unicode 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.
- data/README +107 -0
- data/bin/mkunidata.rb +128 -0
- data/ext/extconf.rb +3 -0
- data/ext/unicode.c +666 -0
- data/ext/unidata.map +10642 -0
- data/ext/ustring.c +206 -0
- data/ext/ustring.h +48 -0
- data/ext/wstring.c +185 -0
- data/ext/wstring.h +41 -0
- data/test/test.rb +68 -0
- metadata +47 -0
data/test/test.rb
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
1
|
+
#! /usr/local/bin/ruby -KU
|
|
2
|
+
|
|
3
|
+
require 'unicode'
|
|
4
|
+
|
|
5
|
+
## dump Unicode string
|
|
6
|
+
class String
|
|
7
|
+
def udump
|
|
8
|
+
ustr = self.unpack("U*")
|
|
9
|
+
ret = []
|
|
10
|
+
ustr.each do |e|
|
|
11
|
+
if e.is_a?(Integer)
|
|
12
|
+
ret << "U+%04X" % e
|
|
13
|
+
else
|
|
14
|
+
ret << e
|
|
15
|
+
end
|
|
16
|
+
end
|
|
17
|
+
ret
|
|
18
|
+
end
|
|
19
|
+
end
|
|
20
|
+
|
|
21
|
+
|
|
22
|
+
print "Canonical decomposition vs compatibility decomposition\n"
|
|
23
|
+
p Unicode::decompose("⑽ o\xef\xac\x83ce").udump
|
|
24
|
+
p Unicode::decompose_compat("⑽ o\xef\xac\x83ce")
|
|
25
|
+
|
|
26
|
+
print "Canonical equivalent vs Compatibility equivalent\n"
|
|
27
|
+
p Unicode::strcmp("ｶﾞ", "ガ")
|
|
28
|
+
p Unicode::strcmp("ガ", "ｶﾞ")
|
|
29
|
+
p Unicode::strcmp_compat("ｶﾞ", "ガ")
|
|
30
|
+
|
|
31
|
+
print "Decomposition/composition\n"
|
|
32
|
+
p Unicode::normalize_D([?c, 0x301, 0x327].pack("U*")).udump
|
|
33
|
+
p Unicode::normalize_D([?c, 0x327, 0x301].pack("U*")).udump
|
|
34
|
+
p Unicode::normalize_D([0x107, 0x327].pack("U*")).udump
|
|
35
|
+
p Unicode::normalize_D([0xe7, 0x301].pack("U*")).udump
|
|
36
|
+
p Unicode::normalize_C([?c, 0x301, 0x327].pack("U*")).udump
|
|
37
|
+
p Unicode::normalize_C([?c, 0x327, 0x301].pack("U*")).udump
|
|
38
|
+
p Unicode::normalize_C([0x107, 0x327].pack("U*")).udump
|
|
39
|
+
p Unicode::normalize_C([0xe7, 0x301].pack("U*")).udump
|
|
40
|
+
|
|
41
|
+
print "Kana Normalization\n"
|
|
42
|
+
p Unicode::normalize_D("ｶﾞガ").udump
|
|
43
|
+
p Unicode::normalize_C("ｶﾞガ").udump
|
|
44
|
+
p Unicode::normalize_KD("ｶﾞガ").udump
|
|
45
|
+
p Unicode::normalize_KC("ｶﾞガ").udump
|
|
46
|
+
|
|
47
|
+
print "Hangul\n"
|
|
48
|
+
p "요시담".udump
|
|
49
|
+
p Unicode::normalize_D("요시담").udump
|
|
50
|
+
p Unicode::normalize_C("요시담").udump
|
|
51
|
+
|
|
52
|
+
print "Composition Exclusion\n"
|
|
53
|
+
print " ANGSTROM SIGN [U+221B]\n"
|
|
54
|
+
p Unicode::normalize_D([0x212b].pack("U")).udump
|
|
55
|
+
p Unicode::normalize_C([0x212b].pack("U")).udump
|
|
56
|
+
print " LATIN CAPITAL LETTER A WITH RING ABOVE [U+00C5]\n"
|
|
57
|
+
p Unicode::normalize_D([0x00c5].pack("U")).udump
|
|
58
|
+
p Unicode::normalize_C([0x00c5].pack("U")).udump
|
|
59
|
+
|
|
60
|
+
print "Case conversion\n"
|
|
61
|
+
p Unicode::normalize_C(Unicode::upcase([?c, 0x301, 0x327, 0xff41].pack("U*"))).udump
|
|
62
|
+
p Unicode::normalize_C(Unicode::downcase([?C, 0x301, 0x327, 0xff21].pack("U*"))).udump
|
|
63
|
+
p Unicode::capitalize([0x1f1, ?A, ?a, 0xff21].pack("U*")).udump
|
|
64
|
+
|
|
65
|
+
|
|
66
|
+
## Local variables:
|
|
67
|
+
## coding: utf-8
|
|
68
|
+
## End:
|
metadata
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
|
2
|
+
rubygems_version: 0.8.10
|
|
3
|
+
specification_version: 1
|
|
4
|
+
name: unicode
|
|
5
|
+
version: !ruby/object:Gem::Version
|
|
6
|
+
version: "0.1"
|
|
7
|
+
date: 2005-08-13
|
|
8
|
+
summary: Unicode normalization library.
|
|
9
|
+
require_paths:
|
|
10
|
+
- "."
|
|
11
|
+
email: yoshidam@yoshidam.net
|
|
12
|
+
homepage: http://www.yoshidam.net/Ruby.html#unicode
|
|
13
|
+
rubyforge_project:
|
|
14
|
+
description:
|
|
15
|
+
autorequire:
|
|
16
|
+
default_executable:
|
|
17
|
+
bindir: bin
|
|
18
|
+
has_rdoc: false
|
|
19
|
+
required_ruby_version: !ruby/object:Gem::Version::Requirement
|
|
20
|
+
requirements:
|
|
21
|
+
-
|
|
22
|
+
- ">"
|
|
23
|
+
- !ruby/object:Gem::Version
|
|
24
|
+
version: 0.0.0
|
|
25
|
+
version:
|
|
26
|
+
platform: ruby
|
|
27
|
+
authors: []
|
|
28
|
+
files:
|
|
29
|
+
- bin/mkunidata.rb
|
|
30
|
+
- ext/ustring.c
|
|
31
|
+
- ext/ustring.h
|
|
32
|
+
- ext/unicode.c
|
|
33
|
+
- ext/wstring.c
|
|
34
|
+
- ext/wstring.h
|
|
35
|
+
- ext/extconf.rb
|
|
36
|
+
- ext/unidata.map
|
|
37
|
+
- test/test.rb
|
|
38
|
+
- README
|
|
39
|
+
test_files: []
|
|
40
|
+
rdoc_options: []
|
|
41
|
+
extra_rdoc_files:
|
|
42
|
+
- README
|
|
43
|
+
executables: []
|
|
44
|
+
extensions:
|
|
45
|
+
- ext/extconf.rb
|
|
46
|
+
requirements: []
|
|
47
|
+
dependencies: []
|