mattt · May 4, 2024 11:22
diff --git a/main.swift b/main.swift
 import NaturalLanguage

 let languages: [NLLanguage] = [
    .amharic, .arabic, .armenian, .bengali, .bulgarian,
    .burmese, .catalan, .cherokee, .croatian, .czech,
    .danish, .dutch, .english, .finnish, .french,
    .georgian, .german, .greek, .gujarati, .hebrew,
    .hindi, .hungarian, .icelandic, .indonesian, .italian,
    .japanese, .kannada, .khmer, .korean, .lao,
    .malay, .malayalam, .marathi, .mongolian, .norwegian,
    .oriya, .persian, .polish, .portuguese, .punjabi,
    .romanian, .russian, .simplifiedChinese, .sinhalese, .slovak,
    .spanish, .swedish, .tamil, .telugu, .thai,
    .tibetan, .traditionalChinese, .turkish, .ukrainian, .urdu,
    .vietnamese
 ].sorted { $0.rawValue < $1.rawValue }

 let tagSchemes: [NLTagScheme] = [
    .language, .script, .tokenType, .lexicalClass, .nameType,
    .nameTypeOrLexicalClass, .lemma, .sentimentScore
 ]

 let columns = ["bcp47"] + tagSchemes.map { $0.rawValue }
 print(columns.joined(separator: ","))

 for language in languages {
    let units: Set<NLTokenUnit> = [.document, .paragraph, .sentence, .word]
    let availableTagSchemes: Set<NLTagScheme> = Set(units.flatMap { unit in
        NLTagger.availableTagSchemes(for: unit, language: language)
    })

    let row = [language.rawValue] + tagSchemes.map { availableTagSchemes.contains($0) ? "true" : "false" }
    print(row.joined(separator: ","))
 }
diff --git a/results-iOS-13.csv b/results-iOS-13.csv
diff --git a/results-macOS-10.15.2.csv b/results-macOS-10.15.2.csv
diff --git a/results-tvOS-13.csv b/results-tvOS-13.csv
	import NaturalLanguage

	let languages: [NLLanguage] = [
	.amharic, .arabic, .armenian, .bengali, .bulgarian,
	.burmese, .catalan, .cherokee, .croatian, .czech,
	.danish, .dutch, .english, .finnish, .french,
	.georgian, .german, .greek, .gujarati, .hebrew,
	.hindi, .hungarian, .icelandic, .indonesian, .italian,
	.japanese, .kannada, .khmer, .korean, .lao,
	.malay, .malayalam, .marathi, .mongolian, .norwegian,
	.oriya, .persian, .polish, .portuguese, .punjabi,
	.romanian, .russian, .simplifiedChinese, .sinhalese, .slovak,
	.spanish, .swedish, .tamil, .telugu, .thai,
	.tibetan, .traditionalChinese, .turkish, .ukrainian, .urdu,
	.vietnamese
	].sorted { $0.rawValue < $1.rawValue }

	let tagSchemes: [NLTagScheme] = [
	.language, .script, .tokenType, .lexicalClass, .nameType,
	.nameTypeOrLexicalClass, .lemma, .sentimentScore
	]

	let columns = ["bcp47"] + tagSchemes.map { $0.rawValue }
	print(columns.joined(separator: ","))

	for language in languages {
	let units: Set<NLTokenUnit> = [.document, .paragraph, .sentence, .word]
	let availableTagSchemes: Set<NLTagScheme> = Set(units.flatMap { unit in
	NLTagger.availableTagSchemes(for: unit, language: language)
	})

	let row = [language.rawValue] + tagSchemes.map { availableTagSchemes.contains($0) ? "true" : "false" }
	print(row.joined(separator: ","))
	}
bcp47	Language	Script	TokenType	LexicalClass	NameType	NameTypeOrLexicalClass	Lemma	Sentiment
am	true	true	true	false	false	false	false	false
ar	true	true	true	false	false	false	false	false
bg	true	true	true	false	false	false	false	false
bn	true	true	true	false	false	false	false	false
bo	true	true	true	false	false	false	false	false
ca	true	true	true	false	false	false	false	false
chr	true	true	true	false	false	false	false	false
cs	true	true	true	false	false	false	false	false
da	true	true	true	false	false	false	false	false
de	true	true	true	true	true	true	true	true
el	true	true	true	false	false	false	false	false
en	true	true	true	true	true	true	true	true
es	true	true	true	true	true	true	true	true
fa	true	true	true	false	false	false	false	false
fi	true	true	true	false	false	false	false	false
fr	true	true	true	true	true	true	true	true
gu	true	true	true	false	false	false	false	false
he	true	true	true	false	false	false	false	false
hi	true	true	true	false	false	false	false	false
hr	true	true	true	false	false	false	false	false
hu	true	true	true	false	false	false	false	false
hy	true	true	true	false	false	false	false	false
id	true	true	true	false	false	false	false	false
is	true	true	true	false	false	false	false	false
it	true	true	true	true	true	true	true	true
ja	true	true	true	false	false	false	false	false
ka	true	true	true	false	false	false	false	false
km	true	true	true	false	false	false	false	false
kn	true	true	true	false	false	false	false	false
ko	true	true	true	false	false	false	false	false
lo	true	true	true	false	false	false	false	false
ml	true	true	true	false	false	false	false	false
mn-Mong	true	true	true	false	false	false	false	false
mr	true	true	true	false	false	false	false	false
ms	true	true	true	false	false	false	false	false
my	true	true	true	false	false	false	false	false
nb	true	true	true	false	false	false	false	false
nl	true	true	true	false	false	false	false	false
or	true	true	true	false	false	false	false	false
pa-Guru	true	true	true	false	false	false	false	false
pl	true	true	true	false	false	false	false	false
pt	true	true	true	true	true	true	true	true
ro	true	true	true	false	false	false	false	false
ru	true	true	true	true	true	true	true	false
si	true	true	true	false	false	false	false	false
sk	true	true	true	false	false	false	false	false
sv	true	true	true	false	false	false	false	false
ta	true	true	true	false	false	false	false	false
te	true	true	true	false	false	false	false	false
th	true	true	true	false	false	false	false	false
tr	true	true	true	true	true	true	true	false
uk	true	true	true	false	false	false	false	false
ur	true	true	true	false	false	false	false	false
vi	true	true	true	false	false	false	false	false
zh-Hans	true	true	true	false	false	false	false	false
zh-Hant	true	true	true	false	false	false	false	false