Skip to content

Converts Markdown files and strings into NSAttributedStrings with lots of customisation options.

License

Notifications You must be signed in to change notification settings

SimonFairbairn/SwiftyMarkdown

Repository files navigation

SwiftyMarkdown 1.0

SwiftyMarkdown converts Markdown files and strings into NSAttributedStrings using sensible defaults and a Swift-style syntax. It uses dynamic type to set the font size correctly with whatever font you'd like to use.

Fully Rebuilt For 2020!

SwiftyMarkdown now features a more robust and reliable rules-based line processing and character tokenisation engine. It has added support for images stored in the bundle (![Image](<Name In bundle>)), codeblocks, blockquotes, and unordered lists!

Line-level attributes can now have a paragraph alignment applied to them (e.g. h2.aligment = .center), and links can be optionally underlined by setting underlineLinks to true.

It also uses the system color .label as the default font color on iOS 13 and above for Dark Mode support out of the box.

Support for all of Apple's platforms has been enabled.

Installation

CocoaPods:

pod 'SwiftyMarkdown'

SPM:

In Xcode, File -> Swift Packages -> Add Package Dependency and add the GitHub URL.

How To Use SwiftyMarkdown

Read Markdown from a text string...

let md = SwiftyMarkdown(string: "# Heading\nMy *Markdown* string")
md.attributedString()

...or from a URL.

if let url = Bundle.main.url(forResource: "file", withExtension: "md"), md = SwiftyMarkdown(url: url ) {
	md.attributedString()
}

If you want to use a different string once SwiftyMarkdown has been initialised, you can now do so like this:

let md = SwiftyMarkdown(string: "# Heading\nMy *Markdown* string")
md.attributedString(from: "A **SECOND** Markdown string. *Fancy!*")

The attributed string can then be assigned to any label or text control that has support for attributed text.

let md = SwiftyMarkdown(string: "# Heading\nMy *Markdown* string")
let label = UILabel()
label.attributedText = md.attributedString()

Supported Markdown Features

*italics* or _italics_
**bold** or __bold__
~~Linethrough~~Strikethroughs. 
`code`

# Header 1

or

Header 1
====

## Header 2

or

Header 2
---

### Header 3
#### Header 4
##### Header 5 #####
###### Header 6 ######

	Indented code blocks (spaces or tabs)

[Links](http://voyagetravelapps.com/)
![Images](<Name of asset in bundle>)

[Referenced Links][1]
![Referenced Images][2]

[1]: http://voyagetravelapps.com/
[2]: <Name of asset in bundle>

> Blockquotes

- Bulleted
- Lists
	- Including indented lists
		- Up to three levels
- Neat!

1. Ordered
1. Lists
	1. Including indented lists
		- Up to three levels

Compound rules also work, for example:

It recognises **[Bold Links](http://voyagetravelapps.com/)**

Or [**Bold Links**](http://voyagetravelapps.com/)

Images will be inserted into the returned NSAttributedString as an NSTextAttachment (sadly, this will not work on watchOS as NSTextAttachment is not available).

Customisation

Set the attributes of every paragraph and character style type using straightforward dot syntax:

md.body.fontName = "AvenirNextCondensed-Medium"

md.h1.color = UIColor.redColor()
md.h1.fontName = "AvenirNextCondensed-Bold"
md.h1.fontSize = 16
md.h1.alignmnent = .center

md.italic.color = UIColor.blueColor()

md.underlineLinks = true

md.bullet = "🍏"

On iOS, Specified font sizes will be adjusted relative to the the user's dynamic type settings.

Screenshot

Screenshot

There's an example project included in the repository. Open the Example/SwiftyMarkdown.xcodeproj file to get started.

Front Matter

SwiftyMarkdown recognises YAML front matter and will populate the frontMatterAttributes property with the key-value pairs that it fines.

Appendix

A) All Customisable Properties

h1.fontName : String
h1.fontSize : CGFloat
h1.color : UI/NSColor
h1.fontStyle : FontStyle
h1.alignment : NSTextAlignment

h2.fontName : String
h2.fontSize : CGFloat
h2.color : UI/NSColor
h2.fontStyle : FontStyle
h2.alignment : NSTextAlignment

h3.fontName : String
h3.fontSize : CGFloat
h3.color : UI/NSColor
h3.fontStyle : FontStyle
h3.alignment : NSTextAlignment

h4.fontName : String
h4.fontSize : CGFloat
h4.color : UI/NSColor
h4.fontStyle : FontStyle
h4.alignment : NSTextAlignment

h5.fontName : String
h5.fontSize : CGFloat
h5.color : UI/NSColor
h5.fontStyle : FontStyle
h5.alignment : NSTextAlignment

h6.fontName : String
h6.fontSize : CGFloat
h6.color : UI/NSColor
h6.fontStyle : FontStyle
h6.alignment : NSTextAlignment

body.fontName : String
body.fontSize : CGFloat
body.color : UI/NSColor
body.fontStyle : FontStyle
body.alignment : NSTextAlignment

blockquotes.fontName : String
blockquotes.fontSize : CGFloat
blockquotes.color : UI/NSColor
blockquotes.fontStyle : FontStyle
blockquotes.alignment : NSTextAlignment

link.fontName : String
link.fontSize : CGFloat
link.color : UI/NSColor
link.fontStyle : FontStyle

bold.fontName : String
bold.fontSize : CGFloat
bold.color : UI/NSColor
bold.fontStyle : FontStyle

italic.fontName : String
italic.fontSize : CGFloat
italic.color : UI/NSColor
italic.fontStyle : FontStyle

code.fontName : String
code.fontSize : CGFloat
code.color : UI/NSColor
code.fontStyle : FontStyle

strikethrough.fontName : String
strikethrough.fontSize : CGFloat
strikethrough.color : UI/NSColor
strikethrough.fontStyle : FontStyle

underlineLinks : Bool

bullet : String

FontStyle is an enum with these cases: normal, bold, italic, and bolditalic to give you more precise control over how lines and character styles should look. For example, perhaps you want blockquotes to default to having the italic style:

md.blockquotes.fontStyle = .italic

Or, if you like a bit of chaos:

md.bold.fontStyle = .italic
md.italic.fontStyle = .bold

B) Advanced Customisation

SwiftyMarkdown uses a rules-based line processing and customisation engine that is no longer limited to Markdown. Rules are processed in order, from top to bottom. Line processing happens first, then character styles are applied based on the character rules.

For example, here's how a small subset of Markdown line tags are set up within SwiftyMarkdown:

enum MarkdownLineStyle : LineStyling {
	case h1
	case h2
	case previousH1
	case codeblock
	case body
	
	var shouldTokeniseLine: Bool {
		switch self {
		case .codeblock:
			return false
		default:
			return true
		}
	}
	
	func styleIfFoundStyleAffectsPreviousLine() -> LineStyling? {
		switch self {
		case .previousH1:
			return MarkdownLineStyle.h1
		default :
			return nil
		}
	}
}

static public var lineRules = [
	LineRule(token: "    ",type : MarkdownLineStyle.codeblock, removeFrom: .leading),
	LineRule(token: "=",type : MarkdownLineStyle.previousH1, removeFrom: .entireLine, changeAppliesTo: .previous),
	LineRule(token: "## ",type : MarkdownLineStyle.h2, removeFrom: .both),
	LineRule(token: "# ",type : MarkdownLineStyle.h1, removeFrom: .both)
]

let lineProcessor = SwiftyLineProcessor(rules: SwiftyMarkdown.lineRules, default: MarkdownLineStyle.body)

Similarly, the character styles all follow rules:

enum CharacterStyle : CharacterStyling {
	case link, bold, italic, code
}

static public var characterRules = [
    CharacterRule(primaryTag: CharacterRuleTag(tag: "[", type: .open), otherTags: [
			CharacterRuleTag(tag: "]", type: .close),
			CharacterRuleTag(tag: "[", type: .metadataOpen),
			CharacterRuleTag(tag: "]", type: .metadataClose)
	], styles: [1 : CharacterStyle.link], metadataLookup: true, definesBoundary: true),
	CharacterRule(primaryTag: CharacterRuleTag(tag: "`", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.code], shouldCancelRemainingTags: true, balancedTags: true),
	CharacterRule(primaryTag: CharacterRuleTag(tag: "*", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.italic, 2 : CharacterStyle.bold], minTags:1 , maxTags:2),
	CharacterRule(primaryTag: CharacterRuleTag(tag: "_", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.italic, 2 : CharacterStyle.bold], minTags:1 , maxTags:2)
]

These Character Rules are defined by SwiftyMarkdown:

public struct CharacterRule : CustomStringConvertible {

	public let primaryTag : CharacterRuleTag
	public let tags : [CharacterRuleTag]
	public let escapeCharacters : [Character]
	public let styles : [Int : CharacterStyling]
	public let minTags : Int
	public let maxTags : Int
	public var metadataLookup : Bool = false
	public var definesBoundary = false
	public var shouldCancelRemainingRules = false
	public var balancedTags = false
}
  1. primaryTag: Each rule must have at least one tag and it can be one of repeating, open, close, metadataOpen, or metadataClose. repeating tags are tags that have identical open and close characters (and often have more than 1 style depending on how many are in a group). For example, the * tag used in Markdown.
  2. tags: An array of other tags that the rule can look for. This is where you would put the close tag for a custom rule, for example.
  3. escapeCharacters: The characters that appear prior to any of the tag characters that tell the scanner to ignore the tag.
  4. styles: The styles that should be applied to every character between the opening and closing tags.
  5. minTags: The minimum number of repeating characters to be considered a successful match. For example, setting the primaryTag to * and the minTag to 2 would mean that **foo** would be a successful match wheras *bar* would not.
  6. maxTags: The maximum number of repeating characters to be considered a successful match.
  7. metadataLookup: Used for Markdown reference links. Tells the scanner to try to look up the metadata from this dictionary, rather than from the inline result.
  8. definesBoundary: In order for open and close tags to be successful, the boundaryCount for a given location in the string needs to be the same. Setting this property to true means that this rule will increase the boundaryCount for every character between its opening and closing tags. For example, the [ rule defines a boundary. After it is applied, the string *foo[bar*] becomes *foobar* with a boundaryCount 00001111. Applying the * rule results in the output *foobar* because the opening * tag and the closing * tag now have different boundaryCount values. It's basically a way to fix the **[should not be bold**](url) problem in Markdown.
  9. shouldCancelRemainingTags: A successful match will mark every character between the opening and closing tags as complete, thereby preventing any further rules from being applied to those characters.
  10. balancedTags: This flag requires that the opening and closing tags be of exactly equal length. E.g. If this is set to true, **foo* would result in **foo*. If it was false, the output would be *foo.

Rule Subsets

If you want to only support a small subset of Markdown, it's now easy to do.

This example would only process strings with * and _ characters, ignoring links, images, code, and all line-level attributes (headings, blockquotes, etc.)

SwiftyMarkdown.lineRules = []

SwiftyMarkdown.characterRules = [
	CharacterRule(primaryTag: CharacterRuleTag(tag: "*", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.italic, 2 : CharacterStyle.bold], minTags:1 , maxTags:2),
	CharacterRule(primaryTag: CharacterRuleTag(tag: "_", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.italic, 2 : CharacterStyle.bold], minTags:1 , maxTags:2)
]

Custom Rules

If you wanted to create a rule that applied a style of Elf to a range of characters between "The elf will speak now: %Here is my elf speaking%", you could set things up like this:

enum Characters : CharacterStyling {
	case elf

	func isEqualTo( _ other : CharacterStyling) -> Bool {
		if let other = other as? Characters else {
			return false
		}
		return other == self
	}
}

let characterRules = [
	CharacterRule(primaryTag: CharacterRuleTag(tag: "%", type: .repeating), otherTags: [], styles: [1 : CharacterStyle.elf])
]

let processor = SwiftyTokeniser( with : characterRules )
let string = "The elf will speak now: %Here is my elf speaking%"
let tokens = processor.process(string)

The output is an array of tokens would be equivalent to:

[
	Token(type: .string, inputString: "The elf will speak now: ", characterStyles: []),
	Token(type: .repeatingTag, inputString: "%", characterStyles: []),
	Token(type: .string, inputString: "Here is my elf speaking", characterStyles: [.elf]),
	Token(type: .repeatingTag, inputString: "%", characterStyles: [])
]

C) SpriteKit Support

Did you know that SKLabelNode supports attributed text? I didn't.

let smd = SwiftyMarkdown(string: "My Character's **Dialogue**")

let label = SKLabelNode()
label.preferredMaxLayoutWidth = 500
label.numberOfLines = 0
label.attributedText = smd.attributedString()