2017-10-19 09:51:54 +00:00
|
|
|
#!/usr/bin/env node
|
|
|
|
|
2017-10-18 12:39:04 +00:00
|
|
|
/*
|
|
|
|
Copyright 2017 New Vector Ltd
|
|
|
|
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
you may not use this file except in compliance with the License.
|
|
|
|
You may obtain a copy of the License at
|
|
|
|
|
|
|
|
http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
See the License for the specific language governing permissions and
|
|
|
|
limitations under the License.
|
|
|
|
*/
|
|
|
|
|
2017-10-17 18:19:01 +00:00
|
|
|
/**
|
2017-10-18 13:45:04 +00:00
|
|
|
* Regenerates the translations en_EN file by walking the source tree and
|
2017-10-17 18:19:01 +00:00
|
|
|
* parsing each file with flow-parser. Emits a JSON file with the
|
|
|
|
* translatable strings mapped to themselves in the order they appeared
|
|
|
|
* in the files and grouped by the file they appeared in.
|
|
|
|
*
|
|
|
|
* Usage: node scripts/gen-i18n.js
|
|
|
|
*/
|
|
|
|
const fs = require('fs');
|
|
|
|
const path = require('path');
|
|
|
|
|
|
|
|
const walk = require('walk');
|
|
|
|
|
|
|
|
const flowParser = require('flow-parser');
|
|
|
|
const estreeWalker = require('estree-walker');
|
|
|
|
|
|
|
|
const TRANSLATIONS_FUNCS = ['_t', '_td', '_tJsx'];
|
|
|
|
|
2017-10-18 13:30:55 +00:00
|
|
|
const INPUT_TRANSLATIONS_FILE = 'src/i18n/strings/en_EN.json';
|
2017-10-20 17:38:22 +00:00
|
|
|
const OUTPUT_FILE = 'src/i18n/strings/en_EN.json';
|
2017-10-17 18:19:01 +00:00
|
|
|
|
2017-10-18 18:36:07 +00:00
|
|
|
// NB. The sync version of walk is broken for single files so we walk
|
|
|
|
// all of res rather than just res/home.html.
|
|
|
|
// https://git.daplie.com/Daplie/node-walk/merge_requests/1 fixes it,
|
|
|
|
// or if we get bored waiting for it to be merged, we could switch
|
|
|
|
// to a project that's actively maintained.
|
|
|
|
const SEARCH_PATHS = ['src', 'res'];
|
|
|
|
|
2017-10-17 18:19:01 +00:00
|
|
|
const FLOW_PARSER_OPTS = {
|
|
|
|
esproposal_class_instance_fields: true,
|
|
|
|
esproposal_class_static_fields: true,
|
|
|
|
esproposal_decorators: true,
|
|
|
|
esproposal_export_star_as: true,
|
|
|
|
types: true,
|
|
|
|
};
|
|
|
|
|
|
|
|
function getObjectValue(obj, key) {
|
|
|
|
for (const prop of obj.properties) {
|
|
|
|
if (prop.key.type == 'Identifier' && prop.key.name == key) {
|
|
|
|
return prop.value;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return null;
|
|
|
|
}
|
|
|
|
|
|
|
|
function getTKey(arg) {
|
|
|
|
if (arg.type == 'Literal') {
|
|
|
|
return arg.value;
|
|
|
|
} else if (arg.type == 'BinaryExpression' && arg.operator == '+') {
|
|
|
|
return getTKey(arg.left) + getTKey(arg.right);
|
|
|
|
} else if (arg.type == 'TemplateLiteral') {
|
|
|
|
return arg.quasis.map((q) => {
|
|
|
|
return q.value.raw;
|
|
|
|
}).join('');
|
|
|
|
}
|
|
|
|
return null;
|
|
|
|
}
|
|
|
|
|
2017-10-20 17:38:22 +00:00
|
|
|
function getFormatStrings(str) {
|
|
|
|
// Match anything that starts with %
|
|
|
|
// We could make a regex that matched the full placeholder, but this
|
|
|
|
// would just not match invalid placeholders and so wouldn't help us
|
|
|
|
// detect the invalid ones.
|
|
|
|
// Also note that for simplicity, this just matches a % character and then
|
|
|
|
// anything up to the next % character (or a single %, or end of string).
|
|
|
|
const formatStringRe = /%([^%]+|%|$)/g;
|
|
|
|
const formatStrings = new Set();
|
|
|
|
|
|
|
|
let match;
|
2017-10-23 13:02:58 +00:00
|
|
|
while ( (match = formatStringRe.exec(str)) !== null ) {
|
2017-10-20 17:38:22 +00:00
|
|
|
const placeholder = match[1]; // Minus the leading '%'
|
|
|
|
if (placeholder === '%') continue; // Literal % is %%
|
|
|
|
|
|
|
|
const placeholderMatch = placeholder.match(/^\((.*?)\)(.)/);
|
|
|
|
if (placeholderMatch === null) {
|
|
|
|
throw new Error("Invalid format specifier: '"+match[0]+"'");
|
|
|
|
}
|
|
|
|
if (placeholderMatch.length < 3) {
|
|
|
|
throw new Error("Malformed format specifier");
|
|
|
|
}
|
2017-10-23 09:18:29 +00:00
|
|
|
const placeholderName = placeholderMatch[1];
|
|
|
|
const placeholderFormat = placeholderMatch[2];
|
2017-10-20 17:38:22 +00:00
|
|
|
|
2017-10-23 09:18:29 +00:00
|
|
|
if (placeholderFormat !== 's') {
|
|
|
|
throw new Error(`'${placeholderFormat}' used as format character: you probably meant 's'`);
|
2017-10-20 17:38:22 +00:00
|
|
|
}
|
|
|
|
|
2017-10-23 09:18:29 +00:00
|
|
|
formatStrings.add(placeholderName);
|
2017-10-20 17:38:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return formatStrings;
|
|
|
|
}
|
|
|
|
|
2017-10-18 18:36:07 +00:00
|
|
|
function getTranslationsJs(file) {
|
2017-10-18 13:33:21 +00:00
|
|
|
const tree = flowParser.parse(fs.readFileSync(file, { encoding: 'utf8' }), FLOW_PARSER_OPTS);
|
2017-10-17 18:19:01 +00:00
|
|
|
|
|
|
|
const trs = new Set();
|
|
|
|
|
|
|
|
estreeWalker.walk(tree, {
|
|
|
|
enter: function(node, parent) {
|
|
|
|
if (
|
|
|
|
node.type == 'CallExpression' &&
|
|
|
|
TRANSLATIONS_FUNCS.includes(node.callee.name)
|
|
|
|
) {
|
|
|
|
const tKey = getTKey(node.arguments[0]);
|
2017-10-18 13:43:18 +00:00
|
|
|
// This happens whenever we call _t with non-literals (ie. whenever we've
|
|
|
|
// had to use a _td to compensate) so is expected.
|
2017-10-17 18:19:01 +00:00
|
|
|
if (tKey === null) return;
|
|
|
|
|
2017-10-20 17:38:22 +00:00
|
|
|
// check the format string against the args
|
|
|
|
// We only check _t: _tJsx is much more complex and _td has no args
|
|
|
|
if (node.callee.name === '_t') {
|
|
|
|
try {
|
|
|
|
const placeholders = getFormatStrings(tKey);
|
|
|
|
for (const placeholder of placeholders) {
|
|
|
|
if (node.arguments.length < 2) {
|
|
|
|
throw new Error(`Placeholder found ('${placeholder}') but no substitutions given`);
|
|
|
|
}
|
|
|
|
const value = getObjectValue(node.arguments[1], placeholder);
|
|
|
|
if (value === null) {
|
|
|
|
throw new Error(`No value found for placeholder '${placeholder}'`);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} catch (e) {
|
|
|
|
console.log();
|
|
|
|
console.error(`ERROR: ${file}:${node.loc.start.line} ${tKey}`);
|
|
|
|
process.exit(1);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-10-17 18:19:01 +00:00
|
|
|
let isPlural = false;
|
|
|
|
if (node.arguments.length > 1 && node.arguments[1].type == 'ObjectExpression') {
|
|
|
|
const countVal = getObjectValue(node.arguments[1], 'count');
|
|
|
|
if (countVal) {
|
|
|
|
isPlural = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (isPlural) {
|
2017-10-18 13:30:55 +00:00
|
|
|
trs.add(tKey + "|other");
|
|
|
|
const plurals = enPlurals[tKey];
|
|
|
|
if (plurals) {
|
|
|
|
for (const pluralType of Object.keys(plurals)) {
|
|
|
|
trs.add(tKey + "|" + pluralType);
|
|
|
|
}
|
2017-10-17 18:19:01 +00:00
|
|
|
}
|
|
|
|
} else {
|
|
|
|
trs.add(tKey);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
});
|
|
|
|
|
|
|
|
return trs;
|
|
|
|
}
|
|
|
|
|
2017-10-18 18:36:07 +00:00
|
|
|
function getTranslationsOther(file) {
|
|
|
|
const contents = fs.readFileSync(file, { encoding: 'utf8' });
|
|
|
|
|
|
|
|
const trs = new Set();
|
|
|
|
|
|
|
|
// Taken from riot-web src/components/structures/HomePage.js
|
|
|
|
const translationsRegex = /_t\(['"]([\s\S]*?)['"]\)/mg;
|
|
|
|
let matches;
|
|
|
|
while (matches = translationsRegex.exec(contents)) {
|
|
|
|
trs.add(matches[1]);
|
|
|
|
}
|
|
|
|
return trs;
|
|
|
|
}
|
|
|
|
|
2017-10-18 13:30:55 +00:00
|
|
|
// gather en_EN plural strings from the input translations file:
|
|
|
|
// the en_EN strings are all in the source with the exception of
|
|
|
|
// pluralised strings, which we need to pull in from elsewhere.
|
|
|
|
const inputTranslationsRaw = JSON.parse(fs.readFileSync(INPUT_TRANSLATIONS_FILE, { encoding: 'utf8' }));
|
|
|
|
const enPlurals = {};
|
|
|
|
|
|
|
|
for (const key of Object.keys(inputTranslationsRaw)) {
|
|
|
|
const parts = key.split("|");
|
|
|
|
if (parts.length > 1) {
|
|
|
|
const plurals = enPlurals[parts[0]] || {};
|
|
|
|
plurals[parts[1]] = inputTranslationsRaw[key];
|
|
|
|
enPlurals[parts[0]] = plurals;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-10-17 18:19:01 +00:00
|
|
|
const translatables = new Set();
|
|
|
|
|
2017-10-18 18:36:07 +00:00
|
|
|
const walkOpts = {
|
2017-10-17 18:19:01 +00:00
|
|
|
listeners: {
|
|
|
|
file: function(root, fileStats, next) {
|
|
|
|
const fullPath = path.join(root, fileStats.name);
|
2017-10-18 18:36:07 +00:00
|
|
|
|
|
|
|
let ltrs;
|
|
|
|
if (fileStats.name.endsWith('.js')) {
|
|
|
|
trs = getTranslationsJs(fullPath);
|
|
|
|
} else if (fileStats.name.endsWith('.html')) {
|
|
|
|
trs = getTranslationsOther(fullPath);
|
|
|
|
} else {
|
|
|
|
return;
|
|
|
|
}
|
2017-10-17 18:19:01 +00:00
|
|
|
console.log(`${fullPath} (${trs.size} strings)`);
|
|
|
|
for (const tr of trs.values()) {
|
|
|
|
translatables.add(tr);
|
|
|
|
}
|
|
|
|
},
|
|
|
|
}
|
2017-10-18 18:36:07 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (const path of SEARCH_PATHS) {
|
|
|
|
if (fs.existsSync(path)) {
|
|
|
|
walk.walkSync(path, walkOpts);
|
|
|
|
}
|
|
|
|
}
|
2017-10-17 18:19:01 +00:00
|
|
|
|
|
|
|
const trObj = {};
|
|
|
|
for (const tr of translatables) {
|
|
|
|
trObj[tr] = tr;
|
2017-10-18 13:30:55 +00:00
|
|
|
if (tr.includes("|")) {
|
|
|
|
trObj[tr] = inputTranslationsRaw[tr];
|
|
|
|
}
|
2017-10-17 18:19:01 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fs.writeFileSync(
|
2017-10-20 17:38:22 +00:00
|
|
|
OUTPUT_FILE,
|
2017-10-17 18:19:01 +00:00
|
|
|
JSON.stringify(trObj, translatables.values(), 4) + "\n"
|
|
|
|
);
|
|
|
|
|
2017-10-20 17:38:22 +00:00
|
|
|
console.log();
|
|
|
|
console.log(`Wrote ${translatables.size} strings to ${OUTPUT_FILE}`);
|