Merge pull request #487 from n-peugnet/patch-1

Fix GitHub actions badge URL

.github/workflows/test.yaml

@@ -5,22 +5,29 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        go-version: [1.18.x, 1.19.x]
+        go-version: [1.21.x, 1.22.x]
         platform: [ubuntu-latest, macos-latest, windows-latest]
     runs-on: ${{ matrix.platform }}
     steps:
     - name: Install Go
-      uses: actions/setup-go@v1
+      uses: actions/setup-go@v4
       with:
         go-version: ${{ matrix.go-version }}
     - name: Checkout code
-      uses: actions/checkout@v1
+      uses: actions/checkout@v3
+    - name: Run lints
+      uses: golangci/golangci-lint-action@v6
+      with:
+        version: latest
+      if: "matrix.platform == 'ubuntu-latest'" # gofmt linter fails on Windows for CRLF problems
     - name: Run tests
+      env:
+        GOLDMARK_TEST_TIMEOUT_MULTIPLIER: 5
       run: go test -v ./... -covermode=count -coverprofile=coverage.out -coverpkg=./...
+    - name: Install goveralls
+      run: go install github.com/mattn/goveralls@latest
     - name: Send coverage
       if: "matrix.platform == 'ubuntu-latest'"
       env:
         COVERALLS_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      run: |
-        GO111MODULE=off go get github.com/mattn/goveralls
-        $(go env GOPATH)/bin/goveralls -coverprofile=coverage.out -service=github
+      run: goveralls -coverprofile=coverage.out -service=github

.golangci.yml

@@ -0,0 +1,105 @@
+run:
+  deadline: 10m
+
+issues:
+  exclude-use-default: false
+  exclude-rules:
+    - path: _test.go
+      linters:
+        - errcheck
+        - lll
+  exclude:
+  - "Package util"
+
+linters:
+  disable-all: true
+  enable:
+    - errcheck
+    - gosimple
+    - govet
+    - ineffassign
+    - staticcheck
+    - typecheck
+    - unused
+    - gofmt
+    - godot
+    - makezero
+    - misspell
+    - revive
+    - wastedassign
+    - lll
+
+linters-settings:
+  revive:
+    severity: "warning"
+    confidence: 0.8
+    rules:
+      - name: blank-imports
+        severity: warning
+        disabled: false
+      - name: context-as-argument
+        severity: warning
+        disabled: false
+      - name: context-keys-type
+        severity: warning
+        disabled: false
+      - name: dot-imports
+        severity: warning
+        disabled: true
+      - name: error-return
+        severity: warning
+        disabled: false
+      - name: error-strings
+        severity: warning
+        disabled: false
+      - name: error-naming
+        severity: warning
+        disabled: false
+      - name: exported
+        severity: warning
+        disabled: false
+      - name: increment-decrement
+        severity: warning
+        disabled: false
+      - name: var-naming
+        severity: warning
+        disabled: false
+      - name: var-declaration
+        severity: warning
+        disabled: false
+      - name: package-comments
+        severity: warning
+        disabled: false
+      - name: range
+        severity: warning
+        disabled: false
+      - name: receiver-naming
+        severity: warning
+        disabled: false
+      - name: time-naming
+        severity: warning
+        disabled: false
+      - name: unexported-return
+        severity: warning
+        disabled: false
+      - name: indent-error-flow
+        severity: warning
+        disabled: false
+      - name: errorf
+        severity: warning
+        disabled: false
+      - name: empty-block
+        severity: warning
+        disabled: true
+      - name: superfluous-else
+        severity: warning
+        disabled: false
+      - name: unused-parameter
+        severity: warning
+        disabled: true
+      - name: unreachable-code
+        severity: warning
+        disabled: false
+      - name: redefines-builtin-id
+        severity: warning
+        disabled: false

Makefile

@@ -1,4 +1,7 @@
-.PHONY: test fuzz
+.PHONY: test fuzz lint
+
+lint:
+	golangci-lint run -c .golangci.yml ./...
 
 test:
 	go test -coverprofile=profile.out -coverpkg=github.com/yuin/goldmark,github.com/yuin/goldmark/ast,github.com/yuin/goldmark/extension,github.com/yuin/goldmark/extension/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 ./...

README.md

@@ -2,13 +2,15 @@ goldmark
 ==========================================
 
 [![https://pkg.go.dev/github.com/yuin/goldmark](https://pkg.go.dev/badge/github.com/yuin/goldmark.svg)](https://pkg.go.dev/github.com/yuin/goldmark)
-[![https://github.com/yuin/goldmark/actions?query=workflow:test](https://github.com/yuin/goldmark/workflows/test/badge.svg?branch=master&event=push)](https://github.com/yuin/goldmark/actions?query=workflow:test)
+[![https://github.com/yuin/goldmark/actions?query=workflow:test](https://github.com/yuin/goldmark/actions/workflows/test.yaml/badge.svg?branch=master&event=push)](https://github.com/yuin/goldmark/actions?query=workflow:test)
 [![https://coveralls.io/github/yuin/goldmark](https://coveralls.io/repos/github/yuin/goldmark/badge.svg?branch=master)](https://coveralls.io/github/yuin/goldmark)
 [![https://goreportcard.com/report/github.com/yuin/goldmark](https://goreportcard.com/badge/github.com/yuin/goldmark)](https://goreportcard.com/report/github.com/yuin/goldmark)
 
 > A Markdown parser written in Go. Easy to extend, standards-compliant, well-structured.
 
-goldmark is compliant with CommonMark 0.30.
+goldmark is compliant with CommonMark 0.31.2.
+
+- [goldmark playground](https://yuin.github.io/goldmark/playground/) : Try goldmark online. This playground is built with WASM(5-10MB).
 
 Motivation
 ----------------------
@@ -260,7 +262,7 @@ You can override autolinking patterns via options.
 
 | Functional option | Type | Description |
 | ----------------- | ---- | ----------- |
-| `extension.WithLinkifyAllowedProtocols` | `[][]byte` | List of allowed protocols such as `[][]byte{ []byte("http:") }` |
+| `extension.WithLinkifyAllowedProtocols` | `[][]byte \| []string` | List of allowed protocols such as `[]string{ "http:" }` |
 | `extension.WithLinkifyURLRegexp` | `*regexp.Regexp` | Regexp that defines URLs, including protocols |
 | `extension.WithLinkifyWWWRegexp` | `*regexp.Regexp` | Regexp that defines URL starting with `www.`. This pattern corresponds to [the extended www autolink](https://github.github.com/gfm/#extended-www-autolink) |
 | `extension.WithLinkifyEmailRegexp` | `*regexp.Regexp` | Regexp that defines email addresses` |
@@ -277,12 +279,12 @@ markdown := goldmark.New(
     ),
     goldmark.WithExtensions(
         extension.NewLinkify(
-            extension.WithLinkifyAllowedProtocols([][]byte{
-                []byte("http:"),
-                []byte("https:"),
+            extension.WithLinkifyAllowedProtocols([]string{
+                "http:",
+                "https:",
             }),
             extension.WithLinkifyURLRegexp(
-                xurls.Strict,
+                xurls.Strict(),
             ),
         ),
     ),
@@ -297,13 +299,13 @@ This extension has some options:
 
 | Functional option | Type | Description |
 | ----------------- | ---- | ----------- |
-| `extension.WithFootnoteIDPrefix` | `[]byte` |  a prefix for the id attributes.|
+| `extension.WithFootnoteIDPrefix` | `[]byte \| string` |  a prefix for the id attributes.|
 | `extension.WithFootnoteIDPrefixFunction` | `func(gast.Node) []byte` |  a function that determines the id attribute for given Node.|
-| `extension.WithFootnoteLinkTitle` | `[]byte` |  an optional title attribute for footnote links.|
-| `extension.WithFootnoteBacklinkTitle` | `[]byte` |  an optional title attribute for footnote backlinks. |
-| `extension.WithFootnoteLinkClass` | `[]byte` |  a class for footnote links. This defaults to `footnote-ref`. |
-| `extension.WithFootnoteBacklinkClass` | `[]byte` |  a class for footnote backlinks. This defaults to `footnote-backref`. |
-| `extension.WithFootnoteBacklinkHTML` | `[]byte` |  a class for footnote backlinks. This defaults to `↩︎`. |
+| `extension.WithFootnoteLinkTitle` | `[]byte \| string` |  an optional title attribute for footnote links.|
+| `extension.WithFootnoteBacklinkTitle` | `[]byte \| string` |  an optional title attribute for footnote backlinks. |
+| `extension.WithFootnoteLinkClass` | `[]byte \| string` |  a class for footnote links. This defaults to `footnote-ref`. |
+| `extension.WithFootnoteBacklinkClass` | `[]byte \| string` |  a class for footnote backlinks. This defaults to `footnote-backref`. |
+| `extension.WithFootnoteBacklinkHTML` | `[]byte \| string` |  a class for footnote backlinks. This defaults to `↩︎`. |
 
 Some options can have special substitutions. Occurrences of “^^” in the string will be replaced by the corresponding footnote number in the HTML output. Occurrences of “%%” will be replaced by a number for the reference (footnotes can have multiple references).
 
@@ -319,7 +321,7 @@ for _, path := range files {
     markdown := goldmark.New(
         goldmark.WithExtensions(
             NewFootnote(
-                WithFootnoteIDPrefix([]byte(path)),
+                WithFootnoteIDPrefix(path),
             ),
         ),
     )
@@ -379,9 +381,47 @@ This extension provides additional options for CJK users.
 
 | Functional option | Type | Description |
 | ----------------- | ---- | ----------- |
-| `extension.WithEastAsianLineBreaks` | `-` | Soft line breaks are rendered as a newline. Some asian users will see it as an unnecessary space. With this option, soft line breaks between east asian wide characters will be ignored. |
+| `extension.WithEastAsianLineBreaks` | `...extension.EastAsianLineBreaksStyle` | Soft line breaks are rendered as a newline. Some asian users will see it as an unnecessary space. With this option, soft line breaks between east asian wide characters will be ignored. This defaults to `EastAsianLineBreaksStyleSimple`. |
 | `extension.WithEscapedSpace` | `-` | Without spaces around an emphasis started with east asian punctuations, it is not interpreted as an emphasis(as defined in CommonMark spec). With this option, you can avoid this inconvenient behavior by putting 'not rendered' spaces around an emphasis like `太郎は\ **「こんにちわ」**\ といった`. |
 
+#### Styles of Line Breaking
+
+| Style | Description |
+| ----- | ----------- |
+| `EastAsianLineBreaksStyleSimple` | Soft line breaks are ignored if both sides of the break are east asian wide character. This behavior is the same as [`east_asian_line_breaks`](https://pandoc.org/MANUAL.html#extension-east_asian_line_breaks) in Pandoc. |
+| `EastAsianLineBreaksCSS3Draft` | This option implements CSS text level3 [Segment Break Transformation Rules](https://drafts.csswg.org/css-text-3/#line-break-transform) with [some enhancements](https://github.com/w3c/csswg-drafts/issues/5086). |
+
+#### Example of `EastAsianLineBreaksStyleSimple`
+
+Input Markdown:
+
+```md
+私はプログラマーです。
+東京の会社に勤めています。
+GoでWebアプリケーションを開発しています。
+```
+
+Output:
+
+```html
+<p>私はプログラマーです。東京の会社に勤めています。\nGoでWebアプリケーションを開発しています。</p>
+```
+
+#### Example of `EastAsianLineBreaksCSS3Draft`
+
+Input Markdown:
+
+```md
+私はプログラマーです。
+東京の会社に勤めています。
+GoでWebアプリケーションを開発しています。
+```
+
+Output:
+
+```html
+<p>私はプログラマーです。東京の会社に勤めています。GoでWebアプリケーションを開発しています。</p>
+```
 
 Security
 --------------------
@@ -429,6 +469,7 @@ As you can see, goldmark's performance is on par with cmark's.
 
 Extensions
 --------------------
+### List of extensions
 
 - [goldmark-meta](https://github.com/yuin/goldmark-meta): A YAML metadata
   extension for the goldmark Markdown parser.
@@ -452,6 +493,14 @@ Extensions
 - [goldmark-d2](https://github.com/FurqanSoftware/goldmark-d2): Adds support for [D2](https://d2lang.com/) diagrams.
 - [goldmark-katex](https://github.com/FurqanSoftware/goldmark-katex): Adds support for [KaTeX](https://katex.org/) math and equations.
 - [goldmark-img64](https://github.com/tenkoh/goldmark-img64): Adds support for embedding images into the document as DataURL (base64 encoded).
+- [goldmark-enclave](https://github.com/quailyquaily/goldmark-enclave): Adds support for embedding youtube/bilibili video, X's [oembed X](https://publish.x.com/), [tradingview chart](https://www.tradingview.com/widget/)'s chart, [quaily widget](https://quaily.com), [spotify embeds](https://developer.spotify.com/documentation/embeds), [dify embed](https://dify.ai/) and html audio into the document.
+- [goldmark-wiki-table](https://github.com/movsb/goldmark-wiki-table): Adds support for embedding Wiki Tables.
+- [goldmark-tgmd](https://github.com/Mad-Pixels/goldmark-tgmd): A Telegram markdown renderer that can be passed to `goldmark.WithRenderer()`.
+
+### Loading extensions at runtime
+[goldmark-dynamic](https://github.com/yuin/goldmark-dynamic) allows you to write a goldmark extension in Lua and load it at runtime without re-compilation.
+
+Please refer to  [goldmark-dynamic](https://github.com/yuin/goldmark-dynamic) for details.
 
 
 goldmark internal(for extension developers)

_benchmark/go/go.mod

@@ -6,7 +6,7 @@ require (
 	github.com/88250/lute v1.7.5
 	github.com/gomarkdown/markdown v0.0.0-20230322041520-c84983bdbf2a
 	github.com/russross/blackfriday/v2 v2.1.0
-	github.com/yuin/goldmark v1.2.1
+	github.com/yuin/goldmark v0.0.0
 	gitlab.com/golang-commonmark/markdown v0.0.0-20211110145824-bf3e522c626a
 )
 
@@ -22,3 +22,4 @@ require (
 )
 
 replace gopkg.in/russross/blackfriday.v2 v2.0.1 => github.com/russross/blackfriday/v2 v2.0.1
+replace github.com/yuin/goldmark v0.0.0 => ../../

_test/extra.txt

@@ -385,7 +385,8 @@ a* b c d *e*
 //- - - - - - - - -//
 		x
 //- - - - - - - - -//
-<pre><code>	x</code></pre>
+<pre><code>	x
+</code></pre>
 //= = = = = = = = = = = = = = = = = = = = = = = =//
 
 26: NUL bytes must be replaced with U+FFFD
@@ -771,3 +772,74 @@ a <!-- b -->
 <p>&lt;img src=./.assets/logo.svg</p>
 <p>/&gt;</p>
 //= = = = = = = = = = = = = = = = = = = = = = = =//
+
+61: Image alt with a new line
+//- - - - - - - - -//
+![alt
+text](logo.png)
+//- - - - - - - - -//
+<p><img src="logo.png" alt="alt
+text" /></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+62: Image alt with an escaped character 
+//- - - - - - - - -//
+![\`alt](https://example.com/img.png)
+//- - - - - - - - -//
+<p><img src="https://example.com/img.png" alt="`alt" /></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+63: Emphasis in link label
+//- - - - - - - - -//
+[*[a]*](b)
+//- - - - - - - - -//
+<p><a href="b"><em>[a]</em></a></p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+64: Nested list under an empty list item
+//- - - - - - - - -//
+- 
+  - foo
+//- - - - - - - - -//
+<ul>
+<li>
+<ul>
+<li>foo</li>
+</ul>
+</li>
+</ul>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+65: Nested fenced code block with tab
+//- - - - - - - - -//
+>  ```
+>	0
+>  ```
+//- - - - - - - - -//
+<blockquote>
+<pre><code> 0
+</code></pre>
+</blockquote>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+66: EOF should be rendered as a newline with an unclosed block(w/ TAB)
+//- - - - - - - - -//
+>  ```
+>	0
+//- - - - - - - - -//
+<blockquote>
+<pre><code> 0
+</code></pre>
+</blockquote>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+67: EOF should be rendered as a newline with an unclosed block
+//- - - - - - - - -//
+>  ```
+>   0
+//- - - - - - - - -//
+<blockquote>
+<pre><code> 0
+</code></pre>
+</blockquote>
+//= = = = = = = = = = = = = = = = = = = = = = = =//

ast/ast.go

@@ -39,7 +39,7 @@ func NewNodeKind(name string) NodeKind {
 	return kindMax
 }
 
-// An Attribute is an attribute of the Node
+// An Attribute is an attribute of the Node.
 type Attribute struct {
 	Name  []byte
 	Value interface{}
@@ -123,6 +123,12 @@ type Node interface {
 	Dump(source []byte, level int)
 
 	// Text returns text values of this node.
+	// This method is valid only for some inline nodes.
+	// If this node is a block node, Text returns a text value as reasonable as possible.
+	// Notice that there are no 'correct' text values for the block nodes.
+	// Result for the block nodes may be different from your expectation.
+	//
+	// Deprecated: Use other properties of the node to get the text value(i.e. Pragraph.Lines, Text.Value).
 	Text(source []byte) []byte
 
 	// HasBlankPreviousLines returns true if the row before this node is blank,
@@ -248,7 +254,7 @@ func (n *BaseNode) RemoveChildren(self Node) {
 	n.childCount = 0
 }
 
-// SortChildren implements Node.SortChildren
+// SortChildren implements Node.SortChildren.
 func (n *BaseNode) SortChildren(comparator func(n1, n2 Node) int) {
 	var sorted Node
 	current := n.firstChild
@@ -358,7 +364,7 @@ func (n *BaseNode) InsertBefore(self, v1, insertee Node) {
 	}
 }
 
-// OwnerDocument implements Node.OwnerDocument
+// OwnerDocument implements Node.OwnerDocument.
 func (n *BaseNode) OwnerDocument() *Document {
 	d := n.Parent()
 	for {
@@ -375,10 +381,17 @@ func (n *BaseNode) OwnerDocument() *Document {
 }
 
 // Text implements Node.Text .
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. Pragraph.Lines, Text.Value).
 func (n *BaseNode) Text(source []byte) []byte {
 	var buf bytes.Buffer
 	for c := n.firstChild; c != nil; c = c.NextSibling() {
 		buf.Write(c.Text(source))
+		if sb, ok := c.(interface {
+			SoftLineBreak() bool
+		}); ok && sb.SoftLineBreak() {
+			buf.WriteByte('\n')
+		}
 	}
 	return buf.Bytes()
 }
@@ -399,7 +412,7 @@ func (n *BaseNode) SetAttribute(name []byte, value interface{}) {
 	n.attributes = append(n.attributes, Attribute{name, value})
 }
 
-// SetAttributeString implements Node.SetAttributeString
+// SetAttributeString implements Node.SetAttributeString.
 func (n *BaseNode) SetAttributeString(name string, value interface{}) {
 	n.SetAttribute(util.StringToReadOnlyBytes(name), value)
 }
@@ -422,12 +435,12 @@ func (n *BaseNode) AttributeString(s string) (interface{}, bool) {
 	return n.Attribute(util.StringToReadOnlyBytes(s))
 }
 
-// Attributes implements Node.Attributes
+// Attributes implements Node.Attributes.
 func (n *BaseNode) Attributes() []Attribute {
 	return n.attributes
 }
 
-// RemoveAttributes implements Node.RemoveAttributes
+// RemoveAttributes implements Node.RemoveAttributes.
 func (n *BaseNode) RemoveAttributes() {
 	n.attributes = nil
 }

ast/ast_test.go

@@ -5,21 +5,6 @@ import (
 	"testing"
 )
 
-func TestRemoveChildren(t *testing.T) {
-	root := NewDocument()
-
-	node1 := NewDocument()
-
-	node2 := NewDocument()
-
-	root.AppendChild(root, node1)
-	root.AppendChild(root, node2)
-
-	root.RemoveChildren(root)
-
-	t.Logf("%+v", node2.PreviousSibling())
-}
-
 func TestWalk(t *testing.T) {
 	tests := []struct {
 		name   string

ast/block.go

@@ -14,12 +14,12 @@ type BaseBlock struct {
 	lines              *textm.Segments
 }
 
-// Type implements Node.Type
+// Type implements Node.Type.
 func (b *BaseBlock) Type() NodeType {
 	return TypeBlock
 }
 
-// IsRaw implements Node.IsRaw
+// IsRaw implements Node.IsRaw.
 func (b *BaseBlock) IsRaw() bool {
 	return false
 }
@@ -34,7 +34,7 @@ func (b *BaseBlock) SetBlankPreviousLines(v bool) {
 	b.blankPreviousLines = v
 }
 
-// Lines implements Node.Lines
+// Lines implements Node.Lines.
 func (b *BaseBlock) Lines() *textm.Segments {
 	if b.lines == nil {
 		b.lines = textm.NewSegments()
@@ -42,7 +42,7 @@ func (b *BaseBlock) Lines() *textm.Segments {
 	return b.lines
 }
 
-// SetLines implements Node.SetLines
+// SetLines implements Node.SetLines.
 func (b *BaseBlock) SetLines(v *textm.Segments) {
 	b.lines = v
 }
@@ -72,7 +72,7 @@ func (n *Document) Kind() NodeKind {
 	return KindDocument
 }
 
-// OwnerDocument implements Node.OwnerDocument
+// OwnerDocument implements Node.OwnerDocument.
 func (n *Document) OwnerDocument() *Document {
 	return n
 }
@@ -130,6 +130,13 @@ func (n *TextBlock) Kind() NodeKind {
 	return KindTextBlock
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. TextBlock.Lines).
+func (n *TextBlock) Text(source []byte) []byte {
+	return n.Lines().Value(source)
+}
+
 // NewTextBlock returns a new TextBlock node.
 func NewTextBlock() *TextBlock {
 	return &TextBlock{
@@ -155,6 +162,13 @@ func (n *Paragraph) Kind() NodeKind {
 	return KindParagraph
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. Paragraph.Lines).
+func (n *Paragraph) Text(source []byte) []byte {
+	return n.Lines().Value(source)
+}
+
 // NewParagraph returns a new Paragraph node.
 func NewParagraph() *Paragraph {
 	return &Paragraph{
@@ -249,6 +263,13 @@ func (n *CodeBlock) Kind() NodeKind {
 	return KindCodeBlock
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. CodeBlock.Lines).
+func (n *CodeBlock) Text(source []byte) []byte {
+	return n.Lines().Value(source)
+}
+
 // NewCodeBlock returns a new CodeBlock node.
 func NewCodeBlock() *CodeBlock {
 	return &CodeBlock{
@@ -304,6 +325,13 @@ func (n *FencedCodeBlock) Kind() NodeKind {
 	return KindFencedCodeBlock
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. FencedCodeBlock.Lines).
+func (n *FencedCodeBlock) Text(source []byte) []byte {
+	return n.Lines().Value(source)
+}
+
 // NewFencedCodeBlock return a new FencedCodeBlock node.
 func NewFencedCodeBlock(info *Text) *FencedCodeBlock {
 	return &FencedCodeBlock{
@@ -431,19 +459,19 @@ func NewListItem(offset int) *ListItem {
 type HTMLBlockType int
 
 const (
-	// HTMLBlockType1 represents type 1 html blocks
+	// HTMLBlockType1 represents type 1 html blocks.
 	HTMLBlockType1 HTMLBlockType = iota + 1
-	// HTMLBlockType2 represents type 2 html blocks
+	// HTMLBlockType2 represents type 2 html blocks.
 	HTMLBlockType2
-	// HTMLBlockType3 represents type 3 html blocks
+	// HTMLBlockType3 represents type 3 html blocks.
 	HTMLBlockType3
-	// HTMLBlockType4 represents type 4 html blocks
+	// HTMLBlockType4 represents type 4 html blocks.
 	HTMLBlockType4
-	// HTMLBlockType5 represents type 5 html blocks
+	// HTMLBlockType5 represents type 5 html blocks.
 	HTMLBlockType5
-	// HTMLBlockType6 represents type 6 html blocks
+	// HTMLBlockType6 represents type 6 html blocks.
 	HTMLBlockType6
-	// HTMLBlockType7 represents type 7 html blocks
+	// HTMLBlockType7 represents type 7 html blocks.
 	HTMLBlockType7
 )
 
@@ -498,6 +526,17 @@ func (n *HTMLBlock) Kind() NodeKind {
 	return KindHTMLBlock
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. HTMLBlock.Lines).
+func (n *HTMLBlock) Text(source []byte) []byte {
+	ret := n.Lines().Value(source)
+	if n.HasClosure() {
+		ret = append(ret, n.ClosureLine.Value(source)...)
+	}
+	return ret
+}
+
 // NewHTMLBlock returns a new HTMLBlock node.
 func NewHTMLBlock(typ HTMLBlockType) *HTMLBlock {
 	return &HTMLBlock{

ast/inline.go

@@ -13,12 +13,12 @@ type BaseInline struct {
 	BaseNode
 }
 
-// Type implements Node.Type
+// Type implements Node.Type.
 func (b *BaseInline) Type() NodeType {
 	return TypeInline
 }
 
-// IsRaw implements Node.IsRaw
+// IsRaw implements Node.IsRaw.
 func (b *BaseInline) IsRaw() bool {
 	return false
 }
@@ -33,12 +33,12 @@ func (b *BaseInline) SetBlankPreviousLines(v bool) {
 	panic("can not call with inline nodes.")
 }
 
-// Lines implements Node.Lines
+// Lines implements Node.Lines.
 func (b *BaseInline) Lines() *textm.Segments {
 	panic("can not call with inline nodes.")
 }
 
-// SetLines implements Node.SetLines
+// SetLines implements Node.SetLines.
 func (b *BaseInline) SetLines(v *textm.Segments) {
 	panic("can not call with inline nodes.")
 }
@@ -132,7 +132,8 @@ func (n *Text) Merge(node Node, source []byte) bool {
 	if !ok {
 		return false
 	}
-	if n.Segment.Stop != t.Segment.Start || t.Segment.Padding != 0 || source[n.Segment.Stop-1] == '\n' || t.IsRaw() != n.IsRaw() {
+	if n.Segment.Stop != t.Segment.Start || t.Segment.Padding != 0 ||
+		source[n.Segment.Stop-1] == '\n' || t.IsRaw() != n.IsRaw() {
 		return false
 	}
 	n.Segment.Stop = t.Segment.Stop
@@ -142,17 +143,25 @@ func (n *Text) Merge(node Node, source []byte) bool {
 }
 
 // Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. Text.Value).
 func (n *Text) Text(source []byte) []byte {
 	return n.Segment.Value(source)
 }
 
+// Value returns a value of this node.
+// SoftLineBreaks are not included in the returned value.
+func (n *Text) Value(source []byte) []byte {
+	return n.Segment.Value(source)
+}
+
 // Dump implements Node.Dump.
 func (n *Text) Dump(source []byte, level int) {
 	fs := textFlagsString(n.flags)
 	if len(fs) != 0 {
 		fs = "(" + fs + ")"
 	}
-	fmt.Printf("%sText%s: \"%s\"\n", strings.Repeat("    ", level), fs, strings.TrimRight(string(n.Text(source)), "\n"))
+	fmt.Printf("%sText%s: \"%s\"\n", strings.Repeat("    ", level), fs, strings.TrimRight(string(n.Value(source)), "\n"))
 }
 
 // KindText is a NodeKind of the Text node.
@@ -214,7 +223,7 @@ func MergeOrReplaceTextSegment(parent Node, n Node, s textm.Segment) {
 	}
 }
 
-// A String struct is a textual content that has a concrete value
+// A String struct is a textual content that has a concrete value.
 type String struct {
 	BaseInline
 
@@ -257,6 +266,8 @@ func (n *String) SetCode(v bool) {
 }
 
 // Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. String.Value).
 func (n *String) Text(source []byte) []byte {
 	return n.Value
 }
@@ -305,7 +316,7 @@ func (n *CodeSpan) IsBlank(source []byte) bool {
 	return true
 }
 
-// Dump implements Node.Dump
+// Dump implements Node.Dump.
 func (n *CodeSpan) Dump(source []byte, level int) {
 	DumpHelper(n, source, level, nil, nil)
 }
@@ -467,7 +478,7 @@ type AutoLink struct {
 // Inline implements Inline.Inline.
 func (n *AutoLink) Inline() {}
 
-// Dump implements Node.Dump
+// Dump implements Node.Dump.
 func (n *AutoLink) Dump(source []byte, level int) {
 	segment := n.value.Segment
 	m := map[string]string{
@@ -491,15 +502,22 @@ func (n *AutoLink) URL(source []byte) []byte {
 		ret := make([]byte, 0, len(n.Protocol)+s.Len()+3)
 		ret = append(ret, n.Protocol...)
 		ret = append(ret, ':', '/', '/')
-		ret = append(ret, n.value.Text(source)...)
+		ret = append(ret, n.value.Value(source)...)
 		return ret
 	}
-	return n.value.Text(source)
+	return n.value.Value(source)
 }
 
 // Label returns a label of this node.
 func (n *AutoLink) Label(source []byte) []byte {
-	return n.value.Text(source)
+	return n.value.Value(source)
+}
+
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. AutoLink.Label).
+func (n *AutoLink) Text(source []byte) []byte {
+	return n.value.Value(source)
 }
 
 // NewAutoLink returns a new AutoLink node.
@@ -540,6 +558,13 @@ func (n *RawHTML) Kind() NodeKind {
 	return KindRawHTML
 }
 
+// Text implements Node.Text.
+//
+// Deprecated: Use other properties of the node to get the text value(i.e. RawHTML.Segments).
+func (n *RawHTML) Text(source []byte) []byte {
+	return n.Segments.Value(source)
+}
+
 // NewRawHTML returns a new RawHTML node.
 func NewRawHTML() *RawHTML {
 	return &RawHTML{

ast_test.go

@@ -0,0 +1,204 @@
+package goldmark_test
+
+import (
+	"bytes"
+	"testing"
+
+	. "github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/testutil"
+	"github.com/yuin/goldmark/text"
+)
+
+func TestASTBlockNodeText(t *testing.T) {
+	var cases = []struct {
+		Name   string
+		Source string
+		T1     string
+		T2     string
+		C      bool
+	}{
+		{
+			Name: "AtxHeading",
+			Source: `# l1
+
+a
+
+# l2`,
+			T1: `l1`,
+			T2: `l2`,
+		},
+		{
+			Name: "SetextHeading",
+			Source: `l1
+l2
+===============
+
+a
+
+l3
+l4
+==============`,
+			T1: `l1
+l2`,
+			T2: `l3
+l4`,
+		},
+		{
+			Name: "CodeBlock",
+			Source: `    l1
+    l2
+
+a
+
+    l3
+	l4`,
+			T1: `l1
+l2
+`,
+			T2: `l3
+l4
+`,
+		},
+		{
+			Name: "FencedCodeBlock",
+			Source: "```" + `
+l1
+l2
+` + "```" + `
+
+a
+
+` + "```" + `
+l3
+l4`,
+			T1: `l1
+l2
+`,
+			T2: `l3
+l4
+`,
+		},
+		{
+			Name: "Blockquote",
+			Source: `> l1
+> l2
+
+a
+
+> l3
+> l4`,
+			T1: `l1
+l2`,
+			T2: `l3
+l4`,
+		},
+		{
+			Name: "List",
+			Source: `- l1
+  l2
+
+a
+
+- l3
+  l4`,
+			T1: `l1
+l2`,
+			T2: `l3
+l4`,
+			C: true,
+		},
+		{
+			Name: "HTMLBlock",
+			Source: `<div>
+l1
+l2
+</div>
+
+a
+
+<div>
+l3
+l4`,
+			T1: `<div>
+l1
+l2
+</div>
+`,
+			T2: `<div>
+l3
+l4`,
+		},
+	}
+
+	for _, cs := range cases {
+		t.Run(cs.Name, func(t *testing.T) {
+			s := []byte(cs.Source)
+			md := New()
+			n := md.Parser().Parse(text.NewReader(s))
+			c1 := n.FirstChild()
+			c2 := c1.NextSibling().NextSibling()
+			if cs.C {
+				c1 = c1.FirstChild()
+				c2 = c2.FirstChild()
+			}
+			if !bytes.Equal(c1.Text(s), []byte(cs.T1)) { // nolint: staticcheck
+
+				t.Errorf("%s unmatch: %s", cs.Name, testutil.DiffPretty(c1.Text(s), []byte(cs.T1))) // nolint: staticcheck
+
+			}
+			if !bytes.Equal(c2.Text(s), []byte(cs.T2)) { // nolint: staticcheck
+
+				t.Errorf("%s(EOF) unmatch: %s", cs.Name, testutil.DiffPretty(c2.Text(s), []byte(cs.T2))) // nolint: staticcheck
+
+			}
+		})
+	}
+
+}
+
+func TestASTInlineNodeText(t *testing.T) {
+	var cases = []struct {
+		Name   string
+		Source string
+		T1     string
+	}{
+		{
+			Name:   "CodeSpan",
+			Source: "`c1`",
+			T1:     `c1`,
+		},
+		{
+			Name:   "Emphasis",
+			Source: `*c1 **c2***`,
+			T1:     `c1 c2`,
+		},
+		{
+			Name:   "Link",
+			Source: `[label](url)`,
+			T1:     `label`,
+		},
+		{
+			Name:   "AutoLink",
+			Source: `<http://url>`,
+			T1:     `http://url`,
+		},
+		{
+			Name:   "RawHTML",
+			Source: `<span>c1</span>`,
+			T1:     `<span>`,
+		},
+	}
+
+	for _, cs := range cases {
+		t.Run(cs.Name, func(t *testing.T) {
+			s := []byte(cs.Source)
+			md := New()
+			n := md.Parser().Parse(text.NewReader(s))
+			c1 := n.FirstChild().FirstChild()
+			if !bytes.Equal(c1.Text(s), []byte(cs.T1)) { // nolint: staticcheck
+				t.Errorf("%s unmatch:\n%s", cs.Name, testutil.DiffPretty(c1.Text(s), []byte(cs.T1))) // nolint: staticcheck
+			}
+		})
+	}
+
+}

commonmark_test.go

@@ -2,7 +2,7 @@ package goldmark_test
 
 import (
 	"encoding/json"
-	"io/ioutil"
+	"os"
 	"testing"
 
 	. "github.com/yuin/goldmark"
@@ -20,7 +20,7 @@ type commonmarkSpecTestCase struct {
 }
 
 func TestSpec(t *testing.T) {
-	bs, err := ioutil.ReadFile("_test/spec.json")
+	bs, err := os.ReadFile("_test/spec.json")
 	if err != nil {
 		panic(err)
 	}

extension/_test/definition_list.txt

@@ -150,7 +150,8 @@ on two lines.</p>
 //- - - - - - - - -//
 <dl>
 <dt>0</dt>
-<dd><pre><code>	0</code></pre>
+<dd><pre><code>	0
+</code></pre>
 </dd>
 </dl>
 //= = = = = = = = = = = = = = = = = = = = = = = =//

extension/_test/strikethrough.txt

@@ -5,8 +5,6 @@
 <p><del>Hi</del> Hello, world!</p>
 //= = = = = = = = = = = = = = = = = = = = = = = =//
 
-
-
 2
 //- - - - - - - - -//
 This ~~has a
@@ -16,3 +14,26 @@ new paragraph~~.
 <p>This ~~has a</p>
 <p>new paragraph~~.</p>
 //= = = = = = = = = = = = = = = = = = = = = = = =//
+
+3
+//- - - - - - - - -//
+~Hi~ Hello, world!
+//- - - - - - - - -//
+<p><del>Hi</del> Hello, world!</p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+4: Three or more tildes do not create a strikethrough
+//- - - - - - - - -//
+This will ~~~not~~~ strike.
+//- - - - - - - - -//
+<p>This will ~~~not~~~ strike.</p>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+5: Leading three or more tildes do not create a strikethrough, create a code block
+//- - - - - - - - -//
+~~~Hi~~~ Hello, world!
+//- - - - - - - - -//
+<pre><code class="language-Hi~~~"></code></pre>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+

extension/_test/tasklist.txt

@@ -28,3 +28,24 @@
 <li><input disabled="" type="checkbox"> bim</li>
 </ul>
 //= = = = = = = = = = = = = = = = = = = = = = = =//
+
+
+
+3
+//- - - - - - - - -//
+- test[x]=[x]
+//- - - - - - - - -//
+<ul>
+<li>test[x]=[x]</li>
+</ul>
+//= = = = = = = = = = = = = = = = = = = = = = = =//
+
+
+4
+//- - - - - - - - -//
++ [x] [x]
+//- - - - - - - - -//
+<ul>
+<li><input checked="" disabled="" type="checkbox"> [x]</li>
+</ul>
+//= = = = = = = = = = = = = = = = = = = = = = = =//

extension/ast/footnote.go

@@ -88,7 +88,7 @@ type Footnote struct {
 func (n *Footnote) Dump(source []byte, level int) {
 	m := map[string]string{}
 	m["Index"] = fmt.Sprintf("%v", n.Index)
-	m["Ref"] = fmt.Sprintf("%s", n.Ref)
+	m["Ref"] = string(n.Ref)
 	gast.DumpHelper(n, source, level, m, nil)
 }
 

extension/ast/table.go

@@ -2,8 +2,9 @@ package ast
 
 import (
 	"fmt"
-	gast "github.com/yuin/goldmark/ast"
 	"strings"
+
+	gast "github.com/yuin/goldmark/ast"
 )
 
 // Alignment is a text alignment of table cells.
@@ -45,7 +46,7 @@ type Table struct {
 	Alignments []Alignment
 }
 
-// Dump implements Node.Dump
+// Dump implements Node.Dump.
 func (n *Table) Dump(source []byte, level int) {
 	gast.DumpHelper(n, source, level, nil, func(level int) {
 		indent := strings.Repeat("    ", level)

extension/ast_test.go

@@ -0,0 +1,123 @@
+package extension
+
+import (
+	"bytes"
+	"testing"
+
+	"github.com/yuin/goldmark"
+	"github.com/yuin/goldmark/renderer/html"
+	"github.com/yuin/goldmark/testutil"
+	"github.com/yuin/goldmark/text"
+)
+
+func TestASTBlockNodeText(t *testing.T) {
+	var cases = []struct {
+		Name   string
+		Source string
+		T1     string
+		T2     string
+		C      bool
+	}{
+		{
+			Name: "DefinitionList",
+			Source: `c1
+:   c2
+    c3
+
+a
+
+c4
+:   c5
+    c6`,
+			T1: `c1c2
+c3`,
+			T2: `c4c5
+c6`,
+		},
+		{
+			Name: "Table",
+			Source: `| h1 | h2 |
+| -- | -- |
+| c1 | c2 |
+
+a
+
+
+| h3 | h4 |
+| -- | -- |
+| c3 | c4 |`,
+
+			T1: `h1h2c1c2`,
+			T2: `h3h4c3c4`,
+		},
+	}
+
+	for _, cs := range cases {
+		t.Run(cs.Name, func(t *testing.T) {
+			s := []byte(cs.Source)
+			md := goldmark.New(
+				goldmark.WithRendererOptions(
+					html.WithUnsafe(),
+				),
+				goldmark.WithExtensions(
+					DefinitionList,
+					Table,
+				),
+			)
+			n := md.Parser().Parse(text.NewReader(s))
+			c1 := n.FirstChild()
+			c2 := c1.NextSibling().NextSibling()
+			if cs.C {
+				c1 = c1.FirstChild()
+				c2 = c2.FirstChild()
+			}
+			if !bytes.Equal(c1.Text(s), []byte(cs.T1)) { // nolint: staticcheck
+
+				t.Errorf("%s unmatch:\n%s", cs.Name, testutil.DiffPretty(c1.Text(s), []byte(cs.T1))) // nolint: staticcheck
+
+			}
+			if !bytes.Equal(c2.Text(s), []byte(cs.T2)) { // nolint: staticcheck
+
+				t.Errorf("%s(EOF) unmatch: %s", cs.Name, testutil.DiffPretty(c2.Text(s), []byte(cs.T2))) // nolint: staticcheck
+
+			}
+		})
+	}
+
+}
+
+func TestASTInlineNodeText(t *testing.T) {
+	var cases = []struct {
+		Name   string
+		Source string
+		T1     string
+	}{
+		{
+			Name:   "Strikethrough",
+			Source: `~c1 *c2*~`,
+			T1:     `c1 c2`,
+		},
+	}
+
+	for _, cs := range cases {
+		t.Run(cs.Name, func(t *testing.T) {
+			s := []byte(cs.Source)
+			md := goldmark.New(
+				goldmark.WithRendererOptions(
+					html.WithUnsafe(),
+				),
+				goldmark.WithExtensions(
+					Strikethrough,
+				),
+			)
+			n := md.Parser().Parse(text.NewReader(s))
+			c1 := n.FirstChild().FirstChild()
+			if !bytes.Equal(c1.Text(s), []byte(cs.T1)) { // nolint: staticcheck
+
+				t.Errorf("%s unmatch:\n%s", cs.Name, testutil.DiffPretty(c1.Text(s), []byte(cs.T1))) // nolint: staticcheck
+
+			}
+		})
+	}
+
+}

extension/cjk.go

@@ -9,11 +9,30 @@ import (
 // A CJKOption sets options for CJK support mostly for HTML based renderers.
 type CJKOption func(*cjk)
 
+// A EastAsianLineBreaks is a style of east asian line breaks.
+type EastAsianLineBreaks int
+
+const (
+	//EastAsianLineBreaksNone renders line breaks as it is.
+	EastAsianLineBreaksNone EastAsianLineBreaks = iota
+	// EastAsianLineBreaksSimple is a style where soft line breaks are ignored
+	// if both sides of the break are east asian wide characters.
+	EastAsianLineBreaksSimple
+	// EastAsianLineBreaksCSS3Draft is a style where soft line breaks are ignored
+	// even if only one side of the break is an east asian wide character.
+	EastAsianLineBreaksCSS3Draft
+)
+
 // WithEastAsianLineBreaks is a functional option that indicates whether softline breaks
 // between east asian wide characters should be ignored.
-func WithEastAsianLineBreaks() CJKOption {
+// style defauts to [EastAsianLineBreaksSimple] .
+func WithEastAsianLineBreaks(style ...EastAsianLineBreaks) CJKOption {
 	return func(c *cjk) {
-		c.EastAsianLineBreaks = true
+		if len(style) == 0 {
+			c.EastAsianLineBreaks = EastAsianLineBreaksSimple
+			return
+		}
+		c.EastAsianLineBreaks = style[0]
 	}
 }
 
@@ -25,15 +44,18 @@ func WithEscapedSpace() CJKOption {
 }
 
 type cjk struct {
-	EastAsianLineBreaks bool
+	EastAsianLineBreaks EastAsianLineBreaks
 	EscapedSpace        bool
 }
 
+// CJK is a goldmark extension that provides functionalities for CJK languages.
 var CJK = NewCJK(WithEastAsianLineBreaks(), WithEscapedSpace())
 
 // NewCJK returns a new extension with given options.
 func NewCJK(opts ...CJKOption) goldmark.Extender {
-	e := &cjk{}
+	e := &cjk{
+		EastAsianLineBreaks: EastAsianLineBreaksNone,
+	}
 	for _, opt := range opts {
 		opt(e)
 	}
@@ -41,9 +63,8 @@ func NewCJK(opts ...CJKOption) goldmark.Extender {
 }
 
 func (e *cjk) Extend(m goldmark.Markdown) {
-	if e.EastAsianLineBreaks {
-		m.Renderer().AddOptions(html.WithEastAsianLineBreaks())
-	}
+	m.Renderer().AddOptions(html.WithEastAsianLineBreaks(
+		html.EastAsianLineBreaks(e.EastAsianLineBreaks)))
 	if e.EscapedSpace {
 		m.Renderer().AddOptions(html.WithWriter(html.NewWriter(html.WithEscapedSpace())))
 		m.Parser().AddOptions(parser.WithEscapedSpace())

extension/cjk_test.go

@@ -177,6 +177,7 @@ func TestEastAsianLineBreaks(t *testing.T) {
 		t,
 	)
 
+	// Tests with EastAsianLineBreaksStyleSimple
 	markdown = goldmark.New(goldmark.WithRendererOptions(
 		html.WithXHTML(),
 		html.WithUnsafe(),
@@ -197,4 +198,72 @@ func TestEastAsianLineBreaks(t *testing.T) {
 		},
 		t,
 	)
+	no = 8
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          no,
+			Description: "Soft line breaks between east asian wide characters or punctuations are ignored",
+			Markdown:    "太郎は\\ **「こんにちわ」**\\ と、\r\n言った\r\nんです",
+			Expected:    "<p>太郎は\\ <strong>「こんにちわ」</strong>\\ と、言ったんです</p>",
+		},
+		t,
+	)
+	no = 9
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          no,
+			Description: "Soft line breaks between an east asian wide character and a western character are ignored",
+			Markdown:    "私はプログラマーです。\n東京の会社に勤めています。\nGoでWebアプリケーションを開発しています。",
+			Expected:    "<p>私はプログラマーです。東京の会社に勤めています。\nGoでWebアプリケーションを開発しています。</p>",
+		},
+		t,
+	)
+
+	// Tests with EastAsianLineBreaksCSS3Draft
+	markdown = goldmark.New(goldmark.WithRendererOptions(
+		html.WithXHTML(),
+		html.WithUnsafe(),
+	),
+		goldmark.WithExtensions(
+			NewCJK(WithEastAsianLineBreaks(EastAsianLineBreaksCSS3Draft)),
+		),
+	)
+	no = 10
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          no,
+			Description: "Soft line breaks between a western character and an east asian wide character are ignored",
+			Markdown:    "太郎は\\ **「こんにちわ」**\\ と言ったa\nんです",
+			Expected:    "<p>太郎は\\ <strong>「こんにちわ」</strong>\\ と言ったaんです</p>",
+		},
+		t,
+	)
+
+	no = 11
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          no,
+			Description: "Soft line breaks between an east asian wide character and a western character are ignored",
+			Markdown:    "太郎は\\ **「こんにちわ」**\\ と言った\nbんです",
+			Expected:    "<p>太郎は\\ <strong>「こんにちわ」</strong>\\ と言ったbんです</p>",
+		},
+		t,
+	)
+
+	no = 12
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          no,
+			Description: "Soft line breaks between an east asian wide character and a western character are ignored",
+			Markdown:    "私はプログラマーです。\n東京の会社に勤めています。\nGoでWebアプリケーションを開発しています。",
+			Expected:    "<p>私はプログラマーです。東京の会社に勤めています。GoでWebアプリケーションを開発しています。</p>",
+		},
+		t,
+	)
+
 }

extension/definition_list.go

@@ -113,7 +113,8 @@ func (b *definitionDescriptionParser) Trigger() []byte {
 	return []byte{':'}
 }
 
-func (b *definitionDescriptionParser) Open(parent gast.Node, reader text.Reader, pc parser.Context) (gast.Node, parser.State) {
+func (b *definitionDescriptionParser) Open(
+	parent gast.Node, reader text.Reader, pc parser.Context) (gast.Node, parser.State) {
 	line, _ := reader.PeekLine()
 	pos := pc.BlockOffset()
 	indent := pc.BlockIndent()
@@ -199,7 +200,8 @@ func (r *DefinitionListHTMLRenderer) RegisterFuncs(reg renderer.NodeRendererFunc
 // DefinitionListAttributeFilter defines attribute names which dl elements can have.
 var DefinitionListAttributeFilter = html.GlobalAttributeFilter
 
-func (r *DefinitionListHTMLRenderer) renderDefinitionList(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *DefinitionListHTMLRenderer) renderDefinitionList(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		if n.Attributes() != nil {
 			_, _ = w.WriteString("<dl")
@@ -217,7 +219,8 @@ func (r *DefinitionListHTMLRenderer) renderDefinitionList(w util.BufWriter, sour
 // DefinitionTermAttributeFilter defines attribute names which dd elements can have.
 var DefinitionTermAttributeFilter = html.GlobalAttributeFilter
 
-func (r *DefinitionListHTMLRenderer) renderDefinitionTerm(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *DefinitionListHTMLRenderer) renderDefinitionTerm(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		if n.Attributes() != nil {
 			_, _ = w.WriteString("<dt")
@@ -235,7 +238,8 @@ func (r *DefinitionListHTMLRenderer) renderDefinitionTerm(w util.BufWriter, sour
 // DefinitionDescriptionAttributeFilter defines attribute names which dd elements can have.
 var DefinitionDescriptionAttributeFilter = html.GlobalAttributeFilter
 
-func (r *DefinitionListHTMLRenderer) renderDefinitionDescription(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *DefinitionListHTMLRenderer) renderDefinitionDescription(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		n := node.(*ast.DefinitionDescription)
 		_, _ = w.WriteString("<dd")

extension/footnote.go

@@ -44,8 +44,8 @@ func (b *footnoteBlockParser) Open(parent gast.Node, reader text.Reader, pc pars
 		return nil, parser.NoChildren
 	}
 	open := pos + 1
-	closes := 0
-	closure := util.FindClosure(line[pos+1:], '[', ']', false, false)
+	var closes int
+	closure := util.FindClosure(line[pos+1:], '[', ']', false, false) //nolint:staticcheck
 	closes = pos + 1 + closure
 	next := closes + 1
 	if closure > -1 {
@@ -136,7 +136,7 @@ func (s *footnoteParser) Parse(parent gast.Node, block text.Reader, pc parser.Co
 		return nil
 	}
 	open := pos
-	closure := util.FindClosure(line[pos:], '[', ']', false, false)
+	closure := util.FindClosure(line[pos:], '[', ']', false, false) //nolint:staticcheck
 	if closure < 0 {
 		return nil
 	}
@@ -156,7 +156,7 @@ func (s *footnoteParser) Parse(parent gast.Node, block text.Reader, pc parser.Co
 		d := def.(*ast.Footnote)
 		if bytes.Equal(d.Ref, value) {
 			if d.Index < 0 {
-				list.Count += 1
+				list.Count++
 				d.Index = list.Count
 			}
 			index = d.Index
@@ -272,9 +272,9 @@ func (a *footnoteASTTransformer) Transform(node *gast.Document, reader text.Read
 // FootnoteConfig holds configuration values for the footnote extension.
 //
 // Link* and Backlink* configurations have some variables:
-// Occurrances of “^^” in the string will be replaced by the
+// Occurrences of “^^” in the string will be replaced by the
 // corresponding footnote number in the HTML output.
-// Occurrances of “%%” will be replaced by a number for the
+// Occurrences of “%%” will be replaced by a number for the
 // reference (footnotes can have multiple references).
 type FootnoteConfig struct {
 	html.Config
@@ -382,8 +382,8 @@ func (o *withFootnoteIDPrefix) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteIDPrefix is a functional option that is a prefix for the id attributes generated by footnotes.
-func WithFootnoteIDPrefix(a []byte) FootnoteOption {
-	return &withFootnoteIDPrefix{a}
+func WithFootnoteIDPrefix[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteIDPrefix{[]byte(a)}
 }
 
 const optFootnoteIDPrefixFunction renderer.OptionName = "FootnoteIDPrefixFunction"
@@ -420,8 +420,8 @@ func (o *withFootnoteLinkTitle) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteLinkTitle is a functional option that is an optional title attribute for footnote links.
-func WithFootnoteLinkTitle(a []byte) FootnoteOption {
-	return &withFootnoteLinkTitle{a}
+func WithFootnoteLinkTitle[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteLinkTitle{[]byte(a)}
 }
 
 const optFootnoteBacklinkTitle renderer.OptionName = "FootnoteBacklinkTitle"
@@ -439,8 +439,8 @@ func (o *withFootnoteBacklinkTitle) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteBacklinkTitle is a functional option that is an optional title attribute for footnote backlinks.
-func WithFootnoteBacklinkTitle(a []byte) FootnoteOption {
-	return &withFootnoteBacklinkTitle{a}
+func WithFootnoteBacklinkTitle[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteBacklinkTitle{[]byte(a)}
 }
 
 const optFootnoteLinkClass renderer.OptionName = "FootnoteLinkClass"
@@ -458,8 +458,8 @@ func (o *withFootnoteLinkClass) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteLinkClass is a functional option that is a class for footnote links.
-func WithFootnoteLinkClass(a []byte) FootnoteOption {
-	return &withFootnoteLinkClass{a}
+func WithFootnoteLinkClass[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteLinkClass{[]byte(a)}
 }
 
 const optFootnoteBacklinkClass renderer.OptionName = "FootnoteBacklinkClass"
@@ -477,8 +477,8 @@ func (o *withFootnoteBacklinkClass) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteBacklinkClass is a functional option that is a class for footnote backlinks.
-func WithFootnoteBacklinkClass(a []byte) FootnoteOption {
-	return &withFootnoteBacklinkClass{a}
+func WithFootnoteBacklinkClass[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteBacklinkClass{[]byte(a)}
 }
 
 const optFootnoteBacklinkHTML renderer.OptionName = "FootnoteBacklinkHTML"
@@ -496,8 +496,8 @@ func (o *withFootnoteBacklinkHTML) SetFootnoteOption(c *FootnoteConfig) {
 }
 
 // WithFootnoteBacklinkHTML is an HTML content for footnote backlinks.
-func WithFootnoteBacklinkHTML(a []byte) FootnoteOption {
-	return &withFootnoteBacklinkHTML{a}
+func WithFootnoteBacklinkHTML[T []byte | string](a T) FootnoteOption {
+	return &withFootnoteBacklinkHTML{[]byte(a)}
 }
 
 // FootnoteHTMLRenderer is a renderer.NodeRenderer implementation that
@@ -525,7 +525,8 @@ func (r *FootnoteHTMLRenderer) RegisterFuncs(reg renderer.NodeRendererFuncRegist
 	reg.Register(ast.KindFootnoteList, r.renderFootnoteList)
 }
 
-func (r *FootnoteHTMLRenderer) renderFootnoteLink(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *FootnoteHTMLRenderer) renderFootnoteLink(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		n := node.(*ast.FootnoteLink)
 		is := strconv.Itoa(n.Index)
@@ -556,7 +557,8 @@ func (r *FootnoteHTMLRenderer) renderFootnoteLink(w util.BufWriter, source []byt
 	return gast.WalkContinue, nil
 }
 
-func (r *FootnoteHTMLRenderer) renderFootnoteBacklink(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *FootnoteHTMLRenderer) renderFootnoteBacklink(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		n := node.(*ast.FootnoteBacklink)
 		is := strconv.Itoa(n.Index)
@@ -581,7 +583,8 @@ func (r *FootnoteHTMLRenderer) renderFootnoteBacklink(w util.BufWriter, source [
 	return gast.WalkContinue, nil
 }
 
-func (r *FootnoteHTMLRenderer) renderFootnote(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *FootnoteHTMLRenderer) renderFootnote(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	n := node.(*ast.Footnote)
 	is := strconv.Itoa(n.Index)
 	if entering {
@@ -600,7 +603,8 @@ func (r *FootnoteHTMLRenderer) renderFootnote(w util.BufWriter, source []byte, n
 	return gast.WalkContinue, nil
 }
 
-func (r *FootnoteHTMLRenderer) renderFootnoteList(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *FootnoteHTMLRenderer) renderFootnoteList(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		_, _ = w.WriteString(`<div class="footnotes" role="doc-endnotes"`)
 		if node.Attributes() != nil {

extension/footnote_test.go

@@ -38,12 +38,12 @@ func TestFootnoteOptions(t *testing.T) {
 		),
 		goldmark.WithExtensions(
 			NewFootnote(
-				WithFootnoteIDPrefix([]byte("article12-")),
-				WithFootnoteLinkClass([]byte("link-class")),
-				WithFootnoteBacklinkClass([]byte("backlink-class")),
-				WithFootnoteLinkTitle([]byte("link-title-%%-^^")),
-				WithFootnoteBacklinkTitle([]byte("backlink-title")),
-				WithFootnoteBacklinkHTML([]byte("^")),
+				WithFootnoteIDPrefix("article12-"),
+				WithFootnoteLinkClass("link-class"),
+				WithFootnoteBacklinkClass("backlink-class"),
+				WithFootnoteLinkTitle("link-title-%%-^^"),
+				WithFootnoteBacklinkTitle("backlink-title"),
+				WithFootnoteBacklinkHTML("^"),
 			),
 		),
 	)

extension/linkify.go

@@ -11,9 +11,9 @@ import (
 	"github.com/yuin/goldmark/util"
 )
 
-var wwwURLRegxp = regexp.MustCompile(`^www\.[-a-zA-Z0-9@:%._\+~#=]{1,256}\.[a-z]+(?:[/#?][-a-zA-Z0-9@:%_\+.~#!?&/=\(\);,'">\^{}\[\]` + "`" + `]*)?`)
+var wwwURLRegxp = regexp.MustCompile(`^www\.[-a-zA-Z0-9@:%._\+~#=]{1,256}\.[a-z]+(?:[/#?][-a-zA-Z0-9@:%_\+.~#!?&/=\(\);,'">\^{}\[\]` + "`" + `]*)?`) //nolint:golint,lll
 
-var urlRegexp = regexp.MustCompile(`^(?:http|https|ftp)://[-a-zA-Z0-9@:%._\+~#=]{1,256}\.[a-z]+(?::\d+)?(?:[/#?][-a-zA-Z0-9@:%_+.~#$!?&/=\(\);,'">\^{}\[\]` + "`" + `]*)?`)
+var urlRegexp = regexp.MustCompile(`^(?:http|https|ftp)://[-a-zA-Z0-9@:%._\+~#=]{1,256}\.[a-z]+(?::\d+)?(?:[/#?][-a-zA-Z0-9@:%_+.~#$!?&/=\(\);,'">\^{}\[\]` + "`" + `]*)?`) //nolint:golint,lll
 
 // An LinkifyConfig struct is a data structure that holds configuration of the
 // Linkify extension.
@@ -66,10 +66,12 @@ func (o *withLinkifyAllowedProtocols) SetLinkifyOption(p *LinkifyConfig) {
 // WithLinkifyAllowedProtocols is a functional option that specify allowed
 // protocols in autolinks. Each protocol must end with ':' like
 // 'http:' .
-func WithLinkifyAllowedProtocols(value [][]byte) LinkifyOption {
-	return &withLinkifyAllowedProtocols{
-		value: value,
+func WithLinkifyAllowedProtocols[T []byte | string](value []T) LinkifyOption {
+	opt := &withLinkifyAllowedProtocols{}
+	for _, v := range value {
+		opt.value = append(opt.value, []byte(v))
 	}
+	return opt
 }
 
 type withLinkifyURLRegexp struct {
@@ -92,9 +94,6 @@ func WithLinkifyURLRegexp(value *regexp.Regexp) LinkifyOption {
 	}
 }
 
-// WithLinkifyWWWRegexp is a functional option that specify
-// a pattern of the URL without a protocol.
-// This pattern must start with 'www.' .
 type withLinkifyWWWRegexp struct {
 	value *regexp.Regexp
 }
@@ -107,14 +106,15 @@ func (o *withLinkifyWWWRegexp) SetLinkifyOption(p *LinkifyConfig) {
 	p.WWWRegexp = o.value
 }
 
+// WithLinkifyWWWRegexp is a functional option that specify
+// a pattern of the URL without a protocol.
+// This pattern must start with 'www.' .
 func WithLinkifyWWWRegexp(value *regexp.Regexp) LinkifyOption {
 	return &withLinkifyWWWRegexp{
 		value: value,
 	}
 }
 
-// WithLinkifyWWWRegexp is a functional otpion that specify
-// a pattern of the email address.
 type withLinkifyEmailRegexp struct {
 	value *regexp.Regexp
 }
@@ -127,6 +127,8 @@ func (o *withLinkifyEmailRegexp) SetLinkifyOption(p *LinkifyConfig) {
 	p.EmailRegexp = o.value
 }
 
+// WithLinkifyEmailRegexp is a functional otpion that specify
+// a pattern of the email address.
 func WithLinkifyEmailRegexp(value *regexp.Regexp) LinkifyOption {
 	return &withLinkifyEmailRegexp{
 		value: value,
@@ -303,6 +305,8 @@ type linkify struct {
 // Linkify is an extension that allow you to parse text that seems like a URL.
 var Linkify = &linkify{}
 
+// NewLinkify creates a new [goldmark.Extender] that
+// allow you to parse text that seems like a URL.
 func NewLinkify(opts ...LinkifyOption) goldmark.Extender {
 	return &linkify{
 		options: opts,

extension/linkify_test.go

@@ -29,8 +29,8 @@ func TestLinkifyWithAllowedProtocols(t *testing.T) {
 		),
 		goldmark.WithExtensions(
 			NewLinkify(
-				WithLinkifyAllowedProtocols([][]byte{
-					[]byte("ssh:"),
+				WithLinkifyAllowedProtocols([]string{
+					"ssh:",
 				}),
 				WithLinkifyURLRegexp(
 					regexp.MustCompile(`\w+://[^\s]+`),

extension/package.go

@@ -0,0 +1,2 @@
+// Package extension is a collection of builtin extensions.
+package extension

extension/strikethrough.go

@@ -46,10 +46,11 @@ func (s *strikethroughParser) Trigger() []byte {
 func (s *strikethroughParser) Parse(parent gast.Node, block text.Reader, pc parser.Context) gast.Node {
 	before := block.PrecendingCharacter()
 	line, segment := block.PeekLine()
-	node := parser.ScanDelimiter(line, before, 2, defaultStrikethroughDelimiterProcessor)
-	if node == nil {
+	node := parser.ScanDelimiter(line, before, 1, defaultStrikethroughDelimiterProcessor)
+	if node == nil || node.OriginalLength > 2 || before == '~' {
 		return nil
 	}
+
 	node.Segment = segment.WithStop(segment.Start + node.OriginalLength)
 	block.Advance(node.OriginalLength)
 	pc.PushDelimiter(node)
@@ -85,7 +86,8 @@ func (r *StrikethroughHTMLRenderer) RegisterFuncs(reg renderer.NodeRendererFuncR
 // StrikethroughAttributeFilter defines attribute names which dd elements can have.
 var StrikethroughAttributeFilter = html.GlobalAttributeFilter
 
-func (r *StrikethroughHTMLRenderer) renderStrikethrough(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *StrikethroughHTMLRenderer) renderStrikethrough(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		if n.Attributes() != nil {
 			_, _ = w.WriteString("<del")

extension/table.go

@@ -23,7 +23,7 @@ type escapedPipeCell struct {
 	Transformed bool
 }
 
-// TableCellAlignMethod indicates how are table cells aligned in HTML format.indicates how are table cells aligned in HTML format.
+// TableCellAlignMethod indicates how are table cells aligned in HTML format.
 type TableCellAlignMethod int
 
 const (
@@ -181,13 +181,14 @@ func (b *tableParagraphTransformer) Transform(node *gast.Paragraph, reader text.
 	}
 }
 
-func (b *tableParagraphTransformer) parseRow(segment text.Segment, alignments []ast.Alignment, isHeader bool, reader text.Reader, pc parser.Context) *ast.TableRow {
+func (b *tableParagraphTransformer) parseRow(segment text.Segment,
+	alignments []ast.Alignment, isHeader bool, reader text.Reader, pc parser.Context) *ast.TableRow {
 	source := reader.Source()
+	segment = segment.TrimLeftSpace(source)
+	segment = segment.TrimRightSpace(source)
 	line := segment.Value(source)
 	pos := 0
-	pos += util.TrimLeftSpaceLength(line)
 	limit := len(line)
-	limit -= util.TrimRightSpaceLength(line)
 	row := ast.NewTableRow(alignments)
 	if len(line) > 0 && line[pos] == '|' {
 		pos++
@@ -369,7 +370,8 @@ var TableAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("width"),       // [Deprecated]
 )
 
-func (r *TableHTMLRenderer) renderTable(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *TableHTMLRenderer) renderTable(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		_, _ = w.WriteString("<table")
 		if n.Attributes() != nil {
@@ -391,7 +393,8 @@ var TableHeaderAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("valign"),  // [Deprecated since HTML4] [Obsolete since HTML5]
 )
 
-func (r *TableHTMLRenderer) renderTableHeader(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *TableHTMLRenderer) renderTableHeader(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		_, _ = w.WriteString("<thead")
 		if n.Attributes() != nil {
@@ -418,7 +421,8 @@ var TableRowAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("valign"),  // [Obsolete since HTML5]
 )
 
-func (r *TableHTMLRenderer) renderTableRow(w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *TableHTMLRenderer) renderTableRow(
+	w util.BufWriter, source []byte, n gast.Node, entering bool) (gast.WalkStatus, error) {
 	if entering {
 		_, _ = w.WriteString("<tr")
 		if n.Attributes() != nil {
@@ -445,12 +449,14 @@ var TableThCellAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("charoff"), // [Obsolete since HTML5]
 
 	[]byte("colspan"), // [OK] Number of columns that the cell is to span
-	[]byte("headers"), // [OK] This attribute contains a list of space-separated strings, each corresponding to the id attribute of the <th> elements that apply to this element
+	[]byte("headers"), // [OK] This attribute contains a list of space-separated
+	// strings, each corresponding to the id attribute of the <th> elements that apply to this element
 
 	[]byte("height"), // [Deprecated since HTML4] [Obsolete since HTML5]
 
 	[]byte("rowspan"), // [OK] Number of rows that the cell is to span
-	[]byte("scope"),   // [OK] This enumerated attribute defines the cells that the header (defined in the <th>) element relates to [NOT OK in <td>]
+	[]byte("scope"),   // [OK] This enumerated attribute defines the cells that
+	// the header (defined in the <th>) element relates to [NOT OK in <td>]
 
 	[]byte("valign"), // [Obsolete since HTML5]
 	[]byte("width"),  // [Deprecated since HTML4] [Obsolete since HTML5]
@@ -466,7 +472,8 @@ var TableTdCellAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("charoff"), // [Obsolete since HTML5]
 
 	[]byte("colspan"), // [OK] Number of columns that the cell is to span
-	[]byte("headers"), // [OK] This attribute contains a list of space-separated strings, each corresponding to the id attribute of the <th> elements that apply to this element
+	[]byte("headers"), // [OK] This attribute contains a list of space-separated
+	// strings, each corresponding to the id attribute of the <th> elements that apply to this element
 
 	[]byte("height"), // [Deprecated since HTML4] [Obsolete since HTML5]
 
@@ -477,14 +484,15 @@ var TableTdCellAttributeFilter = html.GlobalAttributeFilter.Extend(
 	[]byte("width"),  // [Deprecated since HTML4] [Obsolete since HTML5]
 )
 
-func (r *TableHTMLRenderer) renderTableCell(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *TableHTMLRenderer) renderTableCell(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	n := node.(*ast.TableCell)
 	tag := "td"
 	if n.Parent().Kind() == ast.KindTableHeader {
 		tag = "th"
 	}
 	if entering {
-		fmt.Fprintf(w, "<%s", tag)
+		_, _ = fmt.Fprintf(w, "<%s", tag)
 		if n.Alignment != ast.AlignNone {
 			amethod := r.TableConfig.TableCellAlignMethod
 			if amethod == TableCellAlignDefault {
@@ -497,7 +505,7 @@ func (r *TableHTMLRenderer) renderTableCell(w util.BufWriter, source []byte, nod
 			switch amethod {
 			case TableCellAlignAttribute:
 				if _, ok := n.AttributeString("align"); !ok { // Skip align render if overridden
-					fmt.Fprintf(w, ` align="%s"`, n.Alignment.String())
+					_, _ = fmt.Fprintf(w, ` align="%s"`, n.Alignment.String())
 				}
 			case TableCellAlignStyle:
 				v, ok := n.AttributeString("style")
@@ -520,7 +528,7 @@ func (r *TableHTMLRenderer) renderTableCell(w util.BufWriter, source []byte, nod
 		}
 		_ = w.WriteByte('>')
 	} else {
-		fmt.Fprintf(w, "</%s>\n", tag)
+		_, _ = fmt.Fprintf(w, "</%s>\n", tag)
 	}
 	return gast.WalkContinue, nil
 }

extension/table_test.go

@@ -355,3 +355,40 @@ bar | baz
 		t,
 	)
 }
+
+func TestTableFuzzedPanics(t *testing.T) {
+	markdown := goldmark.New(
+		goldmark.WithRendererOptions(
+			html.WithXHTML(),
+			html.WithUnsafe(),
+		),
+		goldmark.WithExtensions(
+			NewTable(),
+		),
+	)
+	testutil.DoTestCase(
+		markdown,
+		testutil.MarkdownTestCase{
+			No:          1,
+			Description: "This should not panic",
+			Markdown:    "* 0\n-|\n\t0",
+			Expected: `<ul>
+<li>
+<table>
+<thead>
+<tr>
+<th>0</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>0</td>
+</tr>
+</tbody>
+</table>
+</li>
+</ul>`,
+		},
+		t,
+	)
+}

extension/tasklist.go

@@ -1,6 +1,8 @@
 package extension
 
 import (
+	"regexp"
+
 	"github.com/yuin/goldmark"
 	gast "github.com/yuin/goldmark/ast"
 	"github.com/yuin/goldmark/extension/ast"
@@ -9,7 +11,6 @@ import (
 	"github.com/yuin/goldmark/renderer/html"
 	"github.com/yuin/goldmark/text"
 	"github.com/yuin/goldmark/util"
-	"regexp"
 )
 
 var taskListRegexp = regexp.MustCompile(`^\[([\sxX])\]\s*`)
@@ -40,6 +41,9 @@ func (s *taskCheckBoxParser) Parse(parent gast.Node, block text.Reader, pc parse
 		return nil
 	}
 
+	if parent.HasChildren() {
+		return nil
+	}
 	if _, ok := parent.Parent().(*gast.ListItem); !ok {
 		return nil
 	}
@@ -80,21 +84,22 @@ func (r *TaskCheckBoxHTMLRenderer) RegisterFuncs(reg renderer.NodeRendererFuncRe
 	reg.Register(ast.KindTaskCheckBox, r.renderTaskCheckBox)
 }
 
-func (r *TaskCheckBoxHTMLRenderer) renderTaskCheckBox(w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
+func (r *TaskCheckBoxHTMLRenderer) renderTaskCheckBox(
+	w util.BufWriter, source []byte, node gast.Node, entering bool) (gast.WalkStatus, error) {
 	if !entering {
 		return gast.WalkContinue, nil
 	}
 	n := node.(*ast.TaskCheckBox)
 
 	if n.IsChecked {
-		w.WriteString(`<input checked="" disabled="" type="checkbox"`)
+		_, _ = w.WriteString(`<input checked="" disabled="" type="checkbox"`)
 	} else {
-		w.WriteString(`<input disabled="" type="checkbox"`)
+		_, _ = w.WriteString(`<input disabled="" type="checkbox"`)
 	}
 	if r.XHTML {
-		w.WriteString(" /> ")
+		_, _ = w.WriteString(" /> ")
 	} else {
-		w.WriteString("> ")
+		_, _ = w.WriteString("> ")
 	}
 	return gast.WalkContinue, nil
 }

extension/typographer.go

@@ -36,25 +36,25 @@ func getUnclosedCounter(pc parser.Context) *unclosedCounter {
 type TypographicPunctuation int
 
 const (
-	// LeftSingleQuote is '
+	// LeftSingleQuote is ' .
 	LeftSingleQuote TypographicPunctuation = iota + 1
-	// RightSingleQuote is '
+	// RightSingleQuote is ' .
 	RightSingleQuote
-	// LeftDoubleQuote is "
+	// LeftDoubleQuote is " .
 	LeftDoubleQuote
-	// RightDoubleQuote is "
+	// RightDoubleQuote is " .
 	RightDoubleQuote
-	// EnDash is --
+	// EnDash is -- .
 	EnDash
-	// EmDash is ---
+	// EmDash is --- .
 	EmDash
-	// Ellipsis is ...
+	// Ellipsis is ... .
 	Ellipsis
-	// LeftAngleQuote is <<
+	// LeftAngleQuote is << .
 	LeftAngleQuote
-	// RightAngleQuote is >>
+	// RightAngleQuote is >> .
 	RightAngleQuote
-	// Apostrophe is '
+	// Apostrophe is ' .
 	Apostrophe
 
 	typographicPunctuationMax
@@ -115,10 +115,10 @@ func (o *withTypographicSubstitutions) SetTypographerOption(p *TypographerConfig
 
 // WithTypographicSubstitutions is a functional otpion that specify replacement text
 // for punctuations.
-func WithTypographicSubstitutions(values map[TypographicPunctuation][]byte) TypographerOption {
+func WithTypographicSubstitutions[T []byte | string](values map[TypographicPunctuation]T) TypographerOption {
 	replacements := newDefaultSubstitutions()
 	for k, v := range values {
-		replacements[k] = v
+		replacements[k] = []byte(v)
 	}
 
 	return &withTypographicSubstitutions{replacements}
@@ -218,7 +218,8 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 		if c == '\'' {
 			if s.Substitutions[Apostrophe] != nil {
 				// Handle decade abbrevations such as '90s
-				if d.CanOpen && !d.CanClose && len(line) > 3 && util.IsNumeric(line[1]) && util.IsNumeric(line[2]) && line[3] == 's' {
+				if d.CanOpen && !d.CanClose && len(line) > 3 &&
+					util.IsNumeric(line[1]) && util.IsNumeric(line[2]) && line[3] == 's' {
 					after := rune(' ')
 					if len(line) > 4 {
 						after = util.ToRune(line, 4)
@@ -231,7 +232,8 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 					}
 				}
 				// special cases: 'twas, 'em, 'net
-				if len(line) > 1 && (unicode.IsPunct(before) || unicode.IsSpace(before)) && (line[1] == 't' || line[1] == 'e' || line[1] == 'n' || line[1] == 'l') {
+				if len(line) > 1 && (unicode.IsPunct(before) || unicode.IsSpace(before)) &&
+					(line[1] == 't' || line[1] == 'e' || line[1] == 'n' || line[1] == 'l') {
 					node := gast.NewString(s.Substitutions[Apostrophe])
 					node.SetCode(true)
 					block.Advance(1)
@@ -239,7 +241,8 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 				}
 				// Convert normal apostrophes. This is probably more flexible than necessary but
 				// converts any apostrophe in between two alphanumerics.
-				if len(line) > 1 && (unicode.IsDigit(before) || unicode.IsLetter(before)) && (unicode.IsLetter(util.ToRune(line, 1))) {
+				if len(line) > 1 && (unicode.IsDigit(before) || unicode.IsLetter(before)) &&
+					(unicode.IsLetter(util.ToRune(line, 1))) {
 					node := gast.NewString(s.Substitutions[Apostrophe])
 					node.SetCode(true)
 					block.Advance(1)
@@ -249,11 +252,14 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 			if s.Substitutions[LeftSingleQuote] != nil && d.CanOpen && !d.CanClose {
 				nt := LeftSingleQuote
 				// special cases: Alice's, I'm, Don't, You'd
-				if len(line) > 1 && (line[1] == 's' || line[1] == 'm' || line[1] == 't' || line[1] == 'd') && (len(line) < 3 || util.IsPunct(line[2]) || util.IsSpace(line[2])) {
+				if len(line) > 1 && (line[1] == 's' || line[1] == 'm' || line[1] == 't' || line[1] == 'd') &&
+					(len(line) < 3 || util.IsPunct(line[2]) || util.IsSpace(line[2])) {
 					nt = RightSingleQuote
 				}
 				// special cases: I've, I'll, You're
-				if len(line) > 2 && ((line[1] == 'v' && line[2] == 'e') || (line[1] == 'l' && line[2] == 'l') || (line[1] == 'r' && line[2] == 'e')) && (len(line) < 4 || util.IsPunct(line[3]) || util.IsSpace(line[3])) {
+				if len(line) > 2 && ((line[1] == 'v' && line[2] == 'e') ||
+					(line[1] == 'l' && line[2] == 'l') || (line[1] == 'r' && line[2] == 'e')) &&
+					(len(line) < 4 || util.IsPunct(line[3]) || util.IsSpace(line[3])) {
 					nt = RightSingleQuote
 				}
 				if nt == LeftSingleQuote {
@@ -266,8 +272,9 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 				return node
 			}
 			if s.Substitutions[RightSingleQuote] != nil {
-				// plural possesives and abbreviations: Smiths', doin'
-				if len(line) > 1 && unicode.IsSpace(util.ToRune(line, 0)) || unicode.IsPunct(util.ToRune(line, 0)) && (len(line) > 2 && !unicode.IsDigit(util.ToRune(line, 1))) {
+				// plural possesive and abbreviations: Smiths', doin'
+				if len(line) > 1 && unicode.IsSpace(util.ToRune(line, 0)) || unicode.IsPunct(util.ToRune(line, 0)) &&
+					(len(line) > 2 && !unicode.IsDigit(util.ToRune(line, 1))) {
 					node := gast.NewString(s.Substitutions[RightSingleQuote])
 					node.SetCode(true)
 					block.Advance(1)
@@ -276,7 +283,8 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 			}
 			if s.Substitutions[RightSingleQuote] != nil && counter.Single > 0 {
 				isClose := d.CanClose && !d.CanOpen
-				maybeClose := d.CanClose && d.CanOpen && len(line) > 1 && unicode.IsPunct(util.ToRune(line, 1)) && (len(line) == 2 || (len(line) > 2 && util.IsPunct(line[2]) || util.IsSpace(line[2])))
+				maybeClose := d.CanClose && d.CanOpen && len(line) > 1 && unicode.IsPunct(util.ToRune(line, 1)) &&
+					(len(line) == 2 || (len(line) > 2 && util.IsPunct(line[2]) || util.IsSpace(line[2])))
 				if isClose || maybeClose {
 					node := gast.NewString(s.Substitutions[RightSingleQuote])
 					node.SetCode(true)
@@ -296,7 +304,8 @@ func (s *typographerParser) Parse(parent gast.Node, block text.Reader, pc parser
 			}
 			if s.Substitutions[RightDoubleQuote] != nil && counter.Double > 0 {
 				isClose := d.CanClose && !d.CanOpen
-				maybeClose := d.CanClose && d.CanOpen && len(line) > 1 && (unicode.IsPunct(util.ToRune(line, 1))) && (len(line) == 2 || (len(line) > 2 && util.IsPunct(line[2]) || util.IsSpace(line[2])))
+				maybeClose := d.CanClose && d.CanOpen && len(line) > 1 && (unicode.IsPunct(util.ToRune(line, 1))) &&
+					(len(line) == 2 || (len(line) > 2 && util.IsPunct(line[2]) || util.IsSpace(line[2])))
 				if isClose || maybeClose {
 					// special case: "Monitor 21""
 					if len(line) > 1 && line[1] == '"' && unicode.IsDigit(before) {

fuzz/fuzz_test.go

@@ -3,7 +3,7 @@ package fuzz
 import (
 	"bytes"
 	"encoding/json"
-	"io/ioutil"
+	"os"
 	"testing"
 
 	"github.com/yuin/goldmark"
@@ -42,7 +42,7 @@ func fuzz(f *testing.F) {
 }
 
 func FuzzDefault(f *testing.F) {
-	bs, err := ioutil.ReadFile("../_test/spec.json")
+	bs, err := os.ReadFile("../_test/spec.json")
 	if err != nil {
 		panic(err)
 	}

go.mod

@@ -1,3 +1,3 @@
 module github.com/yuin/goldmark
 
-go 1.18
+go 1.19

markdown.go

@@ -2,12 +2,13 @@
 package goldmark
 
 import (
+	"io"
+
 	"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"
-	"io"
 )
 
 // DefaultParser returns a new Parser that is configured by default values.

parser/attribute.go

@@ -12,7 +12,7 @@ import (
 var attrNameID = []byte("id")
 var attrNameClass = []byte("class")
 
-// An Attribute is an attribute of the markdown elements
+// An Attribute is an attribute of the markdown elements.
 type Attribute struct {
 	Name  []byte
 	Value interface{}
@@ -93,7 +93,8 @@ func parseAttribute(reader text.Reader) (Attribute, bool) {
 		// CommonMark is basically defined for XHTML(even though it is legacy).
 		// So we restrict id characters.
 		for ; i < len(line) && !util.IsSpace(line[i]) &&
-			(!util.IsPunct(line[i]) || line[i] == '_' || line[i] == '-' || line[i] == ':' || line[i] == '.'); i++ {
+			(!util.IsPunct(line[i]) || line[i] == '_' ||
+				line[i] == '-' || line[i] == ':' || line[i] == '.'); i++ {
 		}
 		name := attrNameClass
 		if c == '#' {
@@ -145,7 +146,7 @@ func parseAttributeValue(reader text.Reader) (interface{}, bool) {
 	reader.SkipSpaces()
 	c := reader.Peek()
 	var value interface{}
-	ok := false
+	var ok bool
 	switch c {
 	case text.EOF:
 		return Attribute{}, false
@@ -244,7 +245,7 @@ func scanAttributeDecimal(reader text.Reader, w io.ByteWriter) {
 	for {
 		c := reader.Peek()
 		if util.IsNumeric(c) {
-			w.WriteByte(c)
+			_ = w.WriteByte(c)
 		} else {
 			return
 		}
@@ -286,7 +287,7 @@ func parseAttributeNumber(reader text.Reader) (float64, bool) {
 		}
 		scanAttributeDecimal(reader, &buf)
 	}
-	f, err := strconv.ParseFloat(buf.String(), 10)
+	f, err := strconv.ParseFloat(buf.String(), 64)
 	if err != nil {
 		return 0, false
 	}

parser/atx_heading.go

@@ -13,7 +13,7 @@ type HeadingConfig struct {
 }
 
 // SetOption implements SetOptioner.
-func (b *HeadingConfig) SetOption(name OptionName, value interface{}) {
+func (b *HeadingConfig) SetOption(name OptionName, _ interface{}) {
 	switch name {
 	case optAutoHeadingID:
 		b.AutoHeadingID = true
@@ -135,7 +135,9 @@ func (b *atxHeadingParser) Open(parent ast.Node, reader text.Reader, pc Context)
 				for _, attr := range attrs {
 					node.SetAttribute(attr.Name, attr.Value)
 				}
-				node.Lines().Append(text.NewSegment(segment.Start+start+1-segment.Padding, segment.Start+closureOpen-segment.Padding))
+				node.Lines().Append(text.NewSegment(
+					segment.Start+start+1-segment.Padding,
+					segment.Start+closureOpen-segment.Padding))
 			}
 		}
 	}

parser/blockquote.go

@@ -28,12 +28,13 @@ func (b *blockquoteParser) process(reader text.Reader) bool {
 		reader.Advance(pos)
 		return true
 	}
-	if line[pos] == ' ' || line[pos] == '\t' {
-		pos++
-	}
 	reader.Advance(pos)
-	if line[pos-1] == '\t' {
-		reader.SetPadding(2)
+	if line[pos] == ' ' || line[pos] == '\t' {
+		padding := 0
+		if line[pos] == '\t' {
+			padding = util.TabWidth(reader.LineOffset()) - 1
+		}
+		reader.AdvanceAndSetPadding(1, padding)
 	}
 	return true
 }

parser/code_block.go

@@ -35,6 +35,7 @@ func (b *codeBlockParser) Open(parent ast.Node, reader text.Reader, pc Context)
 	if segment.Padding != 0 {
 		preserveLeadingTabInCodeBlock(&segment, reader, 0)
 	}
+	segment.ForceNewline = true
 	node.Lines().Append(segment)
 	reader.Advance(segment.Len() - 1)
 	return node, NoChildren
@@ -59,6 +60,7 @@ func (b *codeBlockParser) Continue(node ast.Node, reader text.Reader, pc Context
 		preserveLeadingTabInCodeBlock(&segment, reader, 0)
 	}
 
+	segment.ForceNewline = true
 	node.Lines().Append(segment)
 	reader.Advance(segment.Len() - 1)
 	return Continue | NoChildren

parser/delimiter.go

@@ -66,12 +66,12 @@ func (d *Delimiter) Dump(source []byte, level int) {
 
 var kindDelimiter = ast.NewNodeKind("Delimiter")
 
-// Kind implements Node.Kind
+// Kind implements Node.Kind.
 func (d *Delimiter) Kind() ast.NodeKind {
 	return kindDelimiter
 }
 
-// Text implements Node.Text
+// Text implements Node.Text.
 func (d *Delimiter) Text(source []byte) []byte {
 	return d.Segment.Value(source)
 }
@@ -126,7 +126,7 @@ func ScanDelimiter(line []byte, before rune, min int, processor DelimiterProcess
 			after = util.ToRune(line, j)
 		}
 
-		canOpen, canClose := false, false
+		var canOpen, canClose bool
 		beforeIsPunctuation := util.IsPunctRune(before)
 		beforeIsWhitespace := util.IsSpaceRune(before)
 		afterIsPunctuation := util.IsPunctRune(after)

parser/fcode_block.go

@@ -100,6 +100,7 @@ func (b *fencedCodeBlockParser) Continue(node ast.Node, reader text.Reader, pc C
 	if padding != 0 {
 		preserveLeadingTabInCodeBlock(&seg, reader, fdata.indent)
 	}
+	seg.ForceNewline = true // EOF as newline
 	node.Lines().Append(seg)
 	reader.AdvanceAndSetPadding(segment.Stop-segment.Start-pos-1, padding)
 	return Continue | NoChildren

parser/html_block.go

@@ -61,8 +61,8 @@ var allowedBlockTags = map[string]bool{
 	"option":     true,
 	"p":          true,
 	"param":      true,
+	"search":     true,
 	"section":    true,
-	"source":     true,
 	"summary":    true,
 	"table":      true,
 	"tbody":      true,
@@ -76,7 +76,7 @@ var allowedBlockTags = map[string]bool{
 	"ul":         true,
 }
 
-var htmlBlockType1OpenRegexp = regexp.MustCompile(`(?i)^[ ]{0,3}<(script|pre|style|textarea)(?:\s.*|>.*|/>.*|)(?:\r\n|\n)?$`)
+var htmlBlockType1OpenRegexp = regexp.MustCompile(`(?i)^[ ]{0,3}<(script|pre|style|textarea)(?:\s.*|>.*|/>.*|)(?:\r\n|\n)?$`) //nolint:golint,lll
 var htmlBlockType1CloseRegexp = regexp.MustCompile(`(?i)^.*</(?:script|pre|style|textarea)>.*`)
 
 var htmlBlockType2OpenRegexp = regexp.MustCompile(`^[ ]{0,3}<!\-\-`)
@@ -91,9 +91,9 @@ var htmlBlockType4Close = []byte{'>'}
 var htmlBlockType5OpenRegexp = regexp.MustCompile(`^[ ]{0,3}<\!\[CDATA\[`)
 var htmlBlockType5Close = []byte{']', ']', '>'}
 
-var htmlBlockType6Regexp = regexp.MustCompile(`^[ ]{0,3}<(?:/[ ]*)?([a-zA-Z]+[a-zA-Z0-9\-]*)(?:[ ].*|>.*|/>.*|)(?:\r\n|\n)?$`)
+var htmlBlockType6Regexp = regexp.MustCompile(`^[ ]{0,3}<(?:/[ ]*)?([a-zA-Z]+[a-zA-Z0-9\-]*)(?:[ ].*|>.*|/>.*|)(?:\r\n|\n)?$`) //nolint:golint,lll
 
-var htmlBlockType7Regexp = regexp.MustCompile(`^[ ]{0,3}<(/[ ]*)?([a-zA-Z]+[a-zA-Z0-9\-]*)(` + attributePattern + `*)[ ]*(?:>|/>)[ ]*(?:\r\n|\n)?$`)
+var htmlBlockType7Regexp = regexp.MustCompile(`^[ ]{0,3}<(/[ ]*)?([a-zA-Z]+[a-zA-Z0-9\-]*)(` + attributePattern + `*)[ ]*(?:>|/>)[ ]*(?:\r\n|\n)?$`) //nolint:golint,lll
 
 type htmlBlockParser struct {
 }
@@ -135,7 +135,8 @@ func (b *htmlBlockParser) Open(parent ast.Node, reader text.Reader, pc Context)
 		_, ok := allowedBlockTags[tagName]
 		if ok {
 			node = ast.NewHTMLBlock(ast.HTMLBlockType6)
-		} else if tagName != "script" && tagName != "style" && tagName != "pre" && !ast.IsParagraph(last) && !(isCloseTag && hasAttr) { // type 7 can not interrupt paragraph
+		} else if tagName != "script" && tagName != "style" &&
+			tagName != "pre" && !ast.IsParagraph(last) && !(isCloseTag && hasAttr) { // type 7 can not interrupt paragraph
 			node = ast.NewHTMLBlock(ast.HTMLBlockType7)
 		}
 	}

parser/link.go

@@ -126,13 +126,13 @@ func (s *linkParser) Parse(parent ast.Node, block text.Reader, pc Context) ast.N
 	if line[0] == '!' {
 		if len(line) > 1 && line[1] == '[' {
 			block.Advance(1)
-			pc.Set(linkBottom, pc.LastDelimiter())
+			pushLinkBottom(pc)
 			return processLinkLabelOpen(block, segment.Start+1, true, pc)
 		}
 		return nil
 	}
 	if line[0] == '[' {
-		pc.Set(linkBottom, pc.LastDelimiter())
+		pushLinkBottom(pc)
 		return processLinkLabelOpen(block, segment.Start, false, pc)
 	}
 
@@ -143,6 +143,7 @@ func (s *linkParser) Parse(parent ast.Node, block text.Reader, pc Context) ast.N
 	}
 	last := tlist.(*linkLabelState).Last
 	if last == nil {
+		_ = popLinkBottom(pc)
 		return nil
 	}
 	block.Advance(1)
@@ -151,11 +152,13 @@ func (s *linkParser) Parse(parent ast.Node, block text.Reader, pc Context) ast.N
 	//  > A link label can have at most 999 characters inside the square brackets.
 	if linkLabelStateLength(tlist.(*linkLabelState)) > 998 {
 		ast.MergeOrReplaceTextSegment(last.Parent(), last, last.Segment)
+		_ = popLinkBottom(pc)
 		return nil
 	}
 
 	if !last.IsImage && s.containsLink(last) { // a link in a link text is not allowed
 		ast.MergeOrReplaceTextSegment(last.Parent(), last, last.Segment)
+		_ = popLinkBottom(pc)
 		return nil
 	}
 
@@ -169,6 +172,7 @@ func (s *linkParser) Parse(parent ast.Node, block text.Reader, pc Context) ast.N
 		link, hasValue = s.parseReferenceLink(parent, last, block, pc)
 		if link == nil && hasValue {
 			ast.MergeOrReplaceTextSegment(last.Parent(), last, last.Segment)
+			_ = popLinkBottom(pc)
 			return nil
 		}
 	}
@@ -182,12 +186,14 @@ func (s *linkParser) Parse(parent ast.Node, block text.Reader, pc Context) ast.N
 		//  > A link label can have at most 999 characters inside the square brackets.
 		if len(maybeReference) > 999 {
 			ast.MergeOrReplaceTextSegment(last.Parent(), last, last.Segment)
+			_ = popLinkBottom(pc)
 			return nil
 		}
 
 		ref, ok := pc.Reference(util.ToLinkReference(maybeReference))
 		if !ok {
 			ast.MergeOrReplaceTextSegment(last.Parent(), last, last.Segment)
+			_ = popLinkBottom(pc)
 			return nil
 		}
 		link = ast.NewLink()
@@ -230,11 +236,7 @@ func processLinkLabelOpen(block text.Reader, pos int, isImage bool, pc Context)
 }
 
 func (s *linkParser) processLinkLabel(parent ast.Node, link *ast.Link, last *linkLabelState, pc Context) {
-	var bottom ast.Node
-	if v := pc.Get(linkBottom); v != nil {
-		bottom = v.(ast.Node)
-	}
-	pc.Set(linkBottom, nil)
+	bottom := popLinkBottom(pc)
 	ProcessDelimiters(bottom, pc)
 	for c := last.NextSibling(); c != nil; {
 		next := c.NextSibling()
@@ -250,7 +252,8 @@ var linkFindClosureOptions text.FindClosureOptions = text.FindClosureOptions{
 	Advance: true,
 }
 
-func (s *linkParser) parseReferenceLink(parent ast.Node, last *linkLabelState, block text.Reader, pc Context) (*ast.Link, bool) {
+func (s *linkParser) parseReferenceLink(parent ast.Node, last *linkLabelState,
+	block text.Reader, pc Context) (*ast.Link, bool) {
 	_, orgpos := block.Position()
 	block.Advance(1) // skip '['
 	segments, found := block.FindClosure('[', ']', linkFindClosureOptions)
@@ -394,6 +397,43 @@ func parseLinkTitle(block text.Reader) ([]byte, bool) {
 	return nil, false
 }
 
+func pushLinkBottom(pc Context) {
+	bottoms := pc.Get(linkBottom)
+	b := pc.LastDelimiter()
+	if bottoms == nil {
+		pc.Set(linkBottom, b)
+		return
+	}
+	if s, ok := bottoms.([]ast.Node); ok {
+		pc.Set(linkBottom, append(s, b))
+		return
+	}
+	pc.Set(linkBottom, []ast.Node{bottoms.(ast.Node), b})
+}
+
+func popLinkBottom(pc Context) ast.Node {
+	bottoms := pc.Get(linkBottom)
+	if bottoms == nil {
+		return nil
+	}
+	if v, ok := bottoms.(ast.Node); ok {
+		pc.Set(linkBottom, nil)
+		return v
+	}
+	s := bottoms.([]ast.Node)
+	v := s[len(s)-1]
+	n := s[0 : len(s)-1]
+	switch len(n) {
+	case 0:
+		pc.Set(linkBottom, nil)
+	case 1:
+		pc.Set(linkBottom, n[0])
+	default:
+		pc.Set(linkBottom, s[0:len(s)-1])
+	}
+	return v
+}
+
 func (s *linkParser) CloseBlock(parent ast.Node, block text.Reader, pc Context) {
 	pc.Set(linkBottom, nil)
 	tlist := pc.Get(linkLabelStateKey)

parser/list.go

@@ -22,7 +22,7 @@ var listItemFlagValue interface{} = true
 
 // Same as
 // `^(([ ]*)([\-\*\+]))(\s+.*)?\n?$`.FindSubmatchIndex or
-// `^(([ ]*)(\d{1,9}[\.\)]))(\s+.*)?\n?$`.FindSubmatchIndex
+// `^(([ ]*)(\d{1,9}[\.\)]))(\s+.*)?\n?$`.FindSubmatchIndex.
 func parseListItem(line []byte) ([6]int, listItemType) {
 	i := 0
 	l := len(line)
@@ -89,7 +89,7 @@ func matchesListItem(source []byte, strict bool) ([6]int, listItemType) {
 }
 
 func calcListOffset(source []byte, match [6]int) int {
-	offset := 0
+	var offset int
 	if match[4] < 0 || util.IsBlank(source[match[4]:]) { // list item starts with a blank line
 		offset = 1
 	} else {
@@ -250,14 +250,14 @@ func (b *listParser) Close(node ast.Node, reader text.Reader, pc Context) {
 	for c := node.FirstChild(); c != nil && list.IsTight; c = c.NextSibling() {
 		if c.FirstChild() != nil && c.FirstChild() != c.LastChild() {
 			for c1 := c.FirstChild().NextSibling(); c1 != nil; c1 = c1.NextSibling() {
-				if bl, ok := c1.(ast.Node); ok && bl.HasBlankPreviousLines() {
+				if c1.HasBlankPreviousLines() {
 					list.IsTight = false
 					break
 				}
 			}
 		}
 		if c != node.FirstChild() {
-			if bl, ok := c.(ast.Node); ok && bl.HasBlankPreviousLines() {
+			if c.HasBlankPreviousLines() {
 				list.IsTight = false
 			}
 		}

parser/list_item.go

@@ -58,7 +58,7 @@ func (b *listItemParser) Continue(node ast.Node, reader text.Reader, pc Context)
 	}
 
 	offset := lastOffset(node.Parent())
-	isEmpty := node.ChildCount() == 0
+	isEmpty := node.ChildCount() == 0 && pc.Get(emptyListItemWithBlankLines) != nil
 	indent, _ := util.IndentWidth(line, reader.LineOffset())
 	if (isEmpty || indent < offset) && indent < 4 {
 		_, typ := matchesListItem(line, true)

parser/parser.go

@@ -403,7 +403,8 @@ func (p *parseContext) IsInLinkLabel() bool {
 type State int
 
 const (
-	none State = 1 << iota
+	// None is a default value of the [State].
+	None State = 1 << iota
 
 	// Continue indicates parser can continue parsing.
 	Continue
@@ -881,6 +882,7 @@ func (p *parser) Parse(reader text.Reader, opts ...ParseOption) ast.Node {
 	for _, at := range p.astTransformers {
 		at.Transform(root, reader, pc)
 	}
+
 	// root.Dump(reader.Source(), 0)
 	return root
 }
@@ -1049,7 +1051,7 @@ func isBlankLine(lineNum, level int, stats []lineStat) bool {
 func (p *parser) parseBlocks(parent ast.Node, reader text.Reader, pc Context) {
 	pc.SetOpenedBlocks([]Block{})
 	blankLines := make([]lineStat, 0, 128)
-	isBlank := false
+	var isBlank bool
 	for { // process blocks separated by blank lines
 		_, lines, ok := reader.SkipBlankLines()
 		if !ok {
@@ -1152,18 +1154,23 @@ func (p *parser) parseBlock(block text.BlockReader, parent ast.Node, pc Context)
 			break
 		}
 		lineLength := len(line)
-		var lineBreakFlags uint8 = 0
+		var lineBreakFlags uint8
 		hasNewLine := line[lineLength-1] == '\n'
-		if ((lineLength >= 3 && line[lineLength-2] == '\\' && line[lineLength-3] != '\\') || (lineLength == 2 && line[lineLength-2] == '\\')) && hasNewLine { // ends with \\n
+		if ((lineLength >= 3 && line[lineLength-2] == '\\' &&
+			line[lineLength-3] != '\\') || (lineLength == 2 && line[lineLength-2] == '\\')) && hasNewLine { // ends with \\n
 			lineLength -= 2
 			lineBreakFlags |= lineBreakHard | lineBreakVisible
-		} else if ((lineLength >= 4 && line[lineLength-3] == '\\' && line[lineLength-2] == '\r' && line[lineLength-4] != '\\') || (lineLength == 3 && line[lineLength-3] == '\\' && line[lineLength-2] == '\r')) && hasNewLine { // ends with \\r\n
+		} else if ((lineLength >= 4 && line[lineLength-3] == '\\' && line[lineLength-2] == '\r' &&
+			line[lineLength-4] != '\\') || (lineLength == 3 && line[lineLength-3] == '\\' && line[lineLength-2] == '\r')) &&
+			hasNewLine { // ends with \\r\n
 			lineLength -= 3
 			lineBreakFlags |= lineBreakHard | lineBreakVisible
-		} else if lineLength >= 3 && line[lineLength-3] == ' ' && line[lineLength-2] == ' ' && hasNewLine { // ends with [space][space]\n
+		} else if lineLength >= 3 && line[lineLength-3] == ' ' && line[lineLength-2] == ' ' &&
+			hasNewLine { // ends with [space][space]\n
 			lineLength -= 3
 			lineBreakFlags |= lineBreakHard
-		} else if lineLength >= 4 && line[lineLength-4] == ' ' && line[lineLength-3] == ' ' && line[lineLength-2] == '\r' && hasNewLine { // ends with [space][space]\r\n
+		} else if lineLength >= 4 && line[lineLength-4] == ' ' && line[lineLength-3] == ' ' &&
+			line[lineLength-2] == '\r' && hasNewLine { // ends with [space][space]\r\n
 			lineLength -= 4
 			lineBreakFlags |= lineBreakHard
 		} else if hasNewLine {
@@ -1250,4 +1257,5 @@ func (p *parser) parseBlock(block text.BlockReader, parent ast.Node, pc Context)
 	for _, ip := range p.closeBlockers {
 		ip.CloseBlock(parent, block, pc)
 	}
+
 }

parser/raw_html.go

@@ -15,7 +15,7 @@ type rawHTMLParser struct {
 var defaultRawHTMLParser = &rawHTMLParser{}
 
 // NewRawHTMLParser return a new InlineParser that can parse
-// inline htmls
+// inline htmls.
 func NewRawHTMLParser() InlineParser {
 	return defaultRawHTMLParser
 }
@@ -49,7 +49,7 @@ func (s *rawHTMLParser) Parse(parent ast.Node, block text.Reader, pc Context) as
 
 var tagnamePattern = `([A-Za-z][A-Za-z0-9-]*)`
 var spaceOrOneNewline = `(?:[ \t]|(?:\r\n|\n){0,1})`
-var attributePattern = `(?:[\r\n \t]+[a-zA-Z_:][a-zA-Z0-9:._-]*(?:[\r\n \t]*=[\r\n \t]*(?:[^\"'=<>` + "`" + `\x00-\x20]+|'[^']*'|"[^"]*"))?)`
+var attributePattern = `(?:[\r\n \t]+[a-zA-Z_:][a-zA-Z0-9:._-]*(?:[\r\n \t]*=[\r\n \t]*(?:[^\"'=<>` + "`" + `\x00-\x20]+|'[^']*'|"[^"]*"))?)` //nolint:golint,lll
 var openTagRegexp = regexp.MustCompile("^<" + tagnamePattern + attributePattern + `*` + spaceOrOneNewline + `*/?>`)
 var closeTagRegexp = regexp.MustCompile("^</" + tagnamePattern + spaceOrOneNewline + `*>`)
 
@@ -58,47 +58,38 @@ var closeProcessingInstruction = []byte("?>")
 var openCDATA = []byte("<![CDATA[")
 var closeCDATA = []byte("]]>")
 var closeDecl = []byte(">")
-var emptyComment = []byte("<!---->")
-var invalidComment1 = []byte("<!-->")
-var invalidComment2 = []byte("<!--->")
+var emptyComment1 = []byte("<!-->")
+var emptyComment2 = []byte("<!--->")
 var openComment = []byte("<!--")
 var closeComment = []byte("-->")
-var doubleHyphen = []byte("--")
 
 func (s *rawHTMLParser) parseComment(block text.Reader, pc Context) ast.Node {
 	savedLine, savedSegment := block.Position()
 	node := ast.NewRawHTML()
 	line, segment := block.PeekLine()
-	if bytes.HasPrefix(line, emptyComment) {
-		node.Segments.Append(segment.WithStop(segment.Start + len(emptyComment)))
-		block.Advance(len(emptyComment))
+	if bytes.HasPrefix(line, emptyComment1) {
+		node.Segments.Append(segment.WithStop(segment.Start + len(emptyComment1)))
+		block.Advance(len(emptyComment1))
 		return node
 	}
-	if bytes.HasPrefix(line, invalidComment1) || bytes.HasPrefix(line, invalidComment2) {
-		return nil
+	if bytes.HasPrefix(line, emptyComment2) {
+		node.Segments.Append(segment.WithStop(segment.Start + len(emptyComment2)))
+		block.Advance(len(emptyComment2))
+		return node
 	}
 	offset := len(openComment)
 	line = line[offset:]
 	for {
-		hindex := bytes.Index(line, doubleHyphen)
-		if hindex > -1 {
-			hindex += offset
-		}
-		index := bytes.Index(line, closeComment) + offset
-		if index > -1 && hindex == index {
-			if index == 0 || len(line) < 2 || line[index-offset-1] != '-' {
-				node.Segments.Append(segment.WithStop(segment.Start + index + len(closeComment)))
-				block.Advance(index + len(closeComment))
+		index := bytes.Index(line, closeComment)
+		if index > -1 {
+			node.Segments.Append(segment.WithStop(segment.Start + offset + index + len(closeComment)))
+			block.Advance(offset + index + len(closeComment))
 			return node
 		}
-		}
-		if hindex > 0 {
-			break
-		}
+		offset = 0
 		node.Segments.Append(segment)
 		block.AdvanceLine()
 		line, segment = block.PeekLine()
-		offset = 0
 		if line == nil {
 			break
 		}
@@ -153,9 +144,8 @@ func (s *rawHTMLParser) parseMultiLineRegexp(reg *regexp.Regexp, block text.Read
 			if l == eline {
 				block.Advance(end - start)
 				break
-			} else {
-				block.AdvanceLine()
 			}
+			block.AdvanceLine()
 		}
 		return node
 	}

parser/setext_headings.go

@@ -91,7 +91,7 @@ func (b *setextHeadingParser) Close(node ast.Node, reader text.Reader, pc Contex
 			para.Lines().Append(segment)
 			heading.Parent().InsertAfter(heading.Parent(), heading, para)
 		} else {
-			next.(ast.Node).Lines().Unshift(segment)
+			next.Lines().Unshift(segment)
 		}
 		heading.Parent().RemoveChild(heading.Parent(), heading)
 	} else {

renderer/html/html.go

@@ -1,9 +1,11 @@
+// Package html implements renderer that outputs HTMLs.
 package html
 
 import (
 	"bytes"
 	"fmt"
 	"strconv"
+	"unicode"
 	"unicode/utf8"
 
 	"github.com/yuin/goldmark/ast"
@@ -15,7 +17,7 @@ import (
 type Config struct {
 	Writer              Writer
 	HardWraps           bool
-	EastAsianLineBreaks bool
+	EastAsianLineBreaks EastAsianLineBreaks
 	XHTML               bool
 	Unsafe              bool
 }
@@ -25,7 +27,7 @@ func NewConfig() Config {
 	return Config{
 		Writer:              DefaultWriter,
 		HardWraps:           false,
-		EastAsianLineBreaks: false,
+		EastAsianLineBreaks: EastAsianLineBreaksNone,
 		XHTML:               false,
 		Unsafe:              false,
 	}
@@ -37,7 +39,7 @@ func (c *Config) SetOption(name renderer.OptionName, value interface{}) {
 	case optHardWraps:
 		c.HardWraps = value.(bool)
 	case optEastAsianLineBreaks:
-		c.EastAsianLineBreaks = value.(bool)
+		c.EastAsianLineBreaks = value.(EastAsianLineBreaks)
 	case optXHTML:
 		c.XHTML = value.(bool)
 	case optUnsafe:
@@ -102,24 +104,94 @@ func WithHardWraps() interface {
 // EastAsianLineBreaks is an option name used in WithEastAsianLineBreaks.
 const optEastAsianLineBreaks renderer.OptionName = "EastAsianLineBreaks"
 
+// A EastAsianLineBreaks is a style of east asian line breaks.
+type EastAsianLineBreaks int
+
+const (
+	//EastAsianLineBreaksNone renders line breaks as it is.
+	EastAsianLineBreaksNone EastAsianLineBreaks = iota
+	// EastAsianLineBreaksSimple follows east_asian_line_breaks in Pandoc.
+	EastAsianLineBreaksSimple
+	// EastAsianLineBreaksCSS3Draft follows CSS text level3 "Segment Break Transformation Rules" with some enhancements.
+	EastAsianLineBreaksCSS3Draft
+)
+
+func (b EastAsianLineBreaks) softLineBreak(thisLastRune rune, siblingFirstRune rune) bool {
+	switch b {
+	case EastAsianLineBreaksNone:
+		return false
+	case EastAsianLineBreaksSimple:
+		return !(util.IsEastAsianWideRune(thisLastRune) && util.IsEastAsianWideRune(siblingFirstRune))
+	case EastAsianLineBreaksCSS3Draft:
+		return eastAsianLineBreaksCSS3DraftSoftLineBreak(thisLastRune, siblingFirstRune)
+	}
+	return false
+}
+
+func eastAsianLineBreaksCSS3DraftSoftLineBreak(thisLastRune rune, siblingFirstRune rune) bool {
+	// Implements CSS text level3 Segment Break Transformation Rules with some enhancements.
+	// References:
+	//   - https://www.w3.org/TR/2020/WD-css-text-3-20200429/#line-break-transform
+	//   - https://github.com/w3c/csswg-drafts/issues/5086
+
+	// Rule1:
+	//   If the character immediately before or immediately after the segment break is
+	//   the zero-width space character (U+200B), then the break is removed, leaving behind the zero-width space.
+	if thisLastRune == '\u200B' || siblingFirstRune == '\u200B' {
+		return false
+	}
+
+	// Rule2:
+	//   Otherwise, if the East Asian Width property of both the character before and after the segment break is
+	//   F, W, or H (not A), and neither side is Hangul, then the segment break is removed.
+	thisLastRuneEastAsianWidth := util.EastAsianWidth(thisLastRune)
+	siblingFirstRuneEastAsianWidth := util.EastAsianWidth(siblingFirstRune)
+	if (thisLastRuneEastAsianWidth == "F" ||
+		thisLastRuneEastAsianWidth == "W" ||
+		thisLastRuneEastAsianWidth == "H") &&
+		(siblingFirstRuneEastAsianWidth == "F" ||
+			siblingFirstRuneEastAsianWidth == "W" ||
+			siblingFirstRuneEastAsianWidth == "H") {
+		return unicode.Is(unicode.Hangul, thisLastRune) || unicode.Is(unicode.Hangul, siblingFirstRune)
+	}
+
+	// Rule3:
+	//   Otherwise, if either the character before or after the segment break belongs to
+	//   the space-discarding character set and it is a Unicode Punctuation (P*) or U+3000,
+	//   then the segment break is removed.
+	if util.IsSpaceDiscardingUnicodeRune(thisLastRune) ||
+		unicode.IsPunct(thisLastRune) ||
+		thisLastRune == '\u3000' ||
+		util.IsSpaceDiscardingUnicodeRune(siblingFirstRune) ||
+		unicode.IsPunct(siblingFirstRune) ||
+		siblingFirstRune == '\u3000' {
+		return false
+	}
+
+	// Rule4:
+	//   Otherwise, the segment break is converted to a space (U+0020).
+	return true
+}
+
 type withEastAsianLineBreaks struct {
+	eastAsianLineBreaksStyle EastAsianLineBreaks
 }
 
 func (o *withEastAsianLineBreaks) SetConfig(c *renderer.Config) {
-	c.Options[optEastAsianLineBreaks] = true
+	c.Options[optEastAsianLineBreaks] = o.eastAsianLineBreaksStyle
 }
 
 func (o *withEastAsianLineBreaks) SetHTMLOption(c *Config) {
-	c.EastAsianLineBreaks = true
+	c.EastAsianLineBreaks = o.eastAsianLineBreaksStyle
 }
 
 // WithEastAsianLineBreaks is a functional option that indicates whether softline breaks
 // between east asian wide characters should be ignored.
-func WithEastAsianLineBreaks() interface {
+func WithEastAsianLineBreaks(e EastAsianLineBreaks) interface {
 	renderer.Option
 	Option
 } {
-	return &withEastAsianLineBreaks{}
+	return &withEastAsianLineBreaks{e}
 }
 
 // XHTML is an option name used in WithXHTML.
@@ -253,15 +325,17 @@ var GlobalAttributeFilter = util.NewBytesFilter(
 	[]byte("translate"),
 )
 
-func (r *Renderer) renderDocument(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderDocument(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	// nothing to do
 	return ast.WalkContinue, nil
 }
 
-// HeadingAttributeFilter defines attribute names which heading elements can have
+// HeadingAttributeFilter defines attribute names which heading elements can have.
 var HeadingAttributeFilter = GlobalAttributeFilter
 
-func (r *Renderer) renderHeading(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderHeading(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	n := node.(*ast.Heading)
 	if entering {
 		_, _ = w.WriteString("<h")
@@ -278,12 +352,13 @@ func (r *Renderer) renderHeading(w util.BufWriter, source []byte, node ast.Node,
 	return ast.WalkContinue, nil
 }
 
-// BlockquoteAttributeFilter defines attribute names which blockquote elements can have
+// BlockquoteAttributeFilter defines attribute names which blockquote elements can have.
 var BlockquoteAttributeFilter = GlobalAttributeFilter.Extend(
 	[]byte("cite"),
 )
 
-func (r *Renderer) renderBlockquote(w util.BufWriter, source []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderBlockquote(
+	w util.BufWriter, source []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
 	if entering {
 		if n.Attributes() != nil {
 			_, _ = w.WriteString("<blockquote")
@@ -308,7 +383,8 @@ func (r *Renderer) renderCodeBlock(w util.BufWriter, source []byte, n ast.Node,
 	return ast.WalkContinue, nil
 }
 
-func (r *Renderer) renderFencedCodeBlock(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderFencedCodeBlock(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	n := node.(*ast.FencedCodeBlock)
 	if entering {
 		_, _ = w.WriteString("<pre><code")
@@ -326,7 +402,8 @@ func (r *Renderer) renderFencedCodeBlock(w util.BufWriter, source []byte, node a
 	return ast.WalkContinue, nil
 }
 
-func (r *Renderer) renderHTMLBlock(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderHTMLBlock(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	n := node.(*ast.HTMLBlock)
 	if entering {
 		if r.Unsafe {
@@ -368,7 +445,7 @@ func (r *Renderer) renderList(w util.BufWriter, source []byte, node ast.Node, en
 		_ = w.WriteByte('<')
 		_, _ = w.WriteString(tag)
 		if n.IsOrdered() && n.Start != 1 {
-			fmt.Fprintf(w, " start=\"%d\"", n.Start)
+			_, _ = fmt.Fprintf(w, " start=\"%d\"", n.Start)
 		}
 		if n.Attributes() != nil {
 			RenderAttributes(w, n, ListAttributeFilter)
@@ -428,7 +505,7 @@ func (r *Renderer) renderParagraph(w util.BufWriter, source []byte, n ast.Node,
 
 func (r *Renderer) renderTextBlock(w util.BufWriter, source []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
 	if !entering {
-		if _, ok := n.NextSibling().(ast.Node); ok && n.FirstChild() != nil {
+		if n.NextSibling() != nil && n.FirstChild() != nil {
 			_ = w.WriteByte('\n')
 		}
 	}
@@ -444,7 +521,8 @@ var ThematicAttributeFilter = GlobalAttributeFilter.Extend(
 	[]byte("width"),   // [Deprecated]
 )
 
-func (r *Renderer) renderThematicBreak(w util.BufWriter, source []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderThematicBreak(
+	w util.BufWriter, source []byte, n ast.Node, entering bool) (ast.WalkStatus, error) {
 	if !entering {
 		return ast.WalkContinue, nil
 	}
@@ -473,7 +551,8 @@ var LinkAttributeFilter = GlobalAttributeFilter.Extend(
 	[]byte("target"),
 )
 
-func (r *Renderer) renderAutoLink(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderAutoLink(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	n := node.(*ast.AutoLink)
 	if !entering {
 		return ast.WalkContinue, nil
@@ -528,7 +607,8 @@ func (r *Renderer) renderCodeSpan(w util.BufWriter, source []byte, n ast.Node, e
 // EmphasisAttributeFilter defines attribute names which emphasis elements can have.
 var EmphasisAttributeFilter = GlobalAttributeFilter
 
-func (r *Renderer) renderEmphasis(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderEmphasis(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	n := node.(*ast.Emphasis)
 	tag := "em"
 	if n.Level == 2 {
@@ -600,7 +680,7 @@ func (r *Renderer) renderImage(w util.BufWriter, source []byte, node ast.Node, e
 		_, _ = w.Write(util.EscapeHTML(util.URLEscape(n.Destination, true)))
 	}
 	_, _ = w.WriteString(`" alt="`)
-	_, _ = w.Write(nodeToHTMLText(n, source))
+	r.renderTexts(w, source, n)
 	_ = w.WriteByte('"')
 	if n.Title != nil {
 		_, _ = w.WriteString(` title="`)
@@ -618,7 +698,8 @@ func (r *Renderer) renderImage(w util.BufWriter, source []byte, node ast.Node, e
 	return ast.WalkSkipChildren, nil
 }
 
-func (r *Renderer) renderRawHTML(w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
+func (r *Renderer) renderRawHTML(
+	w util.BufWriter, source []byte, node ast.Node, entering bool) (ast.WalkStatus, error) {
 	if !entering {
 		return ast.WalkSkipChildren, nil
 	}
@@ -653,14 +734,13 @@ func (r *Renderer) renderText(w util.BufWriter, source []byte, node ast.Node, en
 				_, _ = w.WriteString("<br>\n")
 			}
 		} else if n.SoftLineBreak() {
-			if r.EastAsianLineBreaks && len(value) != 0 {
+			if r.EastAsianLineBreaks != EastAsianLineBreaksNone && len(value) != 0 {
 				sibling := node.NextSibling()
 				if sibling != nil && sibling.Kind() == ast.KindText {
-					if siblingText := sibling.(*ast.Text).Text(source); len(siblingText) != 0 {
+					if siblingText := sibling.(*ast.Text).Value(source); len(siblingText) != 0 {
 						thisLastRune := util.ToRune(value, len(value)-1)
 						siblingFirstRune, _ := utf8.DecodeRune(siblingText)
-						if !(util.IsEastAsianWideRune(thisLastRune) &&
-							util.IsEastAsianWideRune(siblingFirstRune)) {
+						if r.EastAsianLineBreaks.softLineBreak(thisLastRune, siblingFirstRune) {
 							_ = w.WriteByte('\n')
 						}
 					}
@@ -690,6 +770,18 @@ func (r *Renderer) renderString(w util.BufWriter, source []byte, node ast.Node,
 	return ast.WalkContinue, nil
 }
 
+func (r *Renderer) renderTexts(w util.BufWriter, source []byte, n ast.Node) {
+	for c := n.FirstChild(); c != nil; c = c.NextSibling() {
+		if s, ok := c.(*ast.String); ok {
+			_, _ = r.renderString(w, source, s, true)
+		} else if t, ok := c.(*ast.Text); ok {
+			_, _ = r.renderText(w, source, t, true)
+		} else {
+			r.renderTexts(w, source, c)
+		}
+	}
+}
+
 var dataPrefix = []byte("data-")
 
 // RenderAttributes renders given node's attributes.
@@ -706,7 +798,14 @@ func RenderAttributes(w util.BufWriter, node ast.Node, filter util.BytesFilter)
 		_, _ = w.Write(attr.Name)
 		_, _ = w.WriteString(`="`)
 		// TODO: convert numeric values to strings
-		_, _ = w.Write(util.EscapeHTML(attr.Value.([]byte)))
+		var value []byte
+		switch typed := attr.Value.(type) {
+		case []byte:
+			value = typed
+		case string:
+			value = util.StringToReadOnlyBytes(typed)
+		}
+		_, _ = w.Write(util.EscapeHTML(value))
 		_ = w.WriteByte('"')
 	}
 }
@@ -920,17 +1019,3 @@ func IsDangerousURL(url []byte) bool {
 	return hasPrefix(url, bJs) || hasPrefix(url, bVb) ||
 		hasPrefix(url, bFile) || hasPrefix(url, bData)
 }
-
-func nodeToHTMLText(n ast.Node, source []byte) []byte {
-	var buf bytes.Buffer
-	for c := n.FirstChild(); c != nil; c = c.NextSibling() {
-		if s, ok := c.(*ast.String); ok && s.IsCode() {
-			buf.Write(s.Text(source))
-		} else if !c.HasChildren() {
-			buf.Write(util.EscapeHTML(c.Text(source)))
-		} else {
-			buf.Write(nodeToHTMLText(c, source))
-		}
-	}
-	return buf.Bytes()
-}

renderer/renderer.go

@@ -16,7 +16,7 @@ type Config struct {
 	NodeRenderers util.PrioritizedSlice
 }
 
-// NewConfig returns a new Config
+// NewConfig returns a new Config.
 func NewConfig() *Config {
 	return &Config{
 		Options:       map[OptionName]interface{}{},
@@ -78,7 +78,7 @@ type NodeRenderer interface {
 	RegisterFuncs(NodeRendererFuncRegisterer)
 }
 
-// A NodeRendererFuncRegisterer registers
+// A NodeRendererFuncRegisterer registers given NodeRendererFunc to this object.
 type NodeRendererFuncRegisterer interface {
 	// Register registers given NodeRendererFunc to this object.
 	Register(ast.NodeKind, NodeRendererFunc)

testutil/testutil.go

@@ -1,3 +1,4 @@
+// Package testutil provides utilities for unit tests.
 package testutil
 
 import (
@@ -65,7 +66,7 @@ type MarkdownTestCaseOptions struct {
 const attributeSeparator = "//- - - - - - - - -//"
 const caseSeparator = "//= = = = = = = = = = = = = = = = = = = = = = = =//"
 
-var optionsRegexp *regexp.Regexp = regexp.MustCompile(`(?i)\s*options:(.*)`)
+var optionsRegexp = regexp.MustCompile(`(?i)\s*options:(.*)`)
 
 // ParseCliCaseArg parses -case command line args.
 func ParseCliCaseArg() []int {
@@ -90,7 +91,9 @@ func DoTestCaseFile(m goldmark.Markdown, filename string, t TestingT, no ...int)
 	if err != nil {
 		panic(err)
 	}
-	defer fp.Close()
+	defer func() {
+		_ = fp.Close()
+	}()
 
 	scanner := bufio.NewScanner(fp)
 	c := MarkdownTestCase{

text/package.go

@@ -0,0 +1,2 @@
+// Package text provides functionalities to manipulate texts.
+package text

text/reader.go

@@ -76,7 +76,7 @@ type Reader interface {
 	FindClosure(opener, closer byte, options FindClosureOptions) (*Segments, bool)
 }
 
-// FindClosureOptions is options for Reader.FindClosure
+// FindClosureOptions is options for Reader.FindClosure.
 type FindClosureOptions struct {
 	// CodeSpan is a flag for the FindClosure. If this is set to true,
 	// FindClosure ignores closers in codespans.
@@ -154,7 +154,7 @@ func (r *reader) PeekLine() ([]byte, Segment) {
 	return nil, r.pos
 }
 
-// io.RuneReader interface
+// io.RuneReader interface.
 func (r *reader) ReadRune() (rune, int, error) {
 	return readRuneReader(r)
 }
@@ -354,7 +354,7 @@ func (r *blockReader) Value(seg Segment) []byte {
 	return ret
 }
 
-// io.RuneReader interface
+// io.RuneReader interface.
 func (r *blockReader) ReadRune() (rune, int, error) {
 	return readRuneReader(r)
 }

text/segment.go

@@ -2,6 +2,7 @@ package text
 
 import (
 	"bytes"
+
 	"github.com/yuin/goldmark/util"
 )
 
@@ -18,6 +19,20 @@ type Segment struct {
 
 	// Padding is a padding length of the segment.
 	Padding int
+
+	// ForceNewline is true if the segment should be ended with a newline.
+	// Some elements(i.e. CodeBlock, FencedCodeBlock) does not trim trailing
+	// newlines. Spec defines that EOF is treated as a newline, so we need to
+	// add a newline to the end of the segment if it is not empty.
+	//
+	// i.e.:
+	//
+	//     ```go
+	//     const test = "test"
+	//
+	// This code does not close the code block and ends with EOF. In this case,
+	// we need to add a newline to the end of the last line like `const test = "test"\n`.
+	ForceNewline bool
 }
 
 // NewSegment return a new Segment.
@@ -40,12 +55,18 @@ func NewSegmentPadding(start, stop, n int) Segment {
 
 // Value returns a value of the segment.
 func (t *Segment) Value(buffer []byte) []byte {
+	var result []byte
 	if t.Padding == 0 {
-		return buffer[t.Start:t.Stop]
-	}
-	result := make([]byte, 0, t.Padding+t.Stop-t.Start+1)
+		result = buffer[t.Start:t.Stop]
+	} else {
+		result = make([]byte, 0, t.Padding+t.Stop-t.Start+1)
 		result = append(result, bytes.Repeat(space, t.Padding)...)
-	return append(result, buffer[t.Start:t.Stop]...)
+		result = append(result, buffer[t.Start:t.Stop]...)
+	}
+	if t.ForceNewline && len(result) > 0 && result[len(result)-1] != '\n' {
+		result = append(result, '\n')
+	}
+	return result
 }
 
 // Len returns a length of the segment.
@@ -207,3 +228,12 @@ func (s *Segments) Unshift(v Segment) {
 	s.values = append(s.values[0:1], s.values[0:]...)
 	s.values[0] = v
 }
+
+// Value returns a string value of the collection.
+func (s *Segments) Value(buffer []byte) []byte {
+	var result []byte
+	for _, v := range s.values {
+		result = append(result, v.Value(buffer)...)
+	}
+	return result
+}

util/html5entities.go

@@ -1,5 +1,8 @@
+//nolint:golint,lll,misspell
 package util
 
+import "sync"
+
 // An HTML5Entity struct represents HTML5 entitites.
 type HTML5Entity struct {
 	Name       string
@@ -8,13 +11,20 @@ type HTML5Entity struct {
 }
 
 // LookUpHTML5EntityByName returns (an HTML5Entity, true) if an entity named
-// given name is found, otherwise (nil, false)
+// given name is found, otherwise (nil, false).
 func LookUpHTML5EntityByName(name string) (*HTML5Entity, bool) {
-	v, ok := html5entities[name]
+	v, ok := html5entities()[name]
 	return v, ok
 }
 
-var html5entities = map[string]*HTML5Entity{
+var html5entitiesOnce sync.Once // TODO: uses sync.OnceValue for future
+
+var _html5entities map[string]*HTML5Entity
+
+func html5entities() map[string]*HTML5Entity {
+	html5entitiesOnce.Do(func() {
+		_html5entities =
+			map[string]*HTML5Entity{
 				"AElig":                           {Name: "AElig", CodePoints: []int{198}, Characters: []byte{0xc3, 0x86}},
 				"AMP":                             {Name: "AMP", CodePoints: []int{38}, Characters: []byte{0x26}},
 				"Aacute":                          {Name: "Aacute", CodePoints: []int{193}, Characters: []byte{0xc3, 0x81}},
@@ -2139,4 +2149,7 @@ var html5entities = map[string]*HTML5Entity{
 				"zscr":                            {Name: "zscr", CodePoints: []int{120015}, Characters: []byte{0xf0, 0x9d, 0x93, 0x8f}},
 				"zwj":                             {Name: "zwj", CodePoints: []int{8205}, Characters: []byte{0xe2, 0x80, 0x8d}},
 				"zwnj":                            {Name: "zwnj", CodePoints: []int{8204}, Characters: []byte{0xe2, 0x80, 0x8c}},
+			}
+	})
+	return _html5entities
 }

util/util.go

@@ -63,12 +63,13 @@ func (b *CopyOnWriteBuffer) AppendString(value string) {
 
 // WriteByte writes the given byte to the buffer.
 // WriteByte allocate new buffer and clears it at the first time.
-func (b *CopyOnWriteBuffer) WriteByte(c byte) {
+func (b *CopyOnWriteBuffer) WriteByte(c byte) error {
 	if !b.copied {
 		b.buffer = make([]byte, 0, len(b.buffer)+20)
 		b.copied = true
 	}
 	b.buffer = append(b.buffer, c)
+	return nil
 }
 
 // AppendByte appends given bytes to the buffer.
@@ -150,7 +151,7 @@ func TabWidth(currentPos int) int {
 //	width:    1234 5678
 //
 // width=2 is in the tab character. In this case, IndentPosition returns
-// (pos=1, padding=2)
+// (pos=1, padding=2).
 func IndentPosition(bs []byte, currentPos, width int) (pos, padding int) {
 	return IndentPositionPadding(bs, currentPos, 0, width)
 }
@@ -165,7 +166,13 @@ func IndentPositionPadding(bs []byte, currentPos, paddingv, width int) (pos, pad
 	w := 0
 	i := 0
 	l := len(bs)
+	p := paddingv
 	for ; i < l; i++ {
+		if p > 0 {
+			p--
+			w++
+			continue
+		}
 		if bs[i] == '\t' && w < width {
 			w += TabWidth(currentPos + w)
 		} else if bs[i] == ' ' && w < width {
@@ -424,7 +431,7 @@ func DoFullUnicodeCaseFolding(v []byte) []byte {
 			if c >= 0x41 && c <= 0x5a {
 				// A-Z to a-z
 				cob.Write(v[n:i])
-				cob.WriteByte(c + 32)
+				_ = cob.WriteByte(c + 32)
 				n = i + 1
 			}
 			continue
@@ -521,7 +528,7 @@ func ToLinkReference(v []byte) string {
 	return string(ReplaceSpaces(v, ' '))
 }
 
-var htmlEscapeTable = [256][]byte{nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, []byte("&quot;"), nil, nil, nil, []byte("&amp;"), nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, []byte("&lt;"), nil, []byte("&gt;"), nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil}
+var htmlEscapeTable = [256][]byte{nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, []byte("&quot;"), nil, nil, nil, []byte("&amp;"), nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, []byte("&lt;"), nil, []byte("&gt;"), nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil} //nolint:golint,lll
 
 // EscapeHTMLByte returns HTML escaped bytes if the given byte should be escaped,
 // otherwise nil.
@@ -557,7 +564,7 @@ func UnescapePunctuations(source []byte) []byte {
 		c := source[i]
 		if i < limit-1 && c == '\\' && IsPunct(source[i+1]) {
 			cob.Write(source[n:i])
-			cob.WriteByte(source[i+1])
+			_ = cob.WriteByte(source[i+1])
 			i += 2
 			n = i
 			continue
@@ -573,9 +580,9 @@ func UnescapePunctuations(source []byte) []byte {
 // ResolveNumericReferences resolve numeric references like '&#1234;" .
 func ResolveNumericReferences(source []byte) []byte {
 	cob := NewCopyOnWriteBuffer(source)
-	buf := make([]byte, 6, 6)
+	buf := make([]byte, 6)
 	limit := len(source)
-	ok := false
+	var ok bool
 	n := 0
 	for i := 0; i < limit; i++ {
 		if source[i] == '&' {
@@ -625,7 +632,7 @@ func ResolveNumericReferences(source []byte) []byte {
 func ResolveEntityNames(source []byte) []byte {
 	cob := NewCopyOnWriteBuffer(source)
 	limit := len(source)
-	ok := false
+	var ok bool
 	n := 0
 	for i := 0; i < limit; i++ {
 		if source[i] == '&' {
@@ -750,7 +757,7 @@ func FindURLIndex(b []byte) int {
 	return i
 }
 
-var emailDomainRegexp = regexp.MustCompile(`^[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*`)
+var emailDomainRegexp = regexp.MustCompile(`^[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*`) //nolint:golint,lll
 
 // FindEmailIndex returns a stop index value if the given bytes seem an email address.
 func FindEmailIndex(b []byte) int {
@@ -781,18 +788,19 @@ func FindEmailIndex(b []byte) int {
 
 var spaces = []byte(" \t\n\x0b\x0c\x0d")
 
-var spaceTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}
+var spaceTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} //nolint:golint,lll
 
-var punctTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}
+var punctTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} //nolint:golint,lll
 
 // a-zA-Z0-9, ;/?:@&=+$,-_.!~*'()#
-var urlEscapeTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}
 
-var utf8lenTable = [256]int8{1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 99, 99, 99, 99, 99, 99, 99, 99}
+var urlEscapeTable = [256]int8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} //nolint:golint,lll
 
-var urlTable = [256]uint8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 5, 1, 5, 5, 1, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 1, 1, 0, 1, 0, 1, 1, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 1, 1, 1, 1, 1, 1, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1}
+var utf8lenTable = [256]int8{1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 99, 99, 99, 99, 99, 99, 99, 99} //nolint:golint,lll
 
-var emailTable = [256]uint8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}
+var urlTable = [256]uint8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 5, 1, 5, 5, 1, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 1, 1, 0, 1, 0, 1, 1, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 1, 1, 1, 1, 1, 1, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1} //nolint:golint,lll
+
+var emailTable = [256]uint8{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} //nolint:golint,lll
 
 // UTF8Len returns a byte length of the utf-8 character.
 func UTF8Len(b byte) int8 {
@@ -806,7 +814,7 @@ func IsPunct(c byte) bool {
 
 // IsPunctRune returns true if the given rune is a punctuation, otherwise false.
 func IsPunctRune(r rune) bool {
-	return int32(r) <= 256 && IsPunct(byte(r)) || unicode.IsPunct(r)
+	return unicode.IsSymbol(r) || unicode.IsPunct(r)
 }
 
 // IsSpace returns true if the given character is a space, otherwise false.
@@ -834,15 +842,6 @@ func IsAlphaNumeric(c byte) bool {
 	return c >= 'a' && c <= 'z' || c >= 'A' && c <= 'Z' || c >= '0' && c <= '9'
 }
 
-// IsEastAsianWideRune returns trhe if the given rune is an east asian wide character, otherwise false.
-func IsEastAsianWideRune(r rune) bool {
-	return unicode.Is(unicode.Hiragana, r) ||
-		unicode.Is(unicode.Katakana, r) ||
-		unicode.Is(unicode.Han, r) ||
-		unicode.Is(unicode.Lm, r) ||
-		unicode.Is(unicode.Hangul, r)
-}
-
 // A BufWriter is a subset of the bufio.Writer .
 type BufWriter interface {
 	io.Writer
@@ -862,7 +861,7 @@ type PrioritizedValue struct {
 	Priority int
 }
 
-// PrioritizedSlice is a slice of the PrioritizedValues
+// PrioritizedSlice is a slice of the PrioritizedValues.
 type PrioritizedSlice []PrioritizedValue
 
 // Sort sorts the PrioritizedSlice in ascending order.
@@ -977,7 +976,7 @@ func (s *bytesFilter) Contains(b []byte) bool {
 	}
 	h := bytesHash(b) % uint64(len(s.slots))
 	slot := s.slots[h]
-	if slot == nil || len(slot) == 0 {
+	if len(slot) == 0 {
 		return false
 	}
 	for _, element := range slot {

util/util_cjk.go

@@ -0,0 +1,469 @@
+package util
+
+import "unicode"
+
+var cjkRadicalsSupplement = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x2E80, 0x2EFF, 1},
+	},
+}
+
+var kangxiRadicals = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x2F00, 0x2FDF, 1},
+	},
+}
+
+var ideographicDescriptionCharacters = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x2FF0, 0x2FFF, 1},
+	},
+}
+
+var cjkSymbolsAndPunctuation = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x3000, 0x303F, 1},
+	},
+}
+
+var hiragana = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x3040, 0x309F, 1},
+	},
+}
+
+var katakana = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x30A0, 0x30FF, 1},
+	},
+}
+
+var kanbun = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x3130, 0x318F, 1},
+		{0x3190, 0x319F, 1},
+	},
+}
+
+var cjkStrokes = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x31C0, 0x31EF, 1},
+	},
+}
+
+var katakanaPhoneticExtensions = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x31F0, 0x31FF, 1},
+	},
+}
+
+var cjkCompatibility = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x3300, 0x33FF, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionA = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x3400, 0x4DBF, 1},
+	},
+}
+
+var cjkUnifiedIdeographs = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0x4E00, 0x9FFF, 1},
+	},
+}
+
+var yiSyllables = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xA000, 0xA48F, 1},
+	},
+}
+
+var yiRadicals = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xA490, 0xA4CF, 1},
+	},
+}
+
+var cjkCompatibilityIdeographs = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xF900, 0xFAFF, 1},
+	},
+}
+
+var verticalForms = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xFE10, 0xFE1F, 1},
+	},
+}
+
+var cjkCompatibilityForms = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xFE30, 0xFE4F, 1},
+	},
+}
+
+var smallFormVariants = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xFE50, 0xFE6F, 1},
+	},
+}
+
+var halfwidthAndFullwidthForms = &unicode.RangeTable{
+	R16: []unicode.Range16{
+		{0xFF00, 0xFFEF, 1},
+	},
+}
+
+var kanaSupplement = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x1B000, 0x1B0FF, 1},
+	},
+}
+
+var kanaExtendedA = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x1B100, 0x1B12F, 1},
+	},
+}
+
+var smallKanaExtension = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x1B130, 0x1B16F, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionB = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x20000, 0x2A6DF, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionC = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x2A700, 0x2B73F, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionD = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x2B740, 0x2B81F, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionE = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x2B820, 0x2CEAF, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionF = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x2CEB0, 0x2EBEF, 1},
+	},
+}
+
+var cjkCompatibilityIdeographsSupplement = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x2F800, 0x2FA1F, 1},
+	},
+}
+
+var cjkUnifiedIdeographsExtensionG = &unicode.RangeTable{
+	R32: []unicode.Range32{
+		{0x30000, 0x3134F, 1},
+	},
+}
+
+// IsEastAsianWideRune returns trhe if the given rune is an east asian wide character, otherwise false.
+func IsEastAsianWideRune(r rune) bool {
+	return unicode.Is(unicode.Hiragana, r) ||
+		unicode.Is(unicode.Katakana, r) ||
+		unicode.Is(unicode.Han, r) ||
+		unicode.Is(unicode.Lm, r) ||
+		unicode.Is(unicode.Hangul, r) ||
+		unicode.Is(cjkSymbolsAndPunctuation, r)
+}
+
+// IsSpaceDiscardingUnicodeRune returns true if the given rune is space-discarding unicode character, otherwise false.
+// See https://www.w3.org/TR/2020/WD-css-text-3-20200429/#space-discard-set
+func IsSpaceDiscardingUnicodeRune(r rune) bool {
+	return unicode.Is(cjkRadicalsSupplement, r) ||
+		unicode.Is(kangxiRadicals, r) ||
+		unicode.Is(ideographicDescriptionCharacters, r) ||
+		unicode.Is(cjkSymbolsAndPunctuation, r) ||
+		unicode.Is(hiragana, r) ||
+		unicode.Is(katakana, r) ||
+		unicode.Is(kanbun, r) ||
+		unicode.Is(cjkStrokes, r) ||
+		unicode.Is(katakanaPhoneticExtensions, r) ||
+		unicode.Is(cjkCompatibility, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionA, r) ||
+		unicode.Is(cjkUnifiedIdeographs, r) ||
+		unicode.Is(yiSyllables, r) ||
+		unicode.Is(yiRadicals, r) ||
+		unicode.Is(cjkCompatibilityIdeographs, r) ||
+		unicode.Is(verticalForms, r) ||
+		unicode.Is(cjkCompatibilityForms, r) ||
+		unicode.Is(smallFormVariants, r) ||
+		unicode.Is(halfwidthAndFullwidthForms, r) ||
+		unicode.Is(kanaSupplement, r) ||
+		unicode.Is(kanaExtendedA, r) ||
+		unicode.Is(smallKanaExtension, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionB, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionC, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionD, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionE, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionF, r) ||
+		unicode.Is(cjkCompatibilityIdeographsSupplement, r) ||
+		unicode.Is(cjkUnifiedIdeographsExtensionG, r)
+}
+
+// EastAsianWidth returns the east asian width of the given rune.
+// See https://www.unicode.org/reports/tr11/tr11-36.html
+func EastAsianWidth(r rune) string {
+	switch {
+	case r == 0x3000,
+		(0xFF01 <= r && r <= 0xFF60),
+		(0xFFE0 <= r && r <= 0xFFE6):
+		return "F"
+
+	case r == 0x20A9,
+		(0xFF61 <= r && r <= 0xFFBE),
+		(0xFFC2 <= r && r <= 0xFFC7),
+		(0xFFCA <= r && r <= 0xFFCF),
+		(0xFFD2 <= r && r <= 0xFFD7),
+		(0xFFDA <= r && r <= 0xFFDC),
+		(0xFFE8 <= r && r <= 0xFFEE):
+		return "H"
+
+	case (0x1100 <= r && r <= 0x115F),
+		(0x11A3 <= r && r <= 0x11A7),
+		(0x11FA <= r && r <= 0x11FF),
+		(0x2329 <= r && r <= 0x232A),
+		(0x2E80 <= r && r <= 0x2E99),
+		(0x2E9B <= r && r <= 0x2EF3),
+		(0x2F00 <= r && r <= 0x2FD5),
+		(0x2FF0 <= r && r <= 0x2FFB),
+		(0x3001 <= r && r <= 0x303E),
+		(0x3041 <= r && r <= 0x3096),
+		(0x3099 <= r && r <= 0x30FF),
+		(0x3105 <= r && r <= 0x312D),
+		(0x3131 <= r && r <= 0x318E),
+		(0x3190 <= r && r <= 0x31BA),
+		(0x31C0 <= r && r <= 0x31E3),
+		(0x31F0 <= r && r <= 0x321E),
+		(0x3220 <= r && r <= 0x3247),
+		(0x3250 <= r && r <= 0x32FE),
+		(0x3300 <= r && r <= 0x4DBF),
+		(0x4E00 <= r && r <= 0xA48C),
+		(0xA490 <= r && r <= 0xA4C6),
+		(0xA960 <= r && r <= 0xA97C),
+		(0xAC00 <= r && r <= 0xD7A3),
+		(0xD7B0 <= r && r <= 0xD7C6),
+		(0xD7CB <= r && r <= 0xD7FB),
+		(0xF900 <= r && r <= 0xFAFF),
+		(0xFE10 <= r && r <= 0xFE19),
+		(0xFE30 <= r && r <= 0xFE52),
+		(0xFE54 <= r && r <= 0xFE66),
+		(0xFE68 <= r && r <= 0xFE6B),
+		(0x1B000 <= r && r <= 0x1B001),
+		(0x1F200 <= r && r <= 0x1F202),
+		(0x1F210 <= r && r <= 0x1F23A),
+		(0x1F240 <= r && r <= 0x1F248),
+		(0x1F250 <= r && r <= 0x1F251),
+		(0x20000 <= r && r <= 0x2F73F),
+		(0x2B740 <= r && r <= 0x2FFFD),
+		(0x30000 <= r && r <= 0x3FFFD):
+		return "W"
+
+	case (0x0020 <= r && r <= 0x007E),
+		(0x00A2 <= r && r <= 0x00A3),
+		(0x00A5 <= r && r <= 0x00A6),
+		r == 0x00AC,
+		r == 0x00AF,
+		(0x27E6 <= r && r <= 0x27ED),
+		(0x2985 <= r && r <= 0x2986):
+		return "Na"
+
+	case (0x00A1 == r),
+		(0x00A4 == r),
+		(0x00A7 <= r && r <= 0x00A8),
+		(0x00AA == r),
+		(0x00AD <= r && r <= 0x00AE),
+		(0x00B0 <= r && r <= 0x00B4),
+		(0x00B6 <= r && r <= 0x00BA),
+		(0x00BC <= r && r <= 0x00BF),
+		(0x00C6 == r),
+		(0x00D0 == r),
+		(0x00D7 <= r && r <= 0x00D8),
+		(0x00DE <= r && r <= 0x00E1),
+		(0x00E6 == r),
+		(0x00E8 <= r && r <= 0x00EA),
+		(0x00EC <= r && r <= 0x00ED),
+		(0x00F0 == r),
+		(0x00F2 <= r && r <= 0x00F3),
+		(0x00F7 <= r && r <= 0x00FA),
+		(0x00FC == r),
+		(0x00FE == r),
+		(0x0101 == r),
+		(0x0111 == r),
+		(0x0113 == r),
+		(0x011B == r),
+		(0x0126 <= r && r <= 0x0127),
+		(0x012B == r),
+		(0x0131 <= r && r <= 0x0133),
+		(0x0138 == r),
+		(0x013F <= r && r <= 0x0142),
+		(0x0144 == r),
+		(0x0148 <= r && r <= 0x014B),
+		(0x014D == r),
+		(0x0152 <= r && r <= 0x0153),
+		(0x0166 <= r && r <= 0x0167),
+		(0x016B == r),
+		(0x01CE == r),
+		(0x01D0 == r),
+		(0x01D2 == r),
+		(0x01D4 == r),
+		(0x01D6 == r),
+		(0x01D8 == r),
+		(0x01DA == r),
+		(0x01DC == r),
+		(0x0251 == r),
+		(0x0261 == r),
+		(0x02C4 == r),
+		(0x02C7 == r),
+		(0x02C9 <= r && r <= 0x02CB),
+		(0x02CD == r),
+		(0x02D0 == r),
+		(0x02D8 <= r && r <= 0x02DB),
+		(0x02DD == r),
+		(0x02DF == r),
+		(0x0300 <= r && r <= 0x036F),
+		(0x0391 <= r && r <= 0x03A1),
+		(0x03A3 <= r && r <= 0x03A9),
+		(0x03B1 <= r && r <= 0x03C1),
+		(0x03C3 <= r && r <= 0x03C9),
+		(0x0401 == r),
+		(0x0410 <= r && r <= 0x044F),
+		(0x0451 == r),
+		(0x2010 == r),
+		(0x2013 <= r && r <= 0x2016),
+		(0x2018 <= r && r <= 0x2019),
+		(0x201C <= r && r <= 0x201D),
+		(0x2020 <= r && r <= 0x2022),
+		(0x2024 <= r && r <= 0x2027),
+		(0x2030 == r),
+		(0x2032 <= r && r <= 0x2033),
+		(0x2035 == r),
+		(0x203B == r),
+		(0x203E == r),
+		(0x2074 == r),
+		(0x207F == r),
+		(0x2081 <= r && r <= 0x2084),
+		(0x20AC == r),
+		(0x2103 == r),
+		(0x2105 == r),
+		(0x2109 == r),
+		(0x2113 == r),
+		(0x2116 == r),
+		(0x2121 <= r && r <= 0x2122),
+		(0x2126 == r),
+		(0x212B == r),
+		(0x2153 <= r && r <= 0x2154),
+		(0x215B <= r && r <= 0x215E),
+		(0x2160 <= r && r <= 0x216B),
+		(0x2170 <= r && r <= 0x2179),
+		(0x2189 == r),
+		(0x2190 <= r && r <= 0x2199),
+		(0x21B8 <= r && r <= 0x21B9),
+		(0x21D2 == r),
+		(0x21D4 == r),
+		(0x21E7 == r),
+		(0x2200 == r),
+		(0x2202 <= r && r <= 0x2203),
+		(0x2207 <= r && r <= 0x2208),
+		(0x220B == r),
+		(0x220F == r),
+		(0x2211 == r),
+		(0x2215 == r),
+		(0x221A == r),
+		(0x221D <= r && r <= 0x2220),
+		(0x2223 == r),
+		(0x2225 == r),
+		(0x2227 <= r && r <= 0x222C),
+		(0x222E == r),
+		(0x2234 <= r && r <= 0x2237),
+		(0x223C <= r && r <= 0x223D),
+		(0x2248 == r),
+		(0x224C == r),
+		(0x2252 == r),
+		(0x2260 <= r && r <= 0x2261),
+		(0x2264 <= r && r <= 0x2267),
+		(0x226A <= r && r <= 0x226B),
+		(0x226E <= r && r <= 0x226F),
+		(0x2282 <= r && r <= 0x2283),
+		(0x2286 <= r && r <= 0x2287),
+		(0x2295 == r),
+		(0x2299 == r),
+		(0x22A5 == r),
+		(0x22BF == r),
+		(0x2312 == r),
+		(0x2460 <= r && r <= 0x24E9),
+		(0x24EB <= r && r <= 0x254B),
+		(0x2550 <= r && r <= 0x2573),
+		(0x2580 <= r && r <= 0x258F),
+		(0x2592 <= r && r <= 0x2595),
+		(0x25A0 <= r && r <= 0x25A1),
+		(0x25A3 <= r && r <= 0x25A9),
+		(0x25B2 <= r && r <= 0x25B3),
+		(0x25B6 <= r && r <= 0x25B7),
+		(0x25BC <= r && r <= 0x25BD),
+		(0x25C0 <= r && r <= 0x25C1),
+		(0x25C6 <= r && r <= 0x25C8),
+		(0x25CB == r),
+		(0x25CE <= r && r <= 0x25D1),
+		(0x25E2 <= r && r <= 0x25E5),
+		(0x25EF == r),
+		(0x2605 <= r && r <= 0x2606),
+		(0x2609 == r),
+		(0x260E <= r && r <= 0x260F),
+		(0x2614 <= r && r <= 0x2615),
+		(0x261C == r),
+		(0x261E == r),
+		(0x2640 == r),
+		(0x2642 == r),
+		(0x2660 <= r && r <= 0x2661),
+		(0x2663 <= r && r <= 0x2665),
+		(0x2667 <= r && r <= 0x266A),
+		(0x266C <= r && r <= 0x266D),
+		(0x266F == r),
+		(0x269E <= r && r <= 0x269F),
+		(0x26BE <= r && r <= 0x26BF),
+		(0x26C4 <= r && r <= 0x26CD),
+		(0x26CF <= r && r <= 0x26E1),
+		(0x26E3 == r),
+		(0x26E8 <= r && r <= 0x26FF),
+		(0x273D == r),
+		(0x2757 == r),
+		(0x2776 <= r && r <= 0x277F),
+		(0x2B55 <= r && r <= 0x2B59),
+		(0x3248 <= r && r <= 0x324F),
+		(0xE000 <= r && r <= 0xF8FF),
+		(0xFE00 <= r && r <= 0xFE0F),
+		(0xFFFD == r),
+		(0x1F100 <= r && r <= 0x1F10A),
+		(0x1F110 <= r && r <= 0x1F12D),
+		(0x1F130 <= r && r <= 0x1F169),
+		(0x1F170 <= r && r <= 0x1F19A),
+		(0xE0100 <= r && r <= 0xE01EF),
+		(0xF0000 <= r && r <= 0xFFFFD),
+		(0x100000 <= r && r <= 0x10FFFD):
+		return "A"
+
+	default:
+		return "N"
+	}
+}

util/util_safe.go

@@ -1,3 +1,4 @@
+//go:build appengine || js
 // +build appengine js
 
 package util

util/util_unsafe_go120.go

@@ -1,4 +1,5 @@
-// +build !appengine,!js
+//go:build !appengine && !js && !go1.21
+// +build !appengine,!js,!go1.21
 
 package util
 

util/util_unsafe_go121.go

@@ -0,0 +1,18 @@
+//go:build !appengine && !js && go1.21
+// +build !appengine,!js,go1.21
+
+package util
+
+import (
+	"unsafe"
+)
+
+// BytesToReadOnlyString returns a string converted from given bytes.
+func BytesToReadOnlyString(b []byte) string {
+	return unsafe.String(unsafe.SliceData(b), len(b))
+}
+
+// StringToReadOnlyBytes returns bytes converted from given string.
+func StringToReadOnlyBytes(s string) []byte {
+	return unsafe.Slice(unsafe.StringData(s), len(s))
+}