gettext.js is a lightweight (3k minified!) yet complete and accurate GNU gettext port for node and the browser. Manage your i18n translations the right way in your javascript projects.
gettext.js aim is to port the famous GNU gettext and ngettext functions to javascript node and browser applications. It should be standards respectful and lightweight (no dictionary loading management, no extra features).
The result is a < 200 lines javascript tiny lib yet implementing fully
gettext()
and ngettext()
and having the lighter json translation files
possible (embeding only translated forms, and not much headers).
There are plenty good i18n libraries out there, notably Jed and i18n-next, but either there are too complex and too heavy, or they do not embrace fully the gettext API and philosophy.
There is also gettext.js which is pretty good and heavily inspired this one, but not active since 2012 and without any tests.
apt-get install libjs-gettext.js
Install the lib with the following command: npm install gettext.js --save
Require it in your project:
var i18n = require('gettext.js');
i18n.gettext('foo');
Download the latest
archive or
get it through bower: bower install gettext.js --save
<script src="/path/to/gettext.js" type="text/javascript"></script>
<script>
var i18n = window.i18n(options);
i18n.gettext('foo');
</script>
You can load your messages these ways:
// i18n.setMessages(domain, locale, messages, plural_form);
i18n.setMessages('messages', 'fr', {
"Welcome": "Bienvenue",
"There is %1 apple": [
"Il y a %1 pomme",
"Il y a %1 pommes"
]
}, 'nplurals=2; plural=n>1;');
// i18n.loadJSON(jsonData /*, domain */);
var json = {
"": {
"locale": "fr",
"plural-forms": "nplurals=2; plural=n>1;"
}
"Welcome": "Bienvenue",
"There is %1 apple": [
"Il y a %1 pomme",
"Il y a %1 pommes"
]
};
i18n.loadJSON(jsonData, 'messages');
See Required JSON format section below for more info.
You could do it from your dom
<html lang="fr">
or from javascript
i18n.setLocale('fr');
Translate a string.
Translate a pluralizable string
gettext('There are %1 in the %2
, 'apples', 'bowl'); -> "There are apples in the bowl
ngettext('One %2', '%1 %2', 10, 'bananas'); -> "10 bananas"`
It uses the public method i18n.strfmt("string", var1, var2, ...)
you could
reuse elsewhere in your project.
You'll find in /bin
a po2json.js
converter, based on the excellent
po2json project that will dump your
.po
files into the proper json format below:
{
"": {
"lang": "en",
"plural_forms": "nplurals=2; plural=(n!=1);"
},
"simple key": "It's tranlation",
"another with %1 parameter": "It's %1 tranlsation",
"a key with plural": [
"a plural form",
"another plural form",
"could have up to 6 forms with some languages"
],
"a context\u0004a contextualized key": "translation here"
}
Use bin/po2json.js input.po output.json
or
bin/po2json.js input.po output.json -p
for pretty format.
You could use xgettext-php parser to parse your files. It provides helpful javascript and handlebars parsers.
MIT