Add superscript and subscript extensions

subscript/go.mod

@@ -0,0 +1,5 @@
+module github.com/gohugoio/hugo-goldmark-extensions/subscript
+
+go 1.21.6
+
+require github.com/yuin/goldmark v1.7.0

subscript/go.sum

@@ -0,0 +1,2 @@
+github.com/yuin/goldmark v1.7.0 h1:EfOIvIMZIzHdB/R/zVrikYLPPwJlfMcNczJFMs1m6sA=
+github.com/yuin/goldmark v1.7.0/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=

subscript/subscript.go

@@ -0,0 +1,146 @@
+package subscript
+
+import (
+	"github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/ast"
+	"github.com/yuin/goldmark/parser"
+	"github.com/yuin/goldmark/renderer"
+	"github.com/yuin/goldmark/renderer/html"
+	"github.com/yuin/goldmark/text"
+	"github.com/yuin/goldmark/util"
+)
+
+// A Subscript struct represents a subscript text segment.
+type Subscript struct {
+	ast.BaseInline
+}
+
+// Dump implements Node.Dump.
+func (n *Subscript) Dump(source []byte, level int) {
+	ast.DumpHelper(n, source, level, nil, nil)
+}
+
+// Kind is a NodeKind of the Subscript node.
+var Kind = ast.NewNodeKind("Subscript")
+
+// Kind implements Node.Kind.
+func (n *Subscript) Kind() ast.NodeKind {
+	return Kind
+}
+
+// New returns a new Subscript node.
+func New() *Subscript {
+	return &Subscript{}
+}
+
+type delimiterProcessor struct {
+}
+
+func (p *delimiterProcessor) IsDelimiter(b byte) bool {
+	return b == '~'
+}
+
+func (p *delimiterProcessor) CanOpenCloser(opener, closer *parser.Delimiter) bool {
+	return opener.Char == closer.Char
+}
+
+func (p *delimiterProcessor) OnMatch(int) ast.Node {
+	return New()
+}
+
+var defaultDelimiterProcessor = &delimiterProcessor{}
+
+type Parser struct {
+}
+
+var defaultParser = &Parser{}
+
+// NewParser returns a new InlineParser that parses subscript expressions.
+func NewParser() parser.InlineParser {
+	return defaultParser
+}
+
+func (s *Parser) Trigger() []byte {
+	return []byte{'~'}
+}
+
+func (s *Parser) Parse(_ ast.Node, block text.Reader, pc parser.Context) ast.Node {
+	before := block.PrecendingCharacter()
+	line, segment := block.PeekLine()
+	node := parser.ScanDelimiter(line, before, 1, defaultDelimiterProcessor)
+	if node == nil {
+		return nil
+	}
+	if node.CanOpen {
+		for i := 1; i < len(line); i++ {
+			c := line[i]
+			if c == line[0] { // Found closing match
+				break
+			}
+			if util.IsSpace(c) {
+				return nil
+			} // No ordinary whitespaces allowed
+		}
+	}
+	node.Segment = segment.WithStop(segment.Start + node.OriginalLength)
+	block.Advance(node.OriginalLength)
+	pc.PushDelimiter(node)
+	return node
+}
+
+func (s *Parser) CloseBlock() {
+	// nothing to do
+}
+
+// HTMLRenderer is a renderer.NodeRenderer implementation that renders
+// Subscript nodes.
+type HTMLRenderer struct {
+	html.Config
+}
+
+// NewHTMLRenderer returns a new HTMLRenderer.
+func NewHTMLRenderer(opts ...html.Option) renderer.NodeRenderer {
+	r := &HTMLRenderer{
+		Config: html.NewConfig(),
+	}
+	for _, opt := range opts {
+		opt.SetHTMLOption(&r.Config)
+	}
+	return r
+}
+
+// RegisterFuncs implements renderer.NodeRenderer.RegisterFuncs.
+func (r *HTMLRenderer) RegisterFuncs(reg renderer.NodeRendererFuncRegisterer) {
+	reg.Register(Kind, r.render)
+}
+
+// AttributeFilter defines attribute names which dd elements can have.
+var AttributeFilter = html.GlobalAttributeFilter
+
+func (r *HTMLRenderer) render(
+	w util.BufWriter, _ []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
+	if entering {
+		if n.Attributes() != nil {
+			_, _ = w.WriteString("<sub")
+			html.RenderAttributes(w, n, AttributeFilter)
+			_ = w.WriteByte('>')
+		} else {
+			_, _ = w.WriteString("<sub>")
+		}
+	} else {
+		_, _ = w.WriteString("</sub>")
+	}
+	return ast.WalkContinue, nil
+}
+
+// The Extension allows you to use a subscript expression like 'H~2~O'.
+var Extension = &Subscript{}
+
+func (n *Subscript) Extend(m goldmark.Markdown) {
+	m.Parser().AddOptions(parser.WithInlineParsers(
+		util.Prioritized(NewParser(), 501),
+	))
+	m.Renderer().AddOptions(renderer.WithNodeRenderers(
+		util.Prioritized(NewHTMLRenderer(), 501),
+	))
+}

subscript/subscript_test.go

@@ -0,0 +1,66 @@
+package subscript
+
+import (
+	"bytes"
+	"github.com/yuin/goldmark/extension"
+	"github.com/yuin/goldmark/text"
+	"testing"
+
+	"github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/testutil"
+)
+
+func buildTestParser() goldmark.Markdown {
+	markdown := goldmark.New(
+		goldmark.WithExtensions(
+			Extension, extension.Strikethrough,
+		),
+	)
+	return markdown
+}
+
+func Test(t *testing.T) {
+	markdown := buildTestParser()
+	testutil.DoTestCaseFile(markdown, "testCases.txt", t, testutil.ParseCliCaseArg()...)
+}
+
+func TestDump(t *testing.T) {
+	input := "**H~2~O** or simply water."
+	markdown := goldmark.New(
+		goldmark.WithExtensions(
+			Extension,
+		),
+	)
+	root := markdown.Parser().Parse(text.NewReader([]byte(input)))
+	root.Dump([]byte(input), 0)
+	// Prints to stdout, so just test that it doesn't crash
+}
+
+func BenchmarkWithAndWithoutOneSubscript(b *testing.B) {
+	const input = `
+## Water formula
+
+The chemical formula for water H~2~O contains one subscript.`
+
+	b.Run("without subscript", func(b *testing.B) {
+		markdown := goldmark.New()
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			var buf bytes.Buffer
+			if err := markdown.Convert([]byte(input), &buf); err != nil {
+				b.Fatal(err)
+			}
+		}
+	})
+
+	b.Run("with subscript", func(b *testing.B) {
+		markdown := buildTestParser()
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			var buf bytes.Buffer
+			if err := markdown.Convert([]byte(input), &buf); err != nil {
+				b.Fatal(err)
+			}
+		}
+	})
+}

subscript/testCases.txt

@@ -0,0 +1,56 @@
+1
+//- - - - - - - - -//
+~foo~
+//- - - - - - - - -//
+<p><sub>foo</sub></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+2
+//- - - - - - - - -//
+H~2~O
+//- - - - - - - - -//
+<p>H<sub>2</sub>O</p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+3
+//- - - - - - - - -//
+x~i~ + x~j~
+//- - - - - - - - -//
+<p>x<sub>i</sub> + x<sub>j</sub></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+4
+//- - - - - - - - -//
+a~4 + a~5
+//- - - - - - - - -//
+<p>a~4 + a~5</p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+5
+//- - - - - - - - -//
+~foo\~
+//- - - - - - - - -//
+<p>~foo~</p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+6
+//- - - - - - - - -//
+~foo&nbsp;bar~
+//- - - - - - - - -//
+<p><sub>foo bar</sub></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+7
+//- - - - - - - - -//
+~foo bar~
+//- - - - - - - - -//
+<p><sub>foo bar</sub></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+
+8
+//- - - - - - - - -//
+~~x~foobar~~~
+//- - - - - - - - -//
+<p><del>x<sub>foobar</sub></del></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//

superscript/go.mod

@@ -0,0 +1,5 @@
+module github.com/gohugoio/hugo-goldmark-extensions/superscript
+
+go 1.21.6
+
+require github.com/yuin/goldmark v1.7.0

superscript/go.sum

@@ -0,0 +1,2 @@
+github.com/yuin/goldmark v1.7.0 h1:EfOIvIMZIzHdB/R/zVrikYLPPwJlfMcNczJFMs1m6sA=
+github.com/yuin/goldmark v1.7.0/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=