medieval_latina 2.0.5 → 3.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.DS_Store +0 -0
- data/.github/workflows/tests.yml +1 -5
- data/.standard.yml +1 -1
- data/.tool-versions +1 -1
- data/Gemfile +0 -5
- data/Gemfile.lock +7 -2
- data/README.md +49 -0
- data/bin/build +34 -0
- data/bin/lint +6 -0
- data/data/dictionary.json +4982 -0
- data/lexicons/Latin00.pls +2210 -0
- data/lexicons/Latin01.pls +2210 -0
- data/lexicons/Latin02.pls +2210 -0
- data/lexicons/Latin03.pls +2210 -0
- data/lexicons/Latin04.pls +2210 -0
- data/lexicons/Latin05.pls +2210 -0
- data/lexicons/Latin06.pls +2210 -0
- data/lexicons/Latin07.pls +2210 -0
- data/lexicons/Latin08.pls +2210 -0
- data/lexicons/Latin09.pls +10 -0
- data/lib/medieval_latina/lexicon.rb +27 -0
- data/lib/medieval_latina/lexicon_builder.rb +44 -0
- data/lib/medieval_latina/version.rb +1 -1
- data/lib/medieval_latina.rb +52 -20
- data/medieval_latina.gemspec +5 -1
- metadata +76 -5
- data/lib/medieval_latina/dictionary.rb +0 -2714
@@ -0,0 +1,27 @@
|
|
1
|
+
class MedievalLatina
|
2
|
+
class Lexicon
|
3
|
+
LEXICON_DIR = File.expand_path("../../lexicons", __dir__)
|
4
|
+
|
5
|
+
def self.list_files
|
6
|
+
Dir.glob(File.join(LEXICON_DIR, "*.pls"))
|
7
|
+
end
|
8
|
+
|
9
|
+
def self.get_file_path(file_name)
|
10
|
+
path = File.join(LEXICON_DIR, file_name)
|
11
|
+
return path if File.exist?(path)
|
12
|
+
raise "Lexicon file not found: #{file_name}"
|
13
|
+
end
|
14
|
+
|
15
|
+
def self.get_file_content(file_name)
|
16
|
+
path = get_file_path(file_name)
|
17
|
+
File.read(path)
|
18
|
+
end
|
19
|
+
|
20
|
+
def self.file_names_with_contents
|
21
|
+
list_files.each_with_object({}) do |file_path, files_content|
|
22
|
+
file_name, extension = File.basename(file_path).split(".")
|
23
|
+
files_content[file_name] = get_file_content("#{file_name}.#{extension}")
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
27
|
+
end
|
@@ -0,0 +1,44 @@
|
|
1
|
+
require "cgi"
|
2
|
+
require "rexml/document"
|
3
|
+
|
4
|
+
class MedievalLatina
|
5
|
+
class LexiconBuilder
|
6
|
+
include REXML
|
7
|
+
|
8
|
+
def initialize(words)
|
9
|
+
@document = Document.new
|
10
|
+
@words = words
|
11
|
+
end
|
12
|
+
|
13
|
+
def call
|
14
|
+
document.add_element "lexicon", SPECIFICATION
|
15
|
+
|
16
|
+
words.each do |word, pronunciation|
|
17
|
+
lexeme = Element.new("lexeme")
|
18
|
+
grapheme = Element.new("grapheme")
|
19
|
+
phoneme = Element.new("phoneme")
|
20
|
+
|
21
|
+
grapheme.text = CGI.unescapeHTML(word)
|
22
|
+
phoneme.text = pronunciation
|
23
|
+
|
24
|
+
lexeme.add_element(grapheme)
|
25
|
+
lexeme.add_element(phoneme)
|
26
|
+
|
27
|
+
document.root.add_element(lexeme)
|
28
|
+
end
|
29
|
+
|
30
|
+
document
|
31
|
+
end
|
32
|
+
|
33
|
+
private
|
34
|
+
|
35
|
+
attr_reader :document, :words
|
36
|
+
|
37
|
+
SPECIFICATION = {
|
38
|
+
"version" => "1.0",
|
39
|
+
"xmlns" => "http://www.w3.org/2005/01/pronunciation-lexicon",
|
40
|
+
"alphabet" => "ipa",
|
41
|
+
"xml:lang" => "en-US"
|
42
|
+
}.freeze
|
43
|
+
end
|
44
|
+
end
|
data/lib/medieval_latina.rb
CHANGED
@@ -1,5 +1,7 @@
|
|
1
|
-
require "
|
1
|
+
require "json"
|
2
2
|
require "medieval_latina/initializer"
|
3
|
+
require "medieval_latina/lexicon"
|
4
|
+
require "medieval_latina/lexicon_builder"
|
3
5
|
require "medieval_latina/version"
|
4
6
|
require "set"
|
5
7
|
|
@@ -9,8 +11,8 @@ class MedievalLatina
|
|
9
11
|
if word?(string)
|
10
12
|
metadata = DICTIONARY.fetch(string, {})
|
11
13
|
|
12
|
-
if metadata.key?(
|
13
|
-
metadata[
|
14
|
+
if metadata.key?("pronunciation")
|
15
|
+
metadata["pronunciation"]
|
14
16
|
else
|
15
17
|
new(string).call
|
16
18
|
end
|
@@ -22,6 +24,15 @@ class MedievalLatina
|
|
22
24
|
rejoin_words(prepared_words)
|
23
25
|
end
|
24
26
|
|
27
|
+
def self.dictionary
|
28
|
+
@data ||= load_data
|
29
|
+
end
|
30
|
+
|
31
|
+
def self.load_data
|
32
|
+
file_path = File.join(File.dirname(__FILE__), "../data/dictionary.json")
|
33
|
+
JSON.parse(File.read(file_path))
|
34
|
+
end
|
35
|
+
|
25
36
|
def self.prepare_text(text)
|
26
37
|
text.scan(/[\p{Alnum}'-]+|[[:punct:]]+/).map do |string|
|
27
38
|
if word?(string)
|
@@ -37,31 +48,47 @@ class MedievalLatina
|
|
37
48
|
end
|
38
49
|
|
39
50
|
def self.adjective?(word)
|
40
|
-
|
51
|
+
adjectives.key?(prepare_word(word))
|
41
52
|
end
|
42
53
|
|
43
54
|
def self.adverb?(word)
|
44
|
-
|
55
|
+
adverbs.key?(prepare_word(word))
|
45
56
|
end
|
46
57
|
|
47
58
|
def self.noun?(word)
|
48
|
-
|
59
|
+
nouns.key?(prepare_word(word))
|
49
60
|
end
|
50
61
|
|
51
62
|
def self.verb?(word)
|
52
|
-
|
63
|
+
verbs.key?(prepare_word(word))
|
53
64
|
end
|
54
65
|
|
55
66
|
def self.adjectives
|
56
|
-
|
67
|
+
DICTIONARY.select do |word, metadata|
|
68
|
+
metadata["part"] == "Adjective"
|
69
|
+
end
|
57
70
|
end
|
58
71
|
|
59
72
|
def self.adverbs
|
60
|
-
|
73
|
+
DICTIONARY.select do |word, metadata|
|
74
|
+
metadata["part"] == "Adverb"
|
75
|
+
end
|
61
76
|
end
|
62
77
|
|
63
78
|
def self.nouns
|
64
|
-
|
79
|
+
DICTIONARY.select do |word, metadata|
|
80
|
+
metadata["part"] == "Noun"
|
81
|
+
end
|
82
|
+
end
|
83
|
+
|
84
|
+
def self.pronunciations_for(words)
|
85
|
+
words.map(&:downcase).each_with_object({}) do |word, hash|
|
86
|
+
metadata = DICTIONARY[word]
|
87
|
+
|
88
|
+
if metadata && metadata["ipa"]
|
89
|
+
hash[word] = metadata["ipa"]
|
90
|
+
end
|
91
|
+
end
|
65
92
|
end
|
66
93
|
|
67
94
|
def self.rejoin_words(array)
|
@@ -74,7 +101,9 @@ class MedievalLatina
|
|
74
101
|
end
|
75
102
|
|
76
103
|
def self.verbs
|
77
|
-
|
104
|
+
DICTIONARY.select do |word, metadata|
|
105
|
+
metadata["part"] == "Verb"
|
106
|
+
end
|
78
107
|
end
|
79
108
|
|
80
109
|
def self.word?(string)
|
@@ -82,13 +111,7 @@ class MedievalLatina
|
|
82
111
|
end
|
83
112
|
|
84
113
|
def self.words
|
85
|
-
|
86
|
-
ADJECTIVES,
|
87
|
-
ADVERBS,
|
88
|
-
DICTIONARY,
|
89
|
-
NOUNS,
|
90
|
-
VERBS
|
91
|
-
].flat_map(&:keys).each_with_object(Set.new) { |word, set| set.add(word) }
|
114
|
+
DICTIONARY.keys.to_set
|
92
115
|
end
|
93
116
|
|
94
117
|
def initialize(word)
|
@@ -122,6 +145,15 @@ class MedievalLatina
|
|
122
145
|
x: ->(rest) { "ks" }
|
123
146
|
}
|
124
147
|
CONSONENT_TEAMS = {gn: "n-y", qu: "kw"}.freeze
|
148
|
+
PARTS_OF_SPEECH = [
|
149
|
+
"Adjective",
|
150
|
+
"Adverb",
|
151
|
+
"Conjunction",
|
152
|
+
"Noun",
|
153
|
+
"Preposition",
|
154
|
+
"Pronoun",
|
155
|
+
"Verb"
|
156
|
+
].to_set.freeze
|
125
157
|
SOFT_C = ["e", "i", "ae", "oe"].freeze
|
126
158
|
SOFT_G = SOFT_C
|
127
159
|
SOFT_T = ["i"].freeze
|
@@ -163,13 +195,13 @@ class MedievalLatina
|
|
163
195
|
end
|
164
196
|
|
165
197
|
def to_team
|
166
|
-
"#{character}#{rest[0]}"
|
198
|
+
:"#{character}#{rest[0]}"
|
167
199
|
end
|
168
200
|
end
|
169
201
|
|
170
202
|
class Error < StandardError; end
|
171
203
|
|
172
|
-
DICTIONARY =
|
204
|
+
DICTIONARY = dictionary.each_with_object({}) do |(word, metadata), hash|
|
173
205
|
hash[word] = metadata
|
174
206
|
|
175
207
|
sanitized_word = I18n.transliterate(word)
|
data/medieval_latina.gemspec
CHANGED
@@ -15,7 +15,7 @@ Gem::Specification.new do |spec|
|
|
15
15
|
spec.description = description
|
16
16
|
spec.homepage = "https://github.com/jaysonvirissimo/medieval_latina"
|
17
17
|
spec.license = "MIT"
|
18
|
-
spec.required_ruby_version = Gem::Requirement.new(">=
|
18
|
+
spec.required_ruby_version = Gem::Requirement.new(">= 3.0.0")
|
19
19
|
spec.metadata["allowed_push_host"] = "https://rubygems.org/"
|
20
20
|
spec.metadata["homepage_uri"] = spec.homepage
|
21
21
|
spec.metadata["source_code_uri"] = github_uri
|
@@ -29,4 +29,8 @@ Gem::Specification.new do |spec|
|
|
29
29
|
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
30
30
|
spec.require_paths = ["lib"]
|
31
31
|
spec.add_dependency "i18n"
|
32
|
+
spec.add_development_dependency "rake", "~> 12.0"
|
33
|
+
spec.add_development_dependency "rspec", "~> 3.0"
|
34
|
+
spec.add_development_dependency "nokogiri"
|
35
|
+
spec.add_development_dependency "standard"
|
32
36
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: medieval_latina
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version:
|
4
|
+
version: 3.0.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Jayson Virissimo
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date:
|
11
|
+
date: 2024-04-28 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: i18n
|
@@ -24,6 +24,62 @@ dependencies:
|
|
24
24
|
- - ">="
|
25
25
|
- !ruby/object:Gem::Version
|
26
26
|
version: '0'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rake
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '12.0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '12.0'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: rspec
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '3.0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '3.0'
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: nokogiri
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - ">="
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
type: :development
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - ">="
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: '0'
|
69
|
+
- !ruby/object:Gem::Dependency
|
70
|
+
name: standard
|
71
|
+
requirement: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - ">="
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '0'
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - ">="
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
27
83
|
description: |
|
28
84
|
There are good text-to-speech engines for English and classical Latin, but none for medieval Latin.
|
29
85
|
MedievalLatina converts Latin text to a kind of phonetic spelling that can be read by English text-to-speech engines.
|
@@ -33,6 +89,7 @@ executables: []
|
|
33
89
|
extensions: []
|
34
90
|
extra_rdoc_files: []
|
35
91
|
files:
|
92
|
+
- ".DS_Store"
|
36
93
|
- ".github/workflows/tests.yml"
|
37
94
|
- ".gitignore"
|
38
95
|
- ".rspec"
|
@@ -43,11 +100,25 @@ files:
|
|
43
100
|
- LICENSE.txt
|
44
101
|
- README.md
|
45
102
|
- Rakefile
|
103
|
+
- bin/build
|
46
104
|
- bin/console
|
105
|
+
- bin/lint
|
47
106
|
- bin/setup
|
107
|
+
- data/dictionary.json
|
108
|
+
- lexicons/Latin00.pls
|
109
|
+
- lexicons/Latin01.pls
|
110
|
+
- lexicons/Latin02.pls
|
111
|
+
- lexicons/Latin03.pls
|
112
|
+
- lexicons/Latin04.pls
|
113
|
+
- lexicons/Latin05.pls
|
114
|
+
- lexicons/Latin06.pls
|
115
|
+
- lexicons/Latin07.pls
|
116
|
+
- lexicons/Latin08.pls
|
117
|
+
- lexicons/Latin09.pls
|
48
118
|
- lib/medieval_latina.rb
|
49
|
-
- lib/medieval_latina/dictionary.rb
|
50
119
|
- lib/medieval_latina/initializer.rb
|
120
|
+
- lib/medieval_latina/lexicon.rb
|
121
|
+
- lib/medieval_latina/lexicon_builder.rb
|
51
122
|
- lib/medieval_latina/version.rb
|
52
123
|
- medieval_latina.gemspec
|
53
124
|
homepage: https://github.com/jaysonvirissimo/medieval_latina
|
@@ -66,14 +137,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
|
|
66
137
|
requirements:
|
67
138
|
- - ">="
|
68
139
|
- !ruby/object:Gem::Version
|
69
|
-
version:
|
140
|
+
version: 3.0.0
|
70
141
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
71
142
|
requirements:
|
72
143
|
- - ">="
|
73
144
|
- !ruby/object:Gem::Version
|
74
145
|
version: '0'
|
75
146
|
requirements: []
|
76
|
-
rubygems_version: 3.
|
147
|
+
rubygems_version: 3.5.3
|
77
148
|
signing_key:
|
78
149
|
specification_version: 4
|
79
150
|
summary: Transform medieval Latin text into phonetic English
|