Markdown to Telegram-HTML n8n Node

This repository contains a custom n8n node that converts Markdown into Telegram-compatible HTML.
It preserves Telegram-specific features (like spoilers using ||spoiler||) and formats standard Markdown into HTML that Telegram accepts.

Initial version based on andresberrios/n8n-nodes-telegram-better-markdown.

---

📘 Overview

This node is designed to:

• Accept a Markdown string and convert it into Telegram-safe HTML.
• Handle Telegram's 4096-character message limit intelligently.
• Allow customization of output field names and overflow strategies.

---

⚙️ Node Properties

Property	Type	Default	Description
Markdown Text	string	—	The Markdown content to convert.
Output Field	string	telegram_html	Field name in the output JSON containing the generated HTML.
Message Limit Strategy	options	truncate	Strategy for handling messages exceeding Telegram's 4096-character limit.
Clean Escaped Characters	boolean	true	Replace literal escape sequences (e.g. "\n") with actual characters (e.g. "\n").
Table Conversion Mode	options	codeBlock	How to display Markdown tables: Monospaced Table, Compact View, Detail View, or Detail View (No Headers).

---

🧩 Message Limit Strategies

Telegram limits messages to 4096 characters.
This node provides two ways to handle longer content:

1. truncate — Truncate Message

Behavior:
The generated HTML is truncated to a safe boundary (avoiding broken tags) and appends [...] to indicate omitted content.

Use Case:
When you only need one message output and can tolerate trimming the end.

---

2. split — Split Message

Behavior:
The HTML is split into multiple parts (each ≤4096 characters).
Splitting preserves HTML integrity and returns multiple output items — one per chunk — with the same original data except for the updated Output Field.

Use Case:
When you want to send the entire message as multiple consecutive Telegram messages without breaking formatting.

---

Notes on Behavior

• The node uses helper functions that tokenize HTML and prefer to cut at safe points (e.g., closed tags or line breaks).
• Open HTML tags are automatically closed before truncation/split boundaries to ensure valid HTML output.
• In split mode, tags are re-opened at the start of each subsequent chunk to maintain formatting continuity (e.g., <b>, <code>, <a href="..."> with attributes preserved).
• In truncate mode, a [...] suffix is appended.
• In split mode, each chunk becomes a separate output item — downstream nodes (like Telegram Send Message) will process each part independently.

🔄 Conversions

This node automatically converts standard Markdown elements into their Telegram-compatible HTML equivalents, including:

Markdown	Telegram HTML Output
**bold** or __bold__	<b>bold</b>
*italic* or _italic_	<i>italic</i>
~~strikethrough~~	<s>strikethrough</s>
__underline__	<u>underline</u>
`inline code`	<code>inline code</code>
```code blocks```	<pre><code>...</code></pre>
> blockquote	<blockquote>...</blockquote>
Lists (-, *, 1.)	Rendered with proper indentation
[text](url)	<a href="url">text</a>
	<tg-emoji emoji-id="12345">🙂</tg-emoji>
Tables (via Table Conversion Mode)	See table modes below
Spoilers (||secret||)	<tg-spoiler>secret</tg-spoiler>
Horizontal rules (---)	------

⚙️ All unsupported Markdown features are safely escaped or ignored to ensure Telegram compatibility.

---

📊 Table Conversion Modes

Tables can be displayed in four different ways using the Table Conversion Mode option in Advanced Fields:

1. Monospaced Table (Default)

Behavior: Classic monospaced table in <pre><code> with pipes and separator line.

Input:

| Name | Age | City |
|------|-----|------|
| Alice | 30 | NYC |
| Bob | 25 | LA |

Output:

<pre><code>| Name | Age | City |
| --- | --- | --- |
| Alice | 30 | NYC |
| Bob | 25 | LA |</code></pre>

Best for: Technical tables, data with alignment requirements, preserving original table structure.

---

2. Compact View

Behavior: Each row as a list item with em dash (—) separator between cells. First cell is bold. Empty cells are skipped.

Output:

- <b>Name</b> — Age — City
- <b>Alice</b> — 30 — NYC
- <b>Bob</b> — 25 — LA

Best for: Quick mobile scanning, simple row data, minimal formatting.

---

3. Detail View

Behavior: Nested list with column headers as labels. First cell is bold with header name. Subsequent cells indented with header names. Empty cells are skipped.

Output:

- <b>Name: Alice</b>
  - Age: 30
  - City: NYC
- <b>Name: Bob</b>
  - Age: 25
  - City: LA

Best for: Key-value presentations, form-like data, detailed information cards.

---

4. Detail View (No Headers)

Behavior: Nested list without header names. First value is bold. Subsequent values indented. Empty cells are skipped.

Output:

- <b>Alice</b>
  - 30
  - NYC
- <b>Bob</b>
  - 25
  - LA

Best for: Clean card layouts, minimal visual hierarchy, product listings.

---

🧪 Example Outputs

truncate Mode

{
  "telegram_html": "<p>This is a long message... [...]</p>"
}

A single output item is returned with truncated HTML.

split Mode

[
  { "telegram_html": "<p>Part 1 of the message...</p>" },
  { "telegram_html": "<p>Part 2 of the message...</p>" }
]

Multiple items are returned, each containing a valid HTML chunk.

---

🧰 Local Development

1. Install Dependencies

npm install

2. Run in Development Mode

npm run dev

3. Run unit test

npm test -- --coverage --runInBand

---

🧠 Usage in n8n

1. Add the Markdown to Telegram-HTML node to your workflow.
2. Provide your Markdown content in the Markdown Text field.
3. Optionally change the Output Field name (default: telegram_html).
4. Choose a Message Limit Strategy:
   • truncate for a single, shortened message.
   • split for multiple message chunks.