telegraf 4.16.0

v4.16.0

on Node.js NPM

• Support for API 7.0. Highlights are Reactions, Replies 2.0, Link Previews, Blockquotes, and Chat Boosts.
• Support for API 7.1.
• All methods and update types from these API versions are now fully supported.

• To listen on reaction addition and removal, use Composer.reaction:

  bot.reaction("👍", (ctx) => {
    // user added a 👍 reaction
  });
  
  // prefix with - to listen to reaction removal
  bot.reaction("-👍", (ctx) => {
    // user removed a 👍 reaction
  });

  This also just works with custom emoji IDs.

  bot.reaction("5368742036629364794", (ctx) => {
    // user added a reaction with the given custom emoji ID
  });
  
  bot.reaction("-5368742036629364794", (ctx) => {
    // user removed a reaction with the given custom emoji ID
  });

• You can probe and inspect the reaction list with the ctx.reactions smart object:

  bot.on("message_reaction", async (ctx) => {
    // remember that ctx.reactions is a smart object, but not an array
  
    // message has a 👍 reaction
    ctx.reactions.has("👍");
  
    // message has a reaction with the given custom emoji ID
    ctx.reactions.has("5368742036629364794");
  
    // number of reactions from this user on the message
    ctx.reaction.count;
  
    // indexed access is allowed:
    const first = ctx.reactions[0];
  
    // the 👍 emoji was added in this update
    if (ctx.reactions.added.has("👍")) {
      // user added a 👍 reaction
      await User.updateOne({ id: ctx.from.id }, { $inc: { likes: 1 } });
    }
  
    // the 👍 emoji was removed in this update
    if (ctx.reactions.removed.has("👍")) {
      // user removed a 👍 reaction
      await User.updateOne({ id: ctx.from.id }, { $inc: { likes: -1 } });
    };
  
    // to copy any of these smart objects into an array, call the `toArray` method
    const reactions = ctx.reactions.toArray();
    const added = ctx.reactions.added.toArray();
    const removed = ctx.reactions.removed.toArray();
  });

• To react to a message, use ctx.react:

  bot.on("text", async (ctx) => {
    await ctx.react("👍");
  });

  You can also react to a message with a custom emoji ID:

  bot.on("text", async (ctx) => {
    await ctx.react("5368742036629364794");
  });

  The bot.telegram.setMessageReaction method is also available.

• Added the ctx.entities() method to fetch entities in any message.

  bot.on("text", (ctx) => {
    // fetch all entities
    const entities = ctx.entities();
    // fetch all command entities
    const commandEntities = ctx.entities("bot_command");
    // fetch all mentions and text mentions
    const mentions = ctx.entities("mention", "text_mention");
  });

  Not only does this method fetch entities from any message, but also works with captions, game text, and poll explanations. In short, if an update has any text and entities, this method will find them.

• ctx.text is the companion to ctx.entities(). It fetches the text from the update, and works with any message type. This includes message text, media captions, game text, and poll explanations.

  bot.on(message("text"), (ctx) => {
    // fetch the text from the update
    const text = ctx.text;
  });
  
  bot.on(message("photo"), (ctx) => {
    // fetch the caption from the photo
    const caption = ctx.text;
  });

• 🎁 Bonus! Every entity in the ctx.entities() array will have a fragment of the text they represent!

  bot.on("text", (ctx) => {
    const entities = ctx.entities("bold");
    for (const entity of entities) {
      // the text that is bold
      const boldText = entity.fragment;
    }
  });

• Context::msg shorthand to get any message in the update.

  bot.use((ctx) => {
    // finds one of:
    // ctx.message ?? ctx.editedMessage ?? ctx.callbackQuery?.message ?? ctx.channelPost ?? ctx.editedChannelPost
    const msg = ctx.msg;
  });

  ctx.msg is decorated with the isAccessible and has methods. The has method works similarly to the message() filter in bot.on. It checks if the message has a certain property.

  if (ctx.msg.isAccessible()) {
    // msg is accessible, not deleted or otherwise unavailable
    // this is a type-guard based on the runtime check for msg.date === 0
  }
  
  if (ctx.msg.has("text")) {
    // ctx.msg.text exists
  }

• Context::msgId shorthand to get any available message ID in the update. This also includes message_id present in updates that do not contain a message, such as message_reaction, and message_reaction_count.

  bot.use((ctx) => {
    // finds one of:
    // ctx.msg.message_id ?? ctx.messageReaction.message_id ?? ctx.messageReactionCount.message_id
    const msgId = ctx.msgId;
  });

• bot.launch now takes an optional callback that is called when the bot is launched.

  bot.launch(() => console.log("Bot is starting!"));

  If you pass LaunchOptions, the callback goes after the options.

  bot.launch({ dropPendingUpdates: true }, () => console.log("Bot is starting!"));

  This is useful for running some code when the bot is launched, such as logging to the console, or sending a message to a channel. Remember that errors thrown in this callback will not be caught by the bot's error handler. You must handle them yourself.

  It's worth noting that the callback is called once the first getMe call is successful. This means network is working, and bot token is valid. Due to how networks work, there isn't a way to define when the bot is "fully" launched. The bot may still crash after the callback is called, for example if another instance of the bot is running elsewhere and polling getUpdates fails. For these reasons, onLaunch callback exists under @experimental, and may receive improvements based on feedback.

• Added blockquote format helper.
• Stricter types for format helpers. For example, it will now be a type-error to try to nest pre within another entity.

• Added ctx.match for Composer.command (#1938).
• Fixed thumbnail uploads in media groups (#1947).
• The shorthand ctx.chat now includes chat from this.messageReaction ?? this.messageReactionCount ?? this.removedChatBoost.
• The shorthand ctx.from now includes the field user from ctx.messageReaction ?? ctx.pollAnswer ?? ctx.chatBoost?.boost.source, in addition to fetching from in other updates.
• useNewReplies uses ctx.msgId instead of ctx.message.message_id to reply to messages, which works for more update types than before.
• The following modules are now directly importable: types, scenes, filters, format, future, utils, markup, session. For example, via import { WizardScene } from "telegraf/scenes". This was previously available in v3, and was removed in v4.

• (Breaking) Markup.button.userRequest will take extra instead of user_is_premium as the third parameter.
• (Breaking) Markup.button.botRequest will take extra before hide as the third parameter.
• (Types breaking) reply_to_message_id and allow_sending_without_reply replaced by reply_parameters.
• (Types breaking) disable_web_page_preview and link_preview_options replaced by link_preview_options.

Telegraf v4 - Last Major Update

This will be the last major update for Telegraf v4.

What to Expect

If you are currently using Telegraf v4, you can continue using it as you have been. Telegraf v4 will be supported until February 2025, with the following commitments:

• Bug fixes and security updates will still be released for Telegraf v4.
• The existing documentation for Telegraf v4 will remain available.
• New API updates will only focus on ensuring compatibility with the latest Telegram Bot API. No new convenience features will be added to Telegraf v4.

Introducing Telegraf v5

In the coming weeks, we plan to release Telegraf v5. v5 will bring a revamped API, new functionalities, numerous convenient helpers, an improved approach to handling updates, and enhanced documentation.

One of the key highlights of Telegraf v5 is its platform-agnostic nature, allowing you to run it on any JavaScript runtime environment, including Deno, Bun, edge runtimes (such as Cloudflare Workers and Vercel), browsers, and more.

Smooth Transition to v5

If you have closely followed the development of v4 in the past year and stayed updated with deprecation notices, the transition to v5 will be relatively seamless for you. For those still using v4, we will provide a comprehensive migration guide to assist you in upgrading to v5. Stay tuned for more information on the release of v5!