|
71
|
+
'\u{01D0}' +
|
72
|
+
'\u{01D2}' +
|
73
|
+
'\u{01D4}' +
|
74
|
+
'\u{01D6}' +
|
75
|
+
'\u{01D8}' +
|
76
|
+
'\u{01DA}' +
|
77
|
+
'\u{01DC}' +
|
78
|
+
'\u{0251}' +
|
79
|
+
'\u{0261}' +
|
80
|
+
'\u{02C4}' +
|
81
|
+
'\u{02C7}' +
|
82
|
+
'\u{02C9}-\u{02CB}' +
|
83
|
+
'\u{02CD}' +
|
84
|
+
'\u{02D0}' +
|
85
|
+
'\u{02D8}-\u{02DB}' +
|
86
|
+
'\u{02DD}' +
|
87
|
+
'\u{02DF}' +
|
88
|
+
'\u{0300}-\u{036F}' +
|
89
|
+
'\u{0391}-\u{03A9}' +
|
90
|
+
'\u{03B1}-\u{03C1}' +
|
91
|
+
'\u{03C3}-\u{03C9}' +
|
92
|
+
'\u{0401}' +
|
93
|
+
'\u{0410}-\u{044F}' +
|
94
|
+
'\u{0451}' +
|
95
|
+
'\u{2010}' +
|
96
|
+
'\u{2013}-\u{2016}' +
|
97
|
+
'\u{2018}-\u{2019}' +
|
98
|
+
'\u{201C}-\u{201D}' +
|
99
|
+
'\u{2020}-\u{2022}' +
|
100
|
+
'\u{2024}-\u{2027}' +
|
101
|
+
'\u{2030}' +
|
102
|
+
'\u{2032}-\u{2033}' +
|
103
|
+
'\u{2035}' +
|
104
|
+
'\u{203B}' +
|
105
|
+
'\u{203E}' +
|
106
|
+
'\u{2074}' +
|
107
|
+
'\u{207F}' +
|
108
|
+
'\u{2081}-\u{2084}' +
|
109
|
+
'\u{20AC}' +
|
110
|
+
'\u{2103}' +
|
111
|
+
'\u{2105}' +
|
112
|
+
'\u{2109}' +
|
113
|
+
'\u{2113}' +
|
114
|
+
'\u{2116}' +
|
115
|
+
'\u{2121}-\u{2122}' +
|
116
|
+
'\u{2126}' +
|
117
|
+
'\u{212B}' +
|
118
|
+
'\u{2153}-\u{2154}' +
|
119
|
+
'\u{215B}-\u{215E}' +
|
120
|
+
'\u{2160}-\u{216B}' +
|
121
|
+
'\u{2170}-\u{2179}' +
|
122
|
+
'\u{2189}-\u{2199}' +
|
123
|
+
'\u{21B8}-\u{21B9}' +
|
124
|
+
'\u{21D2}' +
|
125
|
+
'\u{21D4}' +
|
126
|
+
'\u{21E7}' +
|
127
|
+
'\u{2200}' +
|
128
|
+
'\u{2202}-\u{2203}' +
|
129
|
+
'\u{2207}-\u{2208}' +
|
130
|
+
'\u{220B}' +
|
131
|
+
'\u{220F}' +
|
132
|
+
'\u{2211}' +
|
133
|
+
'\u{2215}' +
|
134
|
+
'\u{221A}' +
|
135
|
+
'\u{221D}-\u{2220}' +
|
136
|
+
'\u{2223}' +
|
137
|
+
'\u{2225}' +
|
138
|
+
'\u{2227}-\u{222C}' +
|
139
|
+
'\u{222E}' +
|
140
|
+
'\u{2234}-\u{2237}' +
|
141
|
+
'\u{223C}-\u{223D}' +
|
142
|
+
'\u{2248}' +
|
143
|
+
'\u{224C}' +
|
144
|
+
'\u{2252}' +
|
145
|
+
'\u{2260}-\u{2261}' +
|
146
|
+
'\u{2264}-\u{2267}' +
|
147
|
+
'\u{226A}-\u{226B}' +
|
148
|
+
'\u{226E}-\u{226F}' +
|
149
|
+
'\u{2282}-\u{2283}' +
|
150
|
+
'\u{2286}-\u{2287}' +
|
151
|
+
'\u{2295}' +
|
152
|
+
'\u{2299}' +
|
153
|
+
'\u{22A5}' +
|
154
|
+
'\u{22BF}' +
|
155
|
+
'\u{2312}' +
|
156
|
+
'\u{2460}-\u{24E9}' +
|
157
|
+
'\u{24EB}-\u{254B}' +
|
158
|
+
'\u{2550}-\u{2573}' +
|
159
|
+
'\u{2580}-\u{258F}' +
|
160
|
+
'\u{2592}-\u{2595}' +
|
161
|
+
'\u{25A0}-\u{25A1}' +
|
162
|
+
'\u{25A3}-\u{25A9}' +
|
163
|
+
'\u{25B2}-\u{25B3}' +
|
164
|
+
'\u{25B6}-\u{25B7}' +
|
165
|
+
'\u{25BC}-\u{25BD}' +
|
166
|
+
'\u{25C0}-\u{25C1}' +
|
167
|
+
'\u{25C6}-\u{25C8}' +
|
168
|
+
'\u{25CB}' +
|
169
|
+
'\u{25CE}-\u{25D1}' +
|
170
|
+
'\u{25E2}-\u{25E5}' +
|
171
|
+
'\u{25EF}' +
|
172
|
+
'\u{2605}-\u{2606}' +
|
173
|
+
'\u{2609}' +
|
174
|
+
'\u{260E}-\u{260F}' +
|
175
|
+
'\u{2614}-\u{2615}' +
|
176
|
+
'\u{261C}' +
|
177
|
+
'\u{261E}' +
|
178
|
+
'\u{2640}' +
|
179
|
+
'\u{2642}' +
|
180
|
+
'\u{2660}-\u{2661}' +
|
181
|
+
'\u{2663}-\u{2665}' +
|
182
|
+
'\u{2667}-\u{266A}' +
|
183
|
+
'\u{266C}-\u{266D}' +
|
184
|
+
'\u{266F}' +
|
185
|
+
'\u{269E}-\u{269F}' +
|
186
|
+
'\u{26BE}-\u{26BF}' +
|
187
|
+
'\u{26C4}-\u{26CD}' +
|
188
|
+
'\u{26CF}-\u{26E1}' +
|
189
|
+
'\u{26E3}' +
|
190
|
+
'\u{26E8}-\u{2700}' +
|
191
|
+
'\u{273D}' +
|
192
|
+
'\u{2757}' +
|
193
|
+
'\u{2776}-\u{277F}' +
|
194
|
+
'\u{2B55}-\u{2BFF}' +
|
195
|
+
'\u{3248}-\u{324F}' +
|
196
|
+
'\u{E000}-\u{F8FF}' +
|
197
|
+
'\u{FE00}-\u{FE0F}' +
|
198
|
+
'\u{FFFD}-\u{FFFF}' +
|
199
|
+
'\u{1F100}-\u{1F12D}' +
|
200
|
+
'\u{1F130}-\u{1F169}' +
|
201
|
+
'\u{1F170}-\u{1F1E5}' +
|
202
|
+
']'
|
203
|
+
end
|
204
|
+
|
data/lib/visual_width.rb
ADDED
@@ -0,0 +1,53 @@
|
|
1
|
+
require "visual_width/version"
|
2
|
+
require "visual_width/data"
|
3
|
+
|
4
|
+
module VisualWidth
|
5
|
+
EAST_ASIAN = true
|
6
|
+
|
7
|
+
@@c1 = /( (?:#{Fullwide} | #{Wide} | #{Ambiguous})+ )/x
|
8
|
+
@@c0 = /( (?:#{Fullwide} | #{Wide} )+ )/x
|
9
|
+
|
10
|
+
@@t1 = /( (?:#{Fullwide} | #{Wide} | #{Ambiguous}) ) | ./x
|
11
|
+
@@t0 = /( (?:#{Fullwide} | #{Wide} ) ) | ./x
|
12
|
+
|
13
|
+
module_function
|
14
|
+
|
15
|
+
def count(str, east_asian: EAST_ASIAN)
|
16
|
+
rx = east_asian ? @@c1 : @@c0
|
17
|
+
full_width = 0
|
18
|
+
str.scan(rx) do |w,|
|
19
|
+
full_width += w.length
|
20
|
+
end
|
21
|
+
(full_width * 2) + (str.length - full_width)
|
22
|
+
end
|
23
|
+
|
24
|
+
def truncate(str, max_length, omission: '...', east_asian: EAST_ASIAN)
|
25
|
+
max = max_length - omission.length
|
26
|
+
rx = east_asian ? @@t1 : @@t0
|
27
|
+
pos = 0
|
28
|
+
width = 0
|
29
|
+
str.scan(rx) do |wide,|
|
30
|
+
if wide
|
31
|
+
width += 2
|
32
|
+
else
|
33
|
+
width += 1
|
34
|
+
end
|
35
|
+
|
36
|
+
if width > max
|
37
|
+
break
|
38
|
+
end
|
39
|
+
|
40
|
+
pos += 1
|
41
|
+
|
42
|
+
if width == max
|
43
|
+
break
|
44
|
+
end
|
45
|
+
end
|
46
|
+
|
47
|
+
if width < str.length
|
48
|
+
str.slice(0, pos) + omission
|
49
|
+
else
|
50
|
+
str
|
51
|
+
end
|
52
|
+
end
|
53
|
+
end
|
@@ -0,0 +1,19 @@
|
|
1
|
+
require 'rspec'
|
2
|
+
|
3
|
+
require 'visual_width/string_refine'
|
4
|
+
|
5
|
+
using VisualWidth
|
6
|
+
|
7
|
+
describe VisualWidth do
|
8
|
+
context "String#width" do
|
9
|
+
it do
|
10
|
+
expect("こんにちは".width).to eql(10)
|
11
|
+
end
|
12
|
+
end
|
13
|
+
|
14
|
+
context "String#width" do
|
15
|
+
it do
|
16
|
+
expect("こんにちは".truncate(5)).to eql('こ...')
|
17
|
+
end
|
18
|
+
end
|
19
|
+
end
|
@@ -0,0 +1,90 @@
|
|
1
|
+
require 'rspec'
|
2
|
+
|
3
|
+
require 'visual_width'
|
4
|
+
require 'visual_width/string_ext'
|
5
|
+
|
6
|
+
describe VisualWidth do
|
7
|
+
context ".count" do
|
8
|
+
it "counts Half characters" do
|
9
|
+
expect(VisualWidth.count("foo")).to eql(3)
|
10
|
+
end
|
11
|
+
it "counts Wide characters" do
|
12
|
+
expect(VisualWidth.count("こんにちは")).to eql(10)
|
13
|
+
end
|
14
|
+
|
15
|
+
it "counts Half characters" do
|
16
|
+
expect(VisualWidth.count("コンニチハ")).to eql(5)
|
17
|
+
end
|
18
|
+
|
19
|
+
it "counts Ambiguous characters" do
|
20
|
+
expect(VisualWidth.count("αβ")).to eql(4)
|
21
|
+
end
|
22
|
+
end
|
23
|
+
|
24
|
+
context ".counnt with east_asian: false" do
|
25
|
+
it "counts Halfwide characters" do
|
26
|
+
expect(VisualWidth.count("foo", east_asian: false)).to eql(3)
|
27
|
+
end
|
28
|
+
it "counts Wide characters" do
|
29
|
+
expect(VisualWidth.count("こんにちは", east_asian: false)).to eql(10)
|
30
|
+
end
|
31
|
+
|
32
|
+
it "counts Half characters" do
|
33
|
+
expect(VisualWidth.count("コンニチハ", east_asian: false)).to eql(5)
|
34
|
+
end
|
35
|
+
|
36
|
+
it "counts Ambiguous characters" do
|
37
|
+
expect(VisualWidth.count("αβ", east_asian: false)).to eql(2)
|
38
|
+
end
|
39
|
+
end
|
40
|
+
|
41
|
+
context ".truncate" do
|
42
|
+
it "does nothing if str is short enough" do
|
43
|
+
expect(VisualWidth.truncate("foo", 20)).to eql("foo")
|
44
|
+
end
|
45
|
+
|
46
|
+
it "truncates str, adding '...'" do
|
47
|
+
str = "The quick brown fox jumps over the lazy dog."
|
48
|
+
s = VisualWidth.truncate(str, 15)
|
49
|
+
expect(s.visual_width).to be <= 15
|
50
|
+
expect(s).to eql('The quick br...')
|
51
|
+
end
|
52
|
+
|
53
|
+
it "truncates str, adding '(snip)'" do
|
54
|
+
str = "The quick brown fox jumps over the lazy dog."
|
55
|
+
s = VisualWidth.truncate(str, 15, omission: '(snip)')
|
56
|
+
expect(s.visual_width).to be <= 15
|
57
|
+
expect(s).to eql('The quick(snip)')
|
58
|
+
end
|
59
|
+
|
60
|
+
it "truncates Wide characters" do
|
61
|
+
str = "くにざかいのながいトンネルを抜けるとゆきぐにであった"
|
62
|
+
s = VisualWidth.truncate(str, 20)
|
63
|
+
expect(s.visual_width).to be <= 20
|
64
|
+
expect(s).to eql('くにざかいのなが...')
|
65
|
+
end
|
66
|
+
|
67
|
+
it "truncates mixed character types" do
|
68
|
+
str = "くにざかいの ながい トンネルを 抜けると ゆきぐにであった"
|
69
|
+
s = VisualWidth.truncate(str, 20)
|
70
|
+
expect(s.visual_width).to be <= 20
|
71
|
+
expect(s).to eql('くにざかいの なが...')
|
72
|
+
end
|
73
|
+
|
74
|
+
it "truncates Ambiguous characters" do
|
75
|
+
str = "αβγδεζηθικλμνξοπρστυφχψω"
|
76
|
+
s = VisualWidth.truncate(str, 10)
|
77
|
+
expect(s.visual_width).to be <= 10
|
78
|
+
expect(s).to eql('αβγ...')
|
79
|
+
end
|
80
|
+
end
|
81
|
+
|
82
|
+
context ".truncate with east_asian: false" do
|
83
|
+
it "truncates str with Ambiguous characters" do
|
84
|
+
str = "αβγδεζηθικλμνξοπρστυφχψω"
|
85
|
+
s = VisualWidth.truncate(str, 10, east_asian: false)
|
86
|
+
expect(s.visual_width(east_asian: false)).to be <= 10
|
87
|
+
expect(s).to eql('αβγδεζη...')
|
88
|
+
end
|
89
|
+
end
|
90
|
+
end
|
data/tool/benchmark.rb
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
require 'benchmark'
|
4
|
+
require 'visual_width'
|
5
|
+
require 'unicode/display_width'
|
6
|
+
|
7
|
+
str = "foo あいうえお bar αβγδε baz"
|
8
|
+
|
9
|
+
VisualWidth.count(str) == str.display_width(2)
|
10
|
+
|
11
|
+
[str, str * 5, str * 10].each do |str|
|
12
|
+
puts "\n", "for length #{str.length}:"
|
13
|
+
Benchmark.bmbm do |x|
|
14
|
+
x.report("visual_width") do
|
15
|
+
1000.times { VisualWidth.count(str) }
|
16
|
+
end
|
17
|
+
x.report("display_width") do
|
18
|
+
1000.times { str.display_width(2) }
|
19
|
+
end
|
20
|
+
end
|
21
|
+
end
|
@@ -0,0 +1,113 @@
|
|
1
|
+
#!/usr/bin/env perl
|
2
|
+
# vim: set expandtab shiftwidth=2 tabstop=2:
|
3
|
+
use 5.14.0;
|
4
|
+
use strict;
|
5
|
+
use warnings;
|
6
|
+
use LWP::Simple qw(get);
|
7
|
+
use Text::Xslate;
|
8
|
+
use Log::Minimal qw(infof);
|
9
|
+
|
10
|
+
my $SPEC = "http://www.unicode.org/reports/tr11/";
|
11
|
+
my $DATA = "http://www.unicode.org/Public/UNIDATA/EastAsianWidth.txt";
|
12
|
+
|
13
|
+
my %params = (
|
14
|
+
spec_uri => $SPEC,
|
15
|
+
data_uri => $DATA,
|
16
|
+
|
17
|
+
F => [],
|
18
|
+
W => [],
|
19
|
+
H => [],
|
20
|
+
N => [],
|
21
|
+
Na => [],
|
22
|
+
A => [],
|
23
|
+
);
|
24
|
+
|
25
|
+
infof 'get data from %s', $DATA;
|
26
|
+
my $data = get($DATA);
|
27
|
+
|
28
|
+
my($version) = $data =~ /EastAsianWidth-([\d\.]+)\.txt/;
|
29
|
+
$params{version} = $version;
|
30
|
+
|
31
|
+
infof 'parse character tratis';
|
32
|
+
my @traits;
|
33
|
+
for my $line(split /\n/, $data) {
|
34
|
+
$line =~ s/\#.*//;
|
35
|
+
if ($line =~ m{ (?<code_first> \w+) \.\.(?<code_last> \w+) ; (?<trait> \w+) }xms) {
|
36
|
+
for my $code(hex($+{code_first}) .. hex($+{code_last})) {
|
37
|
+
$traits[$code] = $+{trait};
|
38
|
+
}
|
39
|
+
}
|
40
|
+
elsif ($line =~ m{ (?<code> \w+) ; (?<trait> \w+) }xms) {
|
41
|
+
$traits[hex $+{code}] = $+{trait};
|
42
|
+
}
|
43
|
+
}
|
44
|
+
infof 'make ranges from traits';
|
45
|
+
|
46
|
+
my $c = 0;
|
47
|
+
my $trait = "";
|
48
|
+
my $first = -1;
|
49
|
+
foreach my $t(@traits) {
|
50
|
+
next unless $t;
|
51
|
+
|
52
|
+
if ($t ne $trait) {
|
53
|
+
if ($trait ne "" && $first != -1) {
|
54
|
+
push $params{$trait}, [ $first, $c - 1 ];
|
55
|
+
}
|
56
|
+
|
57
|
+
$first = $c;
|
58
|
+
$trait = $t;
|
59
|
+
}
|
60
|
+
}
|
61
|
+
continue {
|
62
|
+
++$c;
|
63
|
+
}
|
64
|
+
|
65
|
+
infof 'render it';
|
66
|
+
|
67
|
+
my $xslate = Text::Xslate->new(
|
68
|
+
type => "text",
|
69
|
+
cache => 0,
|
70
|
+
type => 'text',
|
71
|
+
module => ['Text::Xslate::Bridge::Star'],
|
72
|
+
);
|
73
|
+
|
74
|
+
print $xslate->render_string(do {
|
75
|
+
local $/;
|
76
|
+
<DATA>;
|
77
|
+
}, \%params);
|
78
|
+
|
79
|
+
__DATA__
|
80
|
+
# THIS FILE IS AUTOMATICALLY GENERATED.
|
81
|
+
# DO NOT EDIT IT!
|
82
|
+
#
|
83
|
+
# Unicode version <: $version :> of EastAsianWidth characters
|
84
|
+
#
|
85
|
+
# see <: $spec_uri :> for specification
|
86
|
+
# see <: $data_uri :> for data
|
87
|
+
|
88
|
+
module VisualWidth
|
89
|
+
<:-
|
90
|
+
macro range -> $r {
|
91
|
+
if $r[0] == $r[1] {
|
92
|
+
sprintf('\u{%04X}', $r[0]);
|
93
|
+
}
|
94
|
+
else {
|
95
|
+
sprintf('\u{%04X}-\u{%04X}', $r[0], $r[1]);
|
96
|
+
}
|
97
|
+
}
|
98
|
+
-:>
|
99
|
+
: macro make_conditions -> $t, $name {
|
100
|
+
<: $name :> = '[' +
|
101
|
+
: for $t -> $range {
|
102
|
+
'<: range($range) :>' +
|
103
|
+
: }
|
104
|
+
']'
|
105
|
+
: }
|
106
|
+
|
107
|
+
: make_conditions($W, "Wide");
|
108
|
+
|
109
|
+
: make_conditions($F, "Fullwide");
|
110
|
+
|
111
|
+
: make_conditions($A, "Ambiguous");
|
112
|
+
end
|
113
|
+
|
@@ -0,0 +1,23 @@
|
|
1
|
+
# coding: utf-8
|
2
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
require 'visual_width/version'
|
5
|
+
|
6
|
+
Gem::Specification.new do |spec|
|
7
|
+
spec.name = "visual_width"
|
8
|
+
spec.version = VisualWidth::VERSION
|
9
|
+
spec.authors = ["Fuji, Goro (gfx)"]
|
10
|
+
spec.email = ["gfuji@cpan.org"]
|
11
|
+
spec.description = %q{Deals with East Asian Width defined in Unicode}
|
12
|
+
spec.summary = %q{Ruby Implementation of East Asian Width}
|
13
|
+
spec.homepage = "https://github.com/gfx/visual_width.rb"
|
14
|
+
spec.license = "MIT"
|
15
|
+
|
16
|
+
spec.files = `git ls-files`.split($/)
|
17
|
+
spec.executables = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
|
18
|
+
spec.test_files = spec.files.grep(%r{^(test|spec|features)/})
|
19
|
+
spec.require_paths = ["lib"]
|
20
|
+
|
21
|
+
spec.add_development_dependency "bundler", "~> 1.3"
|
22
|
+
spec.add_development_dependency "rake"
|
23
|
+
end
|
metadata
ADDED
@@ -0,0 +1,92 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: visual_width
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Fuji, Goro (gfx)
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2013-10-23 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: bundler
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - ~>
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '1.3'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - ~>
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.3'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rake
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - '>='
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - '>='
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '0'
|
41
|
+
description: Deals with East Asian Width defined in Unicode
|
42
|
+
email:
|
43
|
+
- gfuji@cpan.org
|
44
|
+
executables: []
|
45
|
+
extensions: []
|
46
|
+
extra_rdoc_files: []
|
47
|
+
files:
|
48
|
+
- .gitignore
|
49
|
+
- .rspec
|
50
|
+
- Changes
|
51
|
+
- Gemfile
|
52
|
+
- LICENSE.txt
|
53
|
+
- README.md
|
54
|
+
- Rakefile
|
55
|
+
- lib/visual_width.rb
|
56
|
+
- lib/visual_width/data.rb
|
57
|
+
- lib/visual_width/string_ext.rb
|
58
|
+
- lib/visual_width/string_refine.rb
|
59
|
+
- lib/visual_width/version.rb
|
60
|
+
- spec/visual_width/string_refine_spec.rb
|
61
|
+
- spec/visual_width_spec.rb
|
62
|
+
- tool/benchmark.rb
|
63
|
+
- tool/east-asian-width.pl
|
64
|
+
- visual_width.gemspec
|
65
|
+
homepage: https://github.com/gfx/visual_width.rb
|
66
|
+
licenses:
|
67
|
+
- MIT
|
68
|
+
metadata: {}
|
69
|
+
post_install_message:
|
70
|
+
rdoc_options: []
|
71
|
+
require_paths:
|
72
|
+
- lib
|
73
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
74
|
+
requirements:
|
75
|
+
- - '>='
|
76
|
+
- !ruby/object:Gem::Version
|
77
|
+
version: '0'
|
78
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - '>='
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
83
|
+
requirements: []
|
84
|
+
rubyforge_project:
|
85
|
+
rubygems_version: 2.0.3
|
86
|
+
signing_key:
|
87
|
+
specification_version: 4
|
88
|
+
summary: Ruby Implementation of East Asian Width
|
89
|
+
test_files:
|
90
|
+
- spec/visual_width/string_refine_spec.rb
|
91
|
+
- spec/visual_width_spec.rb
|
92
|
+
has_rdoc:
|