wordcuta 0.2.0 → 0.3.0

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.
Files changed (4) hide show
  1. checksums.yaml +4 -4
  2. data/lib/wordcuta.rb +12 -5
  3. data/wordcuta.gemspec +4 -5
  4. metadata +11 -11
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 3f77edbb75622406c64839dff1ed7c45c5f0d07ac9775f268d9ab36dba0b163c
4
- data.tar.gz: 16c7697a501a4aaa044dde4a0c2626e176192fac42a65c8fcfe29eaacd2b9236
3
+ metadata.gz: 4496824d87c7eb49d72ecc7442dcbfa3a1485dbeff9b70579f1717b016bf006f
4
+ data.tar.gz: 32ae3e5760f83eb64ef3a89235c2f0e41812bb7d63d97fe30a8818b5007f39d3
5
5
  SHA512:
6
- metadata.gz: d1f63dcb36276a0e2118cde0aae5a07c44fc73a78937f787d0802d7ba68977dfcee3549ac7c6ab3c6470ced53aa44a4e19857abaeeeb7cf60fcf80f84430b698
7
- data.tar.gz: 66dc0d88cf17cef24a2ff66140fe1e7b0e6f043f75513be0fb3097103a5cf38beba55e0eaa13d4472cab8d9756a42f3be4d29fefffc40184ef4dade8e9c0fca7
6
+ metadata.gz: 0b4d8503f9a59b61cbd156c97c8b1461d9d3ba732aab2f585ff542932c9868f0d07cd6169ab7bd85cfca33271acb7a1ed88cd3de50cf6fbfac9f6fd8afa1a82c
7
+ data.tar.gz: '08467876b7ade9e6e49c1b06482f8564470b541e94435f8fab6558d5ef0b4e6a4a3dfab4652896ac5efae67b28500d7e93905884f2f2cccdbafd61cd45e2cc72'
data/lib/wordcuta.rb CHANGED
@@ -10,8 +10,9 @@ module WordcutFFI
10
10
  extend FFI::Library
11
11
 
12
12
  ffi_lib "wordcutw"
13
-
13
+
14
14
  attach_function :wordcut_new_with_dict, [:string], :pointer
15
+ attach_function :wordcut_new_with_dict_and_cluster_rules, [:string, :string], :pointer
15
16
  attach_function :wordcut_into_text_ranges, [:pointer, :string, :pointer], :pointer
16
17
  attach_function :wordcut_into_strings, [:pointer, :string, :pointer], :pointer
17
18
  attach_function :wordcut_put_delimiters, [:pointer, :string, :string], :string
@@ -23,11 +24,17 @@ module WordcutA
23
24
  TextRange = Struct.new(:s, :e)
24
25
 
25
26
  DEFAULT_THAI_DICT_PATH = File.expand_path('../../data/thai-dix.txt', __FILE__)
26
-
27
+
27
28
  class Wordcut
28
- def initialize(dict_path)
29
- @wordcut_p = FFI::AutoPointer.new(WordcutFFI.wordcut_new_with_dict(dict_path),
30
- WordcutFFI.method(:delete_wordcut))
29
+ def initialize(dict_path, cluster_rule_path = nil)
30
+ if cluster_rule_path
31
+ @wordcut_p = FFI::AutoPointer.new(
32
+ WordcutFFI.wordcut_new_with_dict_and_cluster_rules(dict_path, cluster_rule_path),
33
+ WordcutFFI.method(:delete_wordcut))
34
+ else
35
+ @wordcut_p = FFI::AutoPointer.new(WordcutFFI.wordcut_new_with_dict(dict_path),
36
+ WordcutFFI.method(:delete_wordcut))
37
+ end
31
38
  end
32
39
 
33
40
  def into_ranges(text)
data/wordcuta.gemspec CHANGED
@@ -1,13 +1,12 @@
1
1
  Gem::Specification.new do |s|
2
2
  s.name = 'wordcuta'
3
- s.version = '0.2.0'
3
+ s.version = '0.3.0'
4
4
  s.authors = ['Vee Satayamas']
5
- s.email = ['5ssgdxltv@relay.firefox.com']
5
+ s.email = ['vsatayamas@gmail.com']
6
6
  s.licenses = ['LGPL-3.0']
7
- s.description = "A word segmentation tools for ASEAN languages wrapper for Ruby"
8
- s.homepage = "https://github.com/veer66/wordcuta"
7
+ s.description = "A Ruby wrapper of wordcut-engine - a word segmentation tools for ASEAN languages, i.e. Khmer, Lao, Myanmar, and Thai"
8
+ s.homepage = "https://codeberg.org/mekong/wordcuta"
9
9
  s.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
10
10
  s.summary = "A word segmentation tools for ASEAN languages wrapper for Ruby"
11
11
  s.files = %w(README.md LICENSE Gemfile wordcuta.gemspec data/thai-dix.txt lib/wordcuta.rb)
12
- s.requirements << 'ffi, v1.15.5'
13
12
  end
metadata CHANGED
@@ -1,18 +1,19 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: wordcuta
3
3
  version: !ruby/object:Gem::Version
4
- version: 0.2.0
4
+ version: 0.3.0
5
5
  platform: ruby
6
6
  authors:
7
7
  - Vee Satayamas
8
- autorequire:
8
+ autorequire:
9
9
  bindir: bin
10
10
  cert_chain: []
11
- date: 2022-02-16 00:00:00.000000000 Z
11
+ date: 2023-12-19 00:00:00.000000000 Z
12
12
  dependencies: []
13
- description: A word segmentation tools for ASEAN languages wrapper for Ruby
13
+ description: A Ruby wrapper of wordcut-engine - a word segmentation tools for ASEAN
14
+ languages, i.e. Khmer, Lao, Myanmar, and Thai
14
15
  email:
15
- - 5ssgdxltv@relay.firefox.com
16
+ - vsatayamas@gmail.com
16
17
  executables: []
17
18
  extensions: []
18
19
  extra_rdoc_files: []
@@ -23,11 +24,11 @@ files:
23
24
  - data/thai-dix.txt
24
25
  - lib/wordcuta.rb
25
26
  - wordcuta.gemspec
26
- homepage: https://github.com/veer66/wordcuta
27
+ homepage: https://codeberg.org/mekong/wordcuta
27
28
  licenses:
28
29
  - LGPL-3.0
29
30
  metadata: {}
30
- post_install_message:
31
+ post_install_message:
31
32
  rdoc_options: []
32
33
  require_paths:
33
34
  - lib
@@ -41,10 +42,9 @@ required_rubygems_version: !ruby/object:Gem::Requirement
41
42
  - - ">="
42
43
  - !ruby/object:Gem::Version
43
44
  version: '0'
44
- requirements:
45
- - ffi, v1.15.5
46
- rubygems_version: 3.2.22
47
- signing_key:
45
+ requirements: []
46
+ rubygems_version: 3.4.10
47
+ signing_key:
48
48
  specification_version: 4
49
49
  summary: A word segmentation tools for ASEAN languages wrapper for Ruby
50
50
  test_files: []