overleaf/services/web/scripts/translations/download.js
Jakob Ackermann 99a414f345 Merge pull request #4656 from overleaf/jpa-rewrite-quotes-in-locales
[misc] rewrite single quote to left/right single quotation mark in locales

GitOrigin-RevId: a021fb6841425555b9af79a9146820299cb93fc2
2021-08-17 08:04:49 +00:00

89 lines
2.8 KiB
JavaScript
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

const { promises: fs } = require('fs')
const oneSky = require('@brainly/onesky-utils')
const sanitizeHtml = require('sanitize-html')
const { withAuth } = require('./config')
async function run() {
try {
// The recommended OneSky set-up appears to require an API request to
// generate files on their side, which you could then request and use. We
// only have 1 such file that appears to be misnamed (en-US, despite our
// translations being marked as GB) and very out-of-date.
// However by requesting the "multilingual file" for this file, we get all
// of the translations
const content = await oneSky.getMultilingualFile(
withAuth({
fileName: 'en-US.json',
})
)
const json = JSON.parse(content)
for (const [code, lang] of Object.entries(json)) {
if (code === 'en-GB') {
// OneSky does not have read-after-write consistency.
// Skip the dump of English locales, which may not include locales
// that were just uploaded.
continue
}
for (let [key, value] of Object.entries(lang.translation)) {
// Handle multi-line strings as arrays by joining on newline
if (Array.isArray(value)) {
value = value.join('\n')
}
lang.translation[key] = sanitize(value)
}
await fs.writeFile(
`${__dirname}/../../locales/${code}.json`,
JSON.stringify(lang.translation, null, 2) + '\n'
)
}
} catch (error) {
console.error(error)
process.exit(1)
}
}
run()
/**
* Sanitize a translation string to prevent injection attacks
*
* @param {string} input
* @returns {string}
*/
function sanitize(input) {
// Block Angular XSS
// Ticket: https://github.com/overleaf/issues/issues/4478
input = input.replace(/'/g, '')
// Use left quote where (likely) appropriate.
input.replace(/ /g, ' ')
return sanitizeHtml(input, {
// Allow "replacement" tags (in the format <0>, <1>, <2>, etc) used by
// react-i18next to allow for HTML insertion via the Trans component.
// See: https://github.com/overleaf/developer-manual/blob/master/code/translations.md
// Unfortunately the sanitizeHtml library does not accept regexes or a
// function for the allowedTags option, so we are limited to a hard-coded
// number of "replacement" tags.
allowedTags: ['b', 'strong', 'a', 'code', ...range(10)],
allowedAttributes: {
a: ['href', 'class'],
},
textFilter(text) {
return text
.replace(/\{\{/, '&#123;&#123;')
.replace(/\}\}/, '&#125;&#125;')
},
})
}
/**
* Generate a range of numbers as strings up to the given size
*
* @param {number} size Size of range
* @returns {string[]}
*/
function range(size) {
return Array.from(Array(size).keys()).map(n => n.toString())
}