Lightweight Telegram API framework for Node.js
✅ Explicit messages handlers
✅ Fallback messages handler (empty string - any messages)
✅ Regex matching on text messages
✅ Buttons support (inline keyboard, reply keyboard and remove reply keyboard)
✅ Fallback handler for any button (empty string - button callback_data)
✅ Edit inline keyboards
✅ Secret token support
✅ Long polling support
✅ Sending chat action
Complete example can be found here
npm install telenode-js
In order to reduce the node_modules
size dotenv
(for environment variables) and express
(for local development /
deployment) packages are defined as devDependencies
.
You don't have to use these packages if you don't want to, and you can use them only for local development if you choose
so.
If you do choose to work with these packages, install them manually:
npm install express dotenv
Or for serverless deployments install express as a dev dependency (to use in your local development environment) instead:
npm install express --save-dev
In order to listen to updates from Telegram servers you have to set up a webhook.
To use the npx set-webhook
command you should provide the webhook parameter and api token.
You can do that by:
- Setting the
WEBHOOK
andAPI_TOKEN
environment variables (SECRET_TOKEN
is optional). - Storing them in
.env
file (SECRET_TOKEN
is optional). - Use the
--apiToken
and--webhook
command arguments (--secretToken
is optional).
You also can provide the secret token parameter if you choose to.
Then you can execute the following command:
npx set-webhook
- If you want to delete a webhook use the command:
npx delete-webhook
If you prefer to use long polling method over creating a server with webhook you can use the startLongPolling
method instead of createServer
. If you have a webhook set up already, you need to delete it, you can use the npx delete-webhook
command to delete it.
The method accepts `pollingDelay` - a number that represents milliseconds (must be at least 50ms).
By default the long polling mechanism will ignore all previous updates (`cleanPreviousUpdates` by default is `true`). You can toggle off `cleanPreviousUpdates` if you want to by setting it as `false` when you call `startLongPolling`.
You also can stop the long polling while it's running by setting `bot.useLongPolling` to `false`.
You can view an example of long polling usage [here](https://github.com/NivEz/telenode/tree/main/examples/long-polling.js).
Note that long polling is usually not recommended and webhook is preferred for most use cases.
const Telenode = require('telenode-js');
require('dotenv').config();
const bot = new Telenode({
apiToken: process.env.API_TOKEN,
});
bot.createServer(); // spins up an express server
bot.onTextMessage('hello', async (messageBody) => {
console.log(messageBody);
await bot.sendTextMessage('hello back', messageBody.chat.id);
});
In this example the bot will listen only to 'hello' text messages and will respond to the user 'hello back'. Any other message will be ignored.
- Note that
bot.createServer()
method requiresexpress
, and we are usingdotenev
as well which both are not installed automatically withTelenode
.
Additional examples can be found in the examples folder.
You can secure your webhook with a secret token via the setWebhook
method. You can do that by creating
a SECRET_TOKEN
variable in the .env
file of your project (or environment variable) and run the npx set-webhook
command. The command will
tell Telegram servers to send the secret token in each request to your webhook as x-telegram-bot-api-secret-token
header.
In order for the bot to use the secret token you need to pass to the Telenode
class you instanciate the secretToken
parameter.
You will have to pass a secretToken
parameter to the telenodeHandler
method as well.
You can pass a third parameter called unauthorizedCallback
- a callback that will fire in case the request wasn't
authorized.
You can find the example in the secretToken.js example and the implementation in src/server.js as well.
After you have set your webhook you can play and test the Telenode
features.
Each feature of Telenode
is demonstrated in an example file inside the examples
folder (inside node_modules
if you
installed Telenode
).
You can run an example from the telenode-js
directory inside node_modules
by using the command:
npm run dev --file=<example>
You might need nodemon
and dotenv
installed as dev dependencies to run the examples with the command above.
Since these days it is common to use serverless backend services, you can choose how the bot will work - or
with express
or with the HTTP
engine of the serverless provider.
In order to spin up an express server you should use the command bot.createServer()
- this is useful for deployments
on VMs / containers / on-premise.
You can pass an object as options for createServer
. Currently, it supports port
and unauthorizedCallback
(if you
use secret token) - e.g:
bot.createServer({ port: 4000 }) // the default is 3000
In the other hand, if you want to deploy on serverless backend you need to use bot.telenodeHandler
method and pass to
it the request object.
You will probably have something like this:
functions.https.onCall((req, res) => {
const secretToken = req.headers['x-telegram-bot-api-secret-token'];
bot.telenodeHandler(req.body, secretToken, unauthorizedHandler);
res.end();
});
Note that on serverless you should extract by your own the secretToken
since every serverless service might process
the req
object differently.
If you want to develop a new feature you should create an example file under the examples' folder that demonstrates how to use the feature.
- Direct respond function in message handler without passing chat ID
- Chat ID handlers
- Arguments validations
- Optimize Telegram API requests
- Add extra security with query params token
- Add tests