Scroll is an extensible alternative to Markdown.

Like Markdown, Scroll is a plaintext language that compiles to HTML.

Scroll is minimal to help you reduce your ideas to their essence.

Scroll is extensible to help you use and build new mini languages to fully develop your ideas.

Scroll debuted in 2021 and is now on Version 72, approximately 1 release every 2 weeks. There is no official spec yet but creating a spec is on the roadmap.

This Tutorial will walk you through most keywords in Scroll AND teach you how to build your own Scroll keywords.

A Scroll document

A Scroll document (or "program") is a list of nodes. Every node is one line and every line is one node.

If you put a space at the beginning of a line, that line becomes a child of the preceding line. You have probably seen this indent trick before in languages like Python. But Scroll pushes it to the max.

If you master the indent trick, you master Scroll. But we're getting ahead of ourselves, let's start with the basics.

Basic Node Types

1. The Thought Node (aka "Paragraph" node)

Let's start with the most common node, the thought node. You can think of this node as similar to a p or div tag in HTML. To use this node you can write out the word thought, or use an asterisk *, OR just start any text that does not begin with a keyword. This paragraph you are reading is written using one of these nodes. The code is:

Let's start with the most common node, the `thought` node. You can think of this node as similar to a `p` or `div` tag in HTML. To use this node you can write out the word `thought`, or use an asterisk `*`, OR just start any text that does not begin with a keyword.  This paragraph you are reading is written using one of these nodes. The code is:

2. Headers

There are a few types of headers in scroll. Let's show the 3 main ones and what they generate:

title This is a title
# This is a section header
## This is a subsection header

This is a title

This is a section header

This is a subsection header

3. Unordered lists

Here's how you write unordered lists:

- Scroll has lists
 - That can be nested

Scroll has lists

That can be nested

4. Checklists

Below is the code for a checklist and its rendered version:

[] Finish full tutorial
 [x] Learn that checklists support nesting

Finish full tutorial

Learn that checklists support nesting

5. Tables

Use the table keyword and include your delimiter for tables:

table ,
 Name,Rank
 Scroll,#1
 Markdown,#2

Name	Rank
Scroll	#1
Markdown	#2

6. Images

To add an image use the image keyword:

image https://scroll.pub/blog/screenshot.png
 caption An image with a caption

7. Footnotes

You can make footnotes like this:

Pau means done^pau
^pau In Hawaiian

Pau means done¹

1 In Hawaiian ⮐

8. Dashboard

If you are building a dashboard you might want to try the dashboard keyword:

#1Lang

2kUsers

300Stars

9. Including HTML and CSS

If you need to jump into regular HTML, use the html keyword. html If you need to jump into regular HTML, use the <code>html</code> keyword.

For CSS, use the css keyword:

css .green {color: green;}
This text should be green.
 class green

This text should be green.

Marking Up Text

Bold, italics and code

Formatting text is similar to Markdown or Textile.

Here's how to *bold*, _italicize_, or denote `code`.

Here's how to bold, italicize, or denote code.

Links

Scroll does links different. Instead of mixing in the link with the content, you put the link after the text along with the text you want the link to match against. For example:

A link to Wikipedia
 https://wikipedia.org Wikipedia

A link to Wikipedia

You can also make the whole paragraph a link by not including any text to match against.

A link to Wikipedia
 https://wikipedia.org

A link to Wikipedia

Columns

You can use the startColumns [maxNumberOfColumns] keyword to start a columns flow and endColumns to end a columns flow. If you don't want a section to break across columns, don't put line breaks in between lines. Line breaks will clear sections.

Advanced

Variables

Use the replace keyword to define variables. Variables definitions are parsed and removed on the first compiler pass.

Our domain is: scroll.pub

Import statements

Scroll files can import other Scroll files. Use the import keyword followed by the path to the file, such as: import header.scroll

Advanced: Custom Themes

A standard Scroll Theme consists of CSS and may also contain suggested pages and additional Grammar extensions. The default scroll theme is called Gazette. The checklist below walks you through creating a complete Scroll Theme like Gazette.

Create a [yourThemeName].grammar file

Add a [yourThemeName]CssParser

You can refer to the gazetteCssParser in gazetteTheme.grammar for an example. Your theme can extend abstractThemeCssParser or be its own tag.

Add CSS. The Gazette theme uses Hakon, a language that compiles to CSS, but you are free to use plain CSS—even in an external file.

Add CSS resets. Like any other normal site you probably want to adjust the base CSS of the browser.

Add any custom styles for any custom keywords you include in your theme. If you don't have any custom keywords you can skip this step.

Add CSS for the standard Scroll keywords. You probably want to customize the styles of Scroll keywords like * and #. There is not currently an exhaustive list or consistent pattern to these class names. See the Gazette theme for reference.

Using Custom Themes

By default Scroll emits HTML with no theming. To use any theme—including the default Gazette theme—the user simply uses the keywords provided by the theme, such as gazetteCss.

Examples

A page using no theme:

# This page has no theme

A page using a theme:

gazetteCss
pageHeader
* This page uses the Gazette theme.
pageFooter

Expert: Adding your own parsers

Scroll is based on the theory that a language should adapt to the domain, not the other way around. So Scroll has extendibility built-in.

Note: Custom Parsers are currently only supported using the npm package. The web editor does not currently support custom keywords.

You can define your own parsers right in your Scroll documents using *Parser. These Parsers are written in the Grammar language.

Here is a simple example that extends Scroll by making p work the same as *:

pParser
 extends thoughtParser
 crux p
p We can then make paragraphs using `p`.

We can then make paragraphs using p.

Let's now make a hiddenMessage Parser that alerts a message when clicked:

messageParser
 cruxFromId
 catchAllCellType stringCell
hiddenMessageParser
 extends thoughtParser
 inScope messageParser
 cruxFromId
 javascript
  compile() {
    return `<span onclick="alert('${this.get('message')}')">${super.compile()}</span>`
  }
hiddenMessage Click me.
 message Hello world

Click me.

As you can see, you can define new parsers with a small amount of code. You probably also can see that the Grammar Langauge is powerful but has lots of sharp edges. While the documentation on Grammar evolves, feel free to get in touch for help in adding your own parsers.

Scroll Tutorial

Edit this in the Scroll Web Editor