wordcuta 0.2.1 → 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.
- checksums.yaml +4 -4
- data/lib/wordcuta.rb +12 -5
- data/wordcuta.gemspec +4 -5
- metadata +11 -24
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 4496824d87c7eb49d72ecc7442dcbfa3a1485dbeff9b70579f1717b016bf006f
|
|
4
|
+
data.tar.gz: 32ae3e5760f83eb64ef3a89235c2f0e41812bb7d63d97fe30a8818b5007f39d3
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
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
|
-
|
|
30
|
-
|
|
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.
|
|
3
|
+
s.version = '0.3.0'
|
|
4
4
|
s.authors = ['Vee Satayamas']
|
|
5
|
-
s.email = ['
|
|
5
|
+
s.email = ['vsatayamas@gmail.com']
|
|
6
6
|
s.licenses = ['LGPL-3.0']
|
|
7
|
-
s.description = "A word segmentation tools for ASEAN languages
|
|
8
|
-
s.homepage = "https://
|
|
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.add_dependency 'ffi', '1.15.5'
|
|
13
12
|
end
|
metadata
CHANGED
|
@@ -1,32 +1,19 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: wordcuta
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 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:
|
|
12
|
-
dependencies:
|
|
13
|
-
-
|
|
14
|
-
|
|
15
|
-
requirement: !ruby/object:Gem::Requirement
|
|
16
|
-
requirements:
|
|
17
|
-
- - '='
|
|
18
|
-
- !ruby/object:Gem::Version
|
|
19
|
-
version: 1.15.5
|
|
20
|
-
type: :runtime
|
|
21
|
-
prerelease: false
|
|
22
|
-
version_requirements: !ruby/object:Gem::Requirement
|
|
23
|
-
requirements:
|
|
24
|
-
- - '='
|
|
25
|
-
- !ruby/object:Gem::Version
|
|
26
|
-
version: 1.15.5
|
|
27
|
-
description: A word segmentation tools for ASEAN languages wrapper for Ruby
|
|
11
|
+
date: 2023-12-19 00:00:00.000000000 Z
|
|
12
|
+
dependencies: []
|
|
13
|
+
description: A Ruby wrapper of wordcut-engine - a word segmentation tools for ASEAN
|
|
14
|
+
languages, i.e. Khmer, Lao, Myanmar, and Thai
|
|
28
15
|
email:
|
|
29
|
-
-
|
|
16
|
+
- vsatayamas@gmail.com
|
|
30
17
|
executables: []
|
|
31
18
|
extensions: []
|
|
32
19
|
extra_rdoc_files: []
|
|
@@ -37,11 +24,11 @@ files:
|
|
|
37
24
|
- data/thai-dix.txt
|
|
38
25
|
- lib/wordcuta.rb
|
|
39
26
|
- wordcuta.gemspec
|
|
40
|
-
homepage: https://
|
|
27
|
+
homepage: https://codeberg.org/mekong/wordcuta
|
|
41
28
|
licenses:
|
|
42
29
|
- LGPL-3.0
|
|
43
30
|
metadata: {}
|
|
44
|
-
post_install_message:
|
|
31
|
+
post_install_message:
|
|
45
32
|
rdoc_options: []
|
|
46
33
|
require_paths:
|
|
47
34
|
- lib
|
|
@@ -56,8 +43,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
|
56
43
|
- !ruby/object:Gem::Version
|
|
57
44
|
version: '0'
|
|
58
45
|
requirements: []
|
|
59
|
-
rubygems_version: 3.
|
|
60
|
-
signing_key:
|
|
46
|
+
rubygems_version: 3.4.10
|
|
47
|
+
signing_key:
|
|
61
48
|
specification_version: 4
|
|
62
49
|
summary: A word segmentation tools for ASEAN languages wrapper for Ruby
|
|
63
50
|
test_files: []
|