feat: implement subscript and superscript parsers

plugin/gomark/ast/ast.go

@@ -30,6 +30,8 @@ const (
 	EscapingCharacterNode
 	MathNode
 	HighlightNode
+	SubscriptNode
+	SuperscriptNode
 )
 
 type Node interface {

plugin/gomark/ast/inline.go

@@ -205,3 +205,31 @@ func (*Highlight) Type() NodeType {
 func (n *Highlight) Restore() string {
 	return fmt.Sprintf("==%s==", n.Content)
 }
+
+type Subscript struct {
+	BaseInline
+
+	Content string
+}
+
+func (*Subscript) Type() NodeType {
+	return SubscriptNode
+}
+
+func (n *Subscript) Restore() string {
+	return fmt.Sprintf("~%s~", n.Content)
+}
+
+type Superscript struct {
+	BaseInline
+
+	Content string
+}
+
+func (*Superscript) Type() NodeType {
+	return SuperscriptNode
+}
+
+func (n *Superscript) Restore() string {
+	return fmt.Sprintf("^%s^", n.Content)
+}

plugin/gomark/parser/parser.go

@@ -83,6 +83,8 @@ var defaultInlineParsers = []InlineParser{
 	NewItalicParser(),
 	NewHighlightParser(),
 	NewCodeParser(),
+	NewSubscriptParser(),
+	NewSuperscriptParser(),
 	NewMathParser(),
 	NewTagParser(),
 	NewStrikethroughParser(),

plugin/gomark/parser/subscript.go

@@ -0,0 +1,53 @@
+package parser
+
+import (
+	"errors"
+
+	"github.com/usememos/memos/plugin/gomark/ast"
+	"github.com/usememos/memos/plugin/gomark/parser/tokenizer"
+)
+
+type SubscriptParser struct{}
+
+func NewSubscriptParser() *SubscriptParser {
+	return &SubscriptParser{}
+}
+
+func (*SubscriptParser) Match(tokens []*tokenizer.Token) (int, bool) {
+	if len(tokens) < 3 {
+		return 0, false
+	}
+	if tokens[0].Type != tokenizer.Tilde {
+		return 0, false
+	}
+
+	contentTokens := []*tokenizer.Token{}
+	matched := false
+	for _, token := range tokens[1:] {
+		if token.Type == tokenizer.Newline {
+			return 0, false
+		}
+		if token.Type == tokenizer.Tilde {
+			matched = true
+			break
+		}
+		contentTokens = append(contentTokens, token)
+	}
+	if !matched || len(contentTokens) == 0 {
+		return 0, false
+	}
+
+	return len(contentTokens) + 2, true
+}
+
+func (p *SubscriptParser) Parse(tokens []*tokenizer.Token) (ast.Node, error) {
+	size, ok := p.Match(tokens)
+	if size == 0 || !ok {
+		return nil, errors.New("not matched")
+	}
+
+	contentTokens := tokens[1 : size-1]
+	return &ast.Subscript{
+		Content: tokenizer.Stringify(contentTokens),
+	}, nil
+}

plugin/gomark/parser/subscript_test.go

@@ -0,0 +1,47 @@
+package parser
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/require"
+
+	"github.com/usememos/memos/plugin/gomark/ast"
+	"github.com/usememos/memos/plugin/gomark/parser/tokenizer"
+	"github.com/usememos/memos/plugin/gomark/restore"
+)
+
+func TestSubscriptParser(t *testing.T) {
+	tests := []struct {
+		text      string
+		subscript ast.Node
+	}{
+		{
+			text:      "~Hello world!",
+			subscript: nil,
+		},
+		{
+			text: "~Hello~",
+			subscript: &ast.Subscript{
+				Content: "Hello",
+			},
+		},
+		{
+			text: "~ Hello ~",
+			subscript: &ast.Subscript{
+				Content: " Hello ",
+			},
+		},
+		{
+			text: "~1~ Hello ~ ~",
+			subscript: &ast.Subscript{
+				Content: "1",
+			},
+		},
+	}
+
+	for _, test := range tests {
+		tokens := tokenizer.Tokenize(test.text)
+		node, _ := NewSubscriptParser().Parse(tokens)
+		require.Equal(t, restore.Restore([]ast.Node{test.subscript}), restore.Restore([]ast.Node{node}))
+	}
+}

plugin/gomark/parser/superscript.go

@@ -0,0 +1,53 @@
+package parser
+
+import (
+	"errors"
+
+	"github.com/usememos/memos/plugin/gomark/ast"
+	"github.com/usememos/memos/plugin/gomark/parser/tokenizer"
+)
+
+type SuperscriptParser struct{}
+
+func NewSuperscriptParser() *SuperscriptParser {
+	return &SuperscriptParser{}
+}
+
+func (*SuperscriptParser) Match(tokens []*tokenizer.Token) (int, bool) {
+	if len(tokens) < 3 {
+		return 0, false
+	}
+	if tokens[0].Type != tokenizer.Caret {
+		return 0, false
+	}
+
+	contentTokens := []*tokenizer.Token{}
+	matched := false
+	for _, token := range tokens[1:] {
+		if token.Type == tokenizer.Newline {
+			return 0, false
+		}
+		if token.Type == tokenizer.Caret {
+			matched = true
+			break
+		}
+		contentTokens = append(contentTokens, token)
+	}
+	if !matched || len(contentTokens) == 0 {
+		return 0, false
+	}
+
+	return len(contentTokens) + 2, true
+}
+
+func (p *SuperscriptParser) Parse(tokens []*tokenizer.Token) (ast.Node, error) {
+	size, ok := p.Match(tokens)
+	if size == 0 || !ok {
+		return nil, errors.New("not matched")
+	}
+
+	contentTokens := tokens[1 : size-1]
+	return &ast.Superscript{
+		Content: tokenizer.Stringify(contentTokens),
+	}, nil
+}

plugin/gomark/parser/superscript_test.go

@@ -0,0 +1,47 @@
+package parser
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/require"
+
+	"github.com/usememos/memos/plugin/gomark/ast"
+	"github.com/usememos/memos/plugin/gomark/parser/tokenizer"
+	"github.com/usememos/memos/plugin/gomark/restore"
+)
+
+func TestSuperscriptParser(t *testing.T) {
+	tests := []struct {
+		text        string
+		superscript ast.Node
+	}{
+		{
+			text:        "^Hello world!",
+			superscript: nil,
+		},
+		{
+			text: "^Hello^",
+			superscript: &ast.Superscript{
+				Content: "Hello",
+			},
+		},
+		{
+			text: "^ Hello ^",
+			superscript: &ast.Superscript{
+				Content: " Hello ",
+			},
+		},
+		{
+			text: "^1^ Hello ^ ^",
+			superscript: &ast.Superscript{
+				Content: "1",
+			},
+		},
+	}
+
+	for _, test := range tests {
+		tokens := tokenizer.Tokenize(test.text)
+		node, _ := NewSuperscriptParser().Parse(tokens)
+		require.Equal(t, restore.Restore([]ast.Node{test.superscript}), restore.Restore([]ast.Node{node}))
+	}
+}

plugin/gomark/parser/tokenizer/tokenizer.go

@@ -22,6 +22,7 @@ const (
 	EqualSign          TokenType = "="
 	Pipe               TokenType = "|"
 	Colon              TokenType = ":"
+	Caret              TokenType = "^"
 	Backslash          TokenType = "\\"
 	Newline            TokenType = "\n"
 	Space              TokenType = " "
@@ -86,6 +87,8 @@ func Tokenize(text string) []*Token {
 			tokens = append(tokens, NewToken(Pipe, "|"))
 		case ':':
 			tokens = append(tokens, NewToken(Colon, ":"))
+		case '^':
+			tokens = append(tokens, NewToken(Caret, "^"))
 		case '\\':
 			tokens = append(tokens, NewToken(Backslash, `\`))
 		case '\n':