Add support for typographer extension

README.md

@@ -39,13 +39,14 @@ log.Print(buf.String()) // # My Document Title
 You can control the style of various markdown elements via functional options that are passed to
 the renderer.
 
-| Functional Option       | Type                         | Description                                                                                                |
-| ----------------------- | ---------------------------- | ---------------------------------------------------------------------------------------------------------- |
-| WithIndentStyle         | markdown.IndentStyle         | Indent nested blocks with spaces or tabs.                                                                  |
-| WithHeadingStyle        | markdown.HeadingStyle        | Render markdown headings as ATX (`#`-based), Setext (underlined with `===` or `---`), or variants thereof. |
-| WithThematicBreakStyle  | markdown.ThematicBreakStyle  | Render thematic breaks with `-`, `*`, or `_`.                                                              |
-| WithThematicBreakLength | markdown.ThematicBreakLength | Number of characters to use in a thematic break (minimum 3).                                               |
-| WithNestedListLength    | markdown.NestedListLength    | Number of characters to use in a nested list indentation (minimum 1).                                      |
+| Functional Option            | Type                              | Description                                                                                                                                                                                                                           |
+| ---------------------------- | --------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------                                                          |
+| WithIndentStyle              | markdown.IndentStyle              | Indent nested blocks with spaces or tabs.                                                                                                                                                                                             |
+| WithHeadingStyle             | markdown.HeadingStyle             | Render markdown headings as ATX (`#`-based), Setext (underlined with `===` or `---`), or variants thereof.                                                                                                                            |
+| WithThematicBreakStyle       | markdown.ThematicBreakStyle       | Render thematic breaks with `-`, `*`, or `_`.                                                                                                                                                                                         |
+| WithThematicBreakLength      | markdown.ThematicBreakLength      | Number of characters to use in a thematic break (minimum 3).                                                                                                                                                                          |
+| WithNestedListLength         | markdown.NestedListLength         | Number of characters to use in a nested list indentation (minimum 1).                                                                                                                                                                 |
+| WithTypographerSubstitutions | markdown.TypographerSubstitutions | Whether characters should be substituted by the typographer extension. This setting has no effect unless the typographer extension is enabled. The renderer must be added as an extension (e.g. via `NewExtension`) for this to work. |
 
 ## As a markdown transformer
 

extension.go

@@ -0,0 +1,63 @@
+package markdown
+
+import (
+	"github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/extension"
+)
+
+type rendererExtension struct{ opts []Option }
+
+// NewExtension returns a new goldmark.Markdown extension that uses the markdown renderer.
+func NewExtension(opts ...Option) *rendererExtension {
+	return &rendererExtension{opts: opts}
+}
+
+// Extend implements goldmark.Extension.Extend
+func (re *rendererExtension) Extend(md goldmark.Markdown) {
+	renderer := NewRenderer(re.opts...)
+	md.SetRenderer(renderer)
+	if renderer.config.TypographerSubstitutions {
+		enableTypographicSubstitutions(md)
+	} else {
+		disableTypographicSubstitutions(md)
+	}
+}
+
+// enableTypographicSubstitutions configures the typographer extension
+// to substitute punctuations with the corresponding unicode character,
+// instead of the default HTML escape sequence.
+// These substitutions are only applied IF the typographer extension is enabled.
+func enableTypographicSubstitutions(md goldmark.Markdown) {
+	subs := make(extension.TypographicSubstitutions)
+	subs[extension.LeftSingleQuote] = []byte("‘")
+	subs[extension.RightSingleQuote] = []byte("’")
+	subs[extension.LeftDoubleQuote] = []byte("“")
+	subs[extension.RightDoubleQuote] = []byte("”")
+	subs[extension.EnDash] = []byte("–")
+	subs[extension.EmDash] = []byte("—")
+	subs[extension.Ellipsis] = []byte("…")
+	subs[extension.LeftAngleQuote] = []byte("«")
+	subs[extension.RightAngleQuote] = []byte("»")
+	subs[extension.Apostrophe] = []byte("'")
+
+	md.Parser().AddOptions(extension.WithTypographicSubstitutions(subs))
+}
+
+// disableTypographicSubstitutions configures the typographer extension
+// to substitute punctuations with their original values,
+// effectively turning typographer into a no-op.
+func disableTypographicSubstitutions(md goldmark.Markdown) {
+	subs := make(extension.TypographicSubstitutions)
+	subs[extension.LeftSingleQuote] = []byte("'")
+	subs[extension.RightSingleQuote] = []byte("'")
+	subs[extension.LeftDoubleQuote] = []byte("\"")
+	subs[extension.RightDoubleQuote] = []byte("\"")
+	subs[extension.EnDash] = []byte("--")
+	subs[extension.EmDash] = []byte("---")
+	subs[extension.Ellipsis] = []byte("...")
+	subs[extension.LeftAngleQuote] = []byte("<<")
+	subs[extension.RightAngleQuote] = []byte(">>")
+	subs[extension.Apostrophe] = []byte("'")
+
+	md.Parser().AddOptions(extension.WithTypographicSubstitutions(subs))
+}

extension_test.go

@@ -0,0 +1,67 @@
+package markdown
+
+import (
+	"bytes"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/extension"
+)
+
+func TestTypographerExtensionDisabled(t *testing.T) {
+	assert := assert.New(t)
+	md := goldmark.New(
+		goldmark.WithExtensions(NewExtension(), extension.NewTypographer()),
+	)
+	buf := bytes.Buffer{}
+	source := `'LeftSingleQuote
+RightSingleQuote'
+"LeftDoubleQuote
+RightDoubleQuote"
+EnDash --
+EmDash ---
+Ellipsis ...
+LeftAngleQuote <<
+RightAngleQuote >>
+Apostrophe 'twas
+`
+
+	err := md.Convert([]byte(source), &buf)
+	assert.NoError(err)
+	assert.Equal(source, buf.String())
+}
+
+func TestTypographerExtensionEnabled(t *testing.T) {
+	assert := assert.New(t)
+	md := goldmark.New(
+		goldmark.WithExtensions(NewExtension(WithTypographerSubstitutions(true)), extension.NewTypographer()),
+	)
+	buf := bytes.Buffer{}
+	source := `'LeftSingleQuote
+RightSingleQuote'
+"LeftDoubleQuote
+RightDoubleQuote"
+EnDash --
+EmDash ---
+Ellipsis ...
+LeftAngleQuote <<
+RightAngleQuote >>
+Apostrophe 'twas
+`
+	expected := `‘LeftSingleQuote
+RightSingleQuote’
+“LeftDoubleQuote
+RightDoubleQuote”
+EnDash –
+EmDash —
+Ellipsis …
+LeftAngleQuote «
+RightAngleQuote »
+Apostrophe 'twas
+`
+
+	err := md.Convert([]byte(source), &buf)
+	assert.NoError(err)
+	assert.Equal(expected, buf.String())
+}

options.go

@@ -11,6 +11,7 @@ type Config struct {
 	ThematicBreakStyle
 	ThematicBreakLength
 	NestedListLength
+	TypographerSubstitutions
 }
 
 // NewConfig returns a new Config with defaults and the given options.
@@ -269,3 +270,35 @@ func WithNestedListLength(style NestedListLength) interface {
 } {
 	return &withNestedListLength{style}
 }
+
+// ============================================================================
+// TypographerSubstitutions Option
+// ============================================================================
+
+// optTypographerSubstitutions is an option name used in WithTypographerSubstitutions
+const optTypographerSubstitutions renderer.OptionName = "TypographerSubstitutions"
+
+// TypographerSubstitutions specifies whether characters should be substituted by the typographer extension.
+type TypographerSubstitutions bool
+
+type withTypographerSubstitutions struct {
+	value TypographerSubstitutions
+}
+
+func (o *withTypographerSubstitutions) SetConfig(c *renderer.Config) {
+	c.Options[optTypographerSubstitutions] = o.value
+}
+
+// SetMarkdownOption implements renderer.Option
+func (o *withTypographerSubstitutions) SetMarkdownOption(c *Config) {
+	c.TypographerSubstitutions = o.value
+}
+
+// WithTypographerSubstitutions is a functional option that determines whether characters should be substituted by the typographer extension.
+// This setting has no effect unless the typographer extension is added to the goldmark.Markdown object.
+func WithTypographerSubstitutions(enabled TypographerSubstitutions) interface {
+	renderer.Option
+	Option
+} {
+	return &withTypographerSubstitutions{enabled}
+}

options_test.go

@@ -4,6 +4,7 @@ import (
 	"testing"
 
 	"github.com/stretchr/testify/assert"
+	"github.com/yuin/goldmark"
 	"github.com/yuin/goldmark/renderer"
 )
 
@@ -27,6 +28,7 @@ func TestRendererOptions(t *testing.T) {
 				WithThematicBreakStyle(ThematicBreakStyleDashed),
 				WithThematicBreakLength(ThematicBreakLengthMinimum),
 				WithNestedListLength(NestedListLengthMinimum),
+				WithTypographerSubstitutions(false),
 			},
 			NewConfig(),
 		},
@@ -50,6 +52,11 @@ func TestRendererOptions(t *testing.T) {
 			r := NewRenderer(tc.options...)
 			assert.Equal(tc.expected, r.config)
 
+			// Set options by passing them to the renderer extension
+			md := goldmark.New(goldmark.WithExtensions(NewExtension(tc.options...)))
+			r = md.Renderer().(*Renderer)
+			assert.Equal(tc.expected, r.config)
+
 			// Set options by name using AddOptions
 			r = NewRenderer()
 			// Convert markdown Option interface to renderer.Option interface

renderer.go

@@ -94,8 +94,7 @@ func (r *Renderer) Render(w io.Writer, source []byte, n ast.Node) error {
 		r.nodeRendererFuncs[ast.KindLink] = r.renderLink
 		r.nodeRendererFuncs[ast.KindRawHTML] = r.renderRawHTML
 		r.nodeRendererFuncs[ast.KindText] = r.renderText
-		// TODO: add KindString
-		// r.nodeRendererFuncs[ast.KindString] = r.renderString
+		r.nodeRendererFuncs[ast.KindString] = r.renderString
 
 		for kind, fun := range r.nodeRendererFuncsTmp {
 			r.nodeRendererFuncs[kind] = r.transform(fun)
@@ -322,6 +321,14 @@ func (r *Renderer) renderText(node ast.Node, entering bool) ast.WalkStatus {
 	return ast.WalkContinue
 }
 
+func (r *Renderer) renderString(node ast.Node, entering bool) ast.WalkStatus {
+	n := node.(*ast.String)
+	if entering {
+		r.rc.writer.WriteBytes(n.Value)
+	}
+	return ast.WalkContinue
+}
+
 func (r *Renderer) renderSegments(segments *text.Segments, asLines bool) {
 	for i := 0; i < segments.Len(); i++ {
 		segment := segments.At(i)