Implements CSS3Draft

README.md

@@ -387,7 +387,7 @@ This extension provides additional options for CJK users.
 | 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` | Soft line breaks are ignored even if only one side of the break is east asian wide character. |
+| `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`
 
@@ -401,7 +401,7 @@ GoでWebアプリケーションを開発しています。
 
 Output:
 
-```md
+```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>
 ```
 
@@ -417,7 +417,7 @@ GoでWebアプリケーションを開発しています。
 
 Output:
 
-```md
+```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>
 ```
 

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(),
@@ -208,8 +209,19 @@ func TestEastAsianLineBreaks(t *testing.T) {
 		},
 		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,
+	)
 
-	// test with EastAsianLineBreaksCSS3Draft
+	// Tests with EastAsianLineBreaksCSS3Draft
 	markdown = goldmark.New(goldmark.WithRendererOptions(
 		html.WithXHTML(),
 		html.WithUnsafe(),
@@ -218,7 +230,7 @@ func TestEastAsianLineBreaks(t *testing.T) {
 			NewCJK(WithEastAsianLineBreaks(EastAsianLineBreaksCSS3Draft)),
 		),
 	)
-	no = 9
+	no = 10
 	testutil.DoTestCase(
 		markdown,
 		testutil.MarkdownTestCase{
@@ -230,7 +242,7 @@ func TestEastAsianLineBreaks(t *testing.T) {
 		t,
 	)
 
-	no = 10
+	no = 11
 	testutil.DoTestCase(
 		markdown,
 		testutil.MarkdownTestCase{
@@ -241,4 +253,17 @@ func TestEastAsianLineBreaks(t *testing.T) {
 		},
 		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,
+	)
+
 }

renderer/html/html.go

@@ -5,6 +5,7 @@ import (
 	"bytes"
 	"fmt"
 	"strconv"
+	"unicode"
 	"unicode/utf8"
 
 	"github.com/yuin/goldmark/ast"
@@ -107,15 +108,13 @@ const optEastAsianLineBreaks renderer.OptionName = "EastAsianLineBreaks"
 type EastAsianLineBreaksStyle int
 
 const (
-	// EastAsianLineBreaksStyleSimple is a style where soft line breaks are ignored
-	// if both sides of the break are east asian wide characters.
+	// EastAsianLineBreaksStyleSimple follows east_asian_line_breaks in Pandoc.
 	EastAsianLineBreaksStyleSimple EastAsianLineBreaksStyle = iota
-	// 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 follows CSS text level3 "Segment Break Transformation Rules" with some enhancements.
 	EastAsianLineBreaksCSS3Draft
 )
 
-type eastAsianLineBreaksFunction interface {
+type eastAsianLineBreaker interface {
 	SoftLineBreak(thisLastRune rune, siblingFirstRune rune) bool
 }
 
@@ -128,12 +127,53 @@ func (e *eastAsianLineBreaksSimple) SoftLineBreak(thisLastRune rune, siblingFirs
 type eastAsianLineBreaksCSS3Draft struct{}
 
 func (e *eastAsianLineBreaksCSS3Draft) SoftLineBreak(thisLastRune rune, siblingFirstRune rune) bool {
-	return !(util.IsEastAsianWideRune(thisLastRune) || util.IsEastAsianWideRune(siblingFirstRune))
+	// 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 eastAsianLineBreaks struct {
 	Enabled                     bool
-	EastAsianLineBreaksFunction eastAsianLineBreaksFunction
+	EastAsianLineBreaksFunction eastAsianLineBreaker
 }
 
 type withEastAsianLineBreaks struct {

util/util.go

@@ -836,22 +836,6 @@ 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 {
-	// https://en.wikipedia.org/wiki/CJK_Symbols_and_Punctuation
-	var CJKSymbolsAndPunctuation = &unicode.RangeTable{
-		R16: []unicode.Range16{
-			{0x3000, 0x303F, 1},
-		},
-	}
-	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)
-}
-
 // A BufWriter is a subset of the bufio.Writer .
 type BufWriter interface {
 	io.Writer