hanzi-to-pinyin
Version:
Convert Hanzi to Pinyin with tone marks or tone numbers
48 lines (43 loc) • 1.3 kB
JavaScript
const getTokens = require('hanzi-tokenizer')
const processToken = (token, numbered, segmented) => {
if (numbered) {
if (segmented) {
return Object.keys(token.definitions).map(v => v.replace(/\s/, ''))
} else {
return Object.keys(token.definitions)
}
} else {
if (segmented) {
return Object.values(token.definitions).map(v => v.pinyin.replace(/\s/, ''))
} else {
return Object.values(token.definitions).map(v => v.pinyin)
}
}
}
const convert = async (text, opts = {}) => {
const tokens = await getTokens(text, Object.assign(opts, { spaces: true }))
const list = []
for (let i = 0; i < tokens.length; i++) {
if (typeof tokens[i] === 'string') {
if (typeof list[list.length - 1] === 'string') {
list[list.length - 1] += tokens[i]
} else {
list.push(tokens[i])
}
} else {
if (Object.keys(tokens[i].definitions).length === 1) {
if (typeof list[list.length - 1] === 'string') {
list[list.length - 1] += processToken(tokens[i], opts.numbered, opts.segmented)[0]
} else {
list.push(processToken(tokens[i], opts.numbered, opts.segmented)[0])
}
} else {
list.push(processToken(tokens[i], opts.numbered, opts.segmented))
}
}
}
return list
}
module.exports = convert
module.exports.init = getTokens.init