markup/goldmark: Add goldmark-cjk-friendly extension

Add `goldmark-cjk-friendly` extension which enhances handling of CJK emphasis and strikethrough.

Fixes #14114

docs/content/en/configuration/markup.md

@@ -46,6 +46,7 @@ The extensions below, excluding Extras and Passthrough, are enabled by default.
 Extension|Documentation|Enabled
 :--|:--|:-:
 `cjk`|[Goldmark Extensions: CJK]|:heavy_check_mark:
+`cjkFriendly`|[Goldmark Extensions: CJK Friendly]| 
 `definitionList`|[PHP Markdown Extra: Definition lists]|:heavy_check_mark:
 `extras`|[Hugo Goldmark Extensions: Extras]| 
 `footnote`|[PHP Markdown Extra: Footnotes]|:heavy_check_mark:
@@ -337,6 +338,7 @@ ordered
 [GitHub Flavored Markdown: Task list items]: https://github.github.com/gfm/#task-list-items-extension-
 [GitHub Flavored Markdown]: https://github.github.com/gfm/
 [Goldmark Extensions: CJK]: https://github.com/yuin/goldmark?tab=readme-ov-file#cjk-extension
+[Goldmark Extensions: CJK Friendly]: https://github.com/tats-u/goldmark-cjk-friendly
 [Goldmark Extensions: Typographer]: https://github.com/yuin/goldmark?tab=readme-ov-file#typographer-extension
 [Goldmark]: https://github.com/yuin/goldmark/
 [Hugo Goldmark Extensions: Extras]: https://github.com/gohugoio/hugo-goldmark-extensions?tab=readme-ov-file#extras-extension

docs/data/docs.yaml

@@ -1232,6 +1232,8 @@ config:
           eastAsianLineBreaksStyle: simple
           enable: false
           escapedSpace: false
+        cjkFriendly:
+          emphasis: false
         definitionList: true
         extras:
           delete:

go.mod

@@ -67,6 +67,7 @@ require (
 	github.com/spf13/cobra v1.10.2
 	github.com/spf13/fsync v0.10.1
 	github.com/spf13/pflag v1.0.9
+	github.com/tats-u/goldmark-cjk-friendly/v2 v2.0.2
 	github.com/tdewolff/minify/v2 v2.24.8
 	github.com/tdewolff/parse/v2 v2.8.5
 	github.com/tetratelabs/wazero v1.10.1

go.sum

@@ -505,6 +505,8 @@ github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO
 github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
 github.com/stretchr/testify v1.10.0 h1:Xv5erBjTwe/5IxqUQTdXv5kgmIvbHo3QQyRwhJsOfJA=
 github.com/stretchr/testify v1.10.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
+github.com/tats-u/goldmark-cjk-friendly/v2 v2.0.2 h1:+N9clTED51Tt55ljeXYXufmqS8wUXpLJW771aIGIRxo=
+github.com/tats-u/goldmark-cjk-friendly/v2 v2.0.2/go.mod h1:1Wm+kJwLMq/sr22CCei+eN593nanPDVn1eH91QdVPEI=
 github.com/tdewolff/minify/v2 v2.24.8 h1:58/VjsbevI4d5FGV0ZSuBrHMSSkH4MCH0sIz/eKIauE=
 github.com/tdewolff/minify/v2 v2.24.8/go.mod h1:0Ukj0CRpo/sW/nd8uZ4ccXaV1rEVIWA3dj8U7+Shhfw=
 github.com/tdewolff/parse/v2 v2.8.5 h1:ZmBiA/8Do5Rpk7bDye0jbbDUpXXbCdc3iah4VeUvwYU=

markup/goldmark/convert.go

@@ -27,6 +27,7 @@ import (
 	"github.com/gohugoio/hugo/markup/goldmark/internal/render"
 	"github.com/gohugoio/hugo/markup/goldmark/passthrough"
 	"github.com/gohugoio/hugo/markup/goldmark/tables"
+	cjkFriendly "github.com/tats-u/goldmark-cjk-friendly/v2"
 	"github.com/yuin/goldmark/util"
 
 	"github.com/yuin/goldmark"
@@ -153,7 +154,7 @@ func newMarkdown(pcfg converter.ProviderConfig) goldmark.Markdown {
 		extensions = append(extensions, tables.New())
 	}
 
-	if cfg.Extensions.Strikethrough {
+	if cfg.Extensions.Strikethrough && !cfg.Extensions.CJKFriendly.Emphasis {
 		extensions = append(extensions, extension.Strikethrough)
 	}
 
@@ -207,6 +208,14 @@ func newMarkdown(pcfg converter.ProviderConfig) goldmark.Markdown {
 		extensions = append(extensions, c)
 	}
 
+	if cfg.Extensions.CJKFriendly.Emphasis {
+		if cfg.Extensions.Strikethrough {
+			extensions = append(extensions, cjkFriendly.CJKFriendlyEmphasisAndStrikethrough)
+		} else {
+			extensions = append(extensions, cjkFriendly.CJKFriendlyEmphasis)
+		}
+	}
+
 	if cfg.Extensions.Passthrough.Enable {
 		extensions = append(extensions, passthrough.New(cfg.Extensions.Passthrough))
 	}

markup/goldmark/convert_test.go

@@ -746,6 +746,140 @@ escapedSpace=true
 	c.Assert(got, qt.Contains, "<p>私は太郎です。\nプログラミングが好きです。運動が苦手です。</p>\n")
 }
 
+func TestConvertCJKFriendlyWithExtensionEmphasisWithExtensionStrikethrough(t *testing.T) {
+	c := qt.New(t)
+
+	content := "Git **（ギット）**Hub\n~~（真美好）~~"
+
+	tests := []struct {
+		name                string
+		strikethrough       bool
+		cjkFriendlyEmphasis bool
+		expect              string
+	}{
+		{"noFriendly_noStrike", false, false, "<p>Git **（ギット）**Hub\n~~（真美好）~~</p>\n"},
+		{"friendly_noStrike", false, true, "<p>Git <strong>（ギット）</strong>Hub\n~~（真美好）~~</p>\n"},
+		{"noFriendly_strike", true, false, "<p>Git **（ギット）**Hub\n<del>（真美好）</del></p>\n"},
+		{"friendly_strike", true, true, "<p>Git <strong>（ギット）</strong>Hub\n<del>（真美好）</del></p>\n"},
+	}
+
+	for _, tt := range tests {
+		c.Run(tt.name, func(c *qt.C) {
+			confStr := fmt.Sprintf(`
+[markup]
+[markup.goldmark]
+[markup.goldmark.extensions]
+strikethrough=%v
+[markup.goldmark.extensions.CJKFriendly]
+emphasis=%v
+`, tt.strikethrough, tt.cjkFriendlyEmphasis)
+
+			cfg := config.FromTOMLConfigString(confStr)
+			conf := testconfig.GetTestConfig(nil, cfg)
+
+			b := convert(c, conf, content)
+			got := string(b.Bytes())
+
+			c.Assert(got, qt.Contains, tt.expect)
+		})
+	}
+}
+
+func TestConvertCJKFriendlyWithExtensionEmphasisWithExtensionStrikethroughWithCJKEscapedSpace(t *testing.T) {
+	c := qt.New(t)
+
+	content := "a\\ **()**\\ a𩸽**（）**𩸽~~（真美好）~~"
+
+	tests := []struct {
+		name                string
+		strikethrough       bool
+		escapedSpace        bool
+		cjkFriendlyEmphasis bool
+		expect              string
+	}{
+		{
+			name:                "noFriendly_noStrike_noEscaped",
+			strikethrough:       false,
+			escapedSpace:        false,
+			cjkFriendlyEmphasis: false,
+			expect:              "<p>a\\ <strong>()</strong>\\ a𩸽**（）**𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "noFriendly_noStrike_escaped",
+			strikethrough:       false,
+			escapedSpace:        true,
+			cjkFriendlyEmphasis: false,
+			expect:              "<p>a<strong>()</strong>a𩸽**（）**𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "noFriendly_strike_noEscaped",
+			strikethrough:       true,
+			escapedSpace:        false,
+			cjkFriendlyEmphasis: false,
+			expect:              "<p>a\\ <strong>()</strong>\\ a𩸽**（）**𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "noFriendly_strike_escaped",
+			strikethrough:       true,
+			escapedSpace:        true,
+			cjkFriendlyEmphasis: false,
+			expect:              "<p>a<strong>()</strong>a𩸽**（）**𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "friendly_noStrike_noEscaped",
+			strikethrough:       false,
+			escapedSpace:        false,
+			cjkFriendlyEmphasis: true,
+			expect:              "<p>a\\ <strong>()</strong>\\ a𩸽<strong>（）</strong>𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "friendly_noStrike_escaped",
+			strikethrough:       false,
+			escapedSpace:        true,
+			cjkFriendlyEmphasis: true,
+			expect:              "<p>a<strong>()</strong>a𩸽<strong>（）</strong>𩸽~~（真美好）~~</p>\n",
+		},
+		{
+			name:                "friendly_strike_noEscaped",
+			strikethrough:       true,
+			escapedSpace:        false,
+			cjkFriendlyEmphasis: true,
+			expect:              "<p>a\\ <strong>()</strong>\\ a𩸽<strong>（）</strong>𩸽<del>（真美好）</del></p>\n",
+		},
+		{
+			name:                "friendly_strike_escaped",
+			strikethrough:       true,
+			escapedSpace:        true,
+			cjkFriendlyEmphasis: true,
+			expect:              "<p>a<strong>()</strong>a𩸽<strong>（）</strong>𩸽<del>（真美好）</del></p>\n",
+		},
+	}
+
+	for _, tt := range tests {
+		c.Run(tt.name, func(c *qt.C) {
+			confStr := fmt.Sprintf(`
+[markup]
+[markup.goldmark]
+[markup.goldmark.extensions]
+strikethrough=%v
+[markup.goldmark.extensions.CJK]
+enable=true
+escapedSpace=%v
+[markup.goldmark.extensions.CJKFriendly]
+emphasis=%v
+`, tt.strikethrough, tt.escapedSpace, tt.cjkFriendlyEmphasis)
+
+			cfg := config.FromTOMLConfigString(confStr)
+			conf := testconfig.GetTestConfig(nil, cfg)
+
+			b := convert(c, conf, content)
+			got := string(b.Bytes())
+
+			c.Assert(got, qt.Contains, tt.expect)
+		})
+	}
+}
+
 type tableRenderer int
 
 func (hr tableRenderer) RenderTable(cctx context.Context, w hugio.FlexiWriter, ctx hooks.TableContext) error {

markup/goldmark/goldmark_config/config.go

@@ -57,6 +57,9 @@ var Default = Config{
 			EastAsianLineBreaksStyle: "simple",
 			EscapedSpace:             false,
 		},
+		CJKFriendly: CJKFriendly{
+			Emphasis: false,
+		},
 		Extras: Extras{
 			Delete: Delete{
 				Enable: false,
@@ -168,6 +171,7 @@ type Extensions struct {
 	LinkifyProtocol string
 	TaskList        bool
 	CJK             CJK
+	CJKFriendly     CJKFriendly `json:"cjkFriendly"`
 }
 
 // Footnote holds footnote configuration.
@@ -270,6 +274,11 @@ type CJK struct {
 	EscapedSpace bool
 }
 
+type CJKFriendly struct {
+	// Emphasis adds support for CJK-friendly 'emphasis'. If "strikethrough" goldmark extension is enabled as well, CJK-friendly 'emphasis and strikethrough' will be used.
+	Emphasis bool
+}
+
 type Renderer struct {
 	// Whether softline breaks should be rendered as '<br>'
 	HardWraps bool

markup/goldmark/goldmark_integration_test.go

@@ -828,6 +828,136 @@ H~2~0
 	)
 }
 
+func TestExtrasExtensionWithCJKFriendlyEmphasis(t *testing.T) {
+	t.Parallel()
+
+	files := `
+-- hugo.toml --
+disableKinds = ['page','rss','section','sitemap','taxonomy','term']
+[markup.goldmark.extensions]
+strikethrough = false
+[markup.goldmark.extensions.extras.delete]
+enable = false
+[markup.goldmark.extensions.extras.insert]
+enable = false
+[markup.goldmark.extensions.extras.mark]
+enable = false
+[markup.goldmark.extensions.extras.subscript]
+enable = false
+[markup.goldmark.extensions.extras.superscript]
+enable = false
+[markup.goldmark.extensions.CJKFriendly]
+emphasis = false
+-- layouts/index.html --
+{{ .Content }}
+-- content/_index.md --
+---
+title: home
+---
+~~削除~~
+
+++挿入++
+
+==マーク==
+
+（H~2~O）
+
+面積は1^乗^です
+
+（~~削除~~）
+
+~~（挿入）~~
+
+混合 **（強）**~~削除~~
+
+空 ~~ ~~
+
+++CJK**「ハロー」**Test++
+
+Mark**==マーク==**Test
+
+MarkParen**（==マーク==）**Test
+`
+
+	b := hugolib.Test(t, files)
+
+	// CJKFriendlyEmphasis disabled, Extras disabled.
+	b.AssertFileContent("public/index.html",
+		"<p>~~削除~~</p>",
+		"<p>++挿入++</p>",
+		"<p>==マーク==</p>",
+		"<p>（H~2~O）</p>",
+		"<p>面積は1^乗^です</p>",
+		"<p>（~~削除~~）</p>",
+		"<p>~~（挿入）~~</p>",
+		"<p>混合 <strong>（強）</strong>~~削除~~</p>",
+		"<p>空 ~~ ~~</p>",
+		"<p>++CJK**「ハロー」**Test++</p>",
+		"<p>Mark**==マーク==**Test</p>",
+		"<p>MarkParen**（==マーク==）**Test</p>",
+	)
+
+	files = strings.ReplaceAll(files, "enable = false", "enable = true")
+
+	b = hugolib.Test(t, files)
+
+	// CJKFriendlyEmphasis disabled, Extras enabled.
+	b.AssertFileContent("public/index.html",
+		"<p><del>削除</del></p>",
+		"<p><ins>挿入</ins></p>",
+		"<p><mark>マーク</mark></p>",
+		"<p>（H<sub>2</sub>O）</p>",
+		"<p>面積は1<sup>乗</sup>です</p>",
+		"<p>（<del>削除</del>）</p>",
+		"<p><del>（挿入）</del></p>",
+		"<p>混合 <strong>（強）</strong><del>削除</del></p>",
+		"<p>空 ~~ ~~</p>",
+		"<p><ins>CJK**「ハロー」**Test</ins></p>",
+		"<p>Mark**<mark>マーク</mark>**Test</p>",
+		"<p>MarkParen**（<mark>マーク</mark>）**Test</p>",
+	)
+
+	filesCJKEnabled := strings.Replace(files, "emphasis = false", "emphasis = true", 1)
+
+	b = hugolib.Test(t, filesCJKEnabled)
+
+	// CJKFriendlyEmphasis enabled, Extras enabled.
+	b.AssertFileContent("public/index.html",
+		"<p><del>削除</del></p>",
+		"<p><ins>挿入</ins></p>",
+		"<p><mark>マーク</mark></p>",
+		"<p>（H<sub>2</sub>O）</p>",
+		"<p>面積は1<sup>乗</sup>です</p>",
+		"<p>（<del>削除</del>）</p>",
+		"<p><del>（挿入）</del></p>",
+		"<p>混合 <strong>（強）</strong><del>削除</del></p>",
+		"<p>空 ~~ ~~</p>",
+		"<p><ins>CJK<strong>「ハロー」</strong>Test</ins></p>",
+		"<p>Mark**<mark>マーク</mark>**Test</p>",
+		"<p>MarkParen<strong>（<mark>マーク</mark>）</strong>Test</p>",
+	)
+
+	filesCJKEnabledExtrasDisabled := strings.ReplaceAll(filesCJKEnabled, "enable = true", "enable = false")
+
+	b = hugolib.Test(t, filesCJKEnabledExtrasDisabled)
+
+	// CJKFriendlyEmphasis enabled, Extras disabled.
+	b.AssertFileContent("public/index.html",
+		"<p>~~削除~~</p>",
+		"<p>++挿入++</p>",
+		"<p>==マーク==</p>",
+		"<p>（H~2~O）</p>",
+		"<p>面積は1^乗^です</p>",
+		"<p>（~~削除~~）</p>",
+		"<p>~~（挿入）~~</p>",
+		"<p>混合 <strong>（強）</strong>~~削除~~</p>",
+		"<p>空 ~~ ~~</p>",
+		"<p>++CJK<strong>「ハロー」</strong>Test++</p>",
+		"<p>Mark**==マーク==**Test</p>",
+		"<p>MarkParen<strong>（==マーク==）</strong>Test</p>",
+	)
+}
+
 // Issue 12997.
 func TestGoldmarkRawHTMLWarningBlocks(t *testing.T) {
 	files := `