markdown: Pass emoji codes to yuin/goldmark-emoji

Removes emoji code conversion from the page and shortcode parsers. Emoji
codes in markdown are now passed to Goldmark, where the goldmark-emoji
extension converts them to decimal numeric character references.

This disables emoji rendering for the alternate content formats: html,
asciidoc, org, pandoc, and rst.

Fixes #7332
Fixes #11587
Closes #11598

config/allconfig/configlanguage.go

@@ -228,3 +228,7 @@ func (c ConfigLanguage) PaginatePath() string {
 func (c ConfigLanguage) StaticDirs() []string {
 	return c.config.staticDirs()
 }
+
+func (c ConfigLanguage) EnableEmoji() bool {
+	return c.config.EnableEmoji
+}

config/configProvider.go

@@ -66,6 +66,7 @@ type AllProvider interface {
 	StaticDirs() []string
 	IgnoredErrors() map[string]bool
 	WorkingDir() string
+	EnableEmoji() bool
 }
 
 // Provider provides the configuration settings for Hugo.

go.mod

@@ -64,6 +64,7 @@ require (
 	github.com/tdewolff/minify/v2 v2.12.9
 	github.com/tdewolff/parse/v2 v2.6.8
 	github.com/yuin/goldmark v1.5.6
+	github.com/yuin/goldmark-emoji v1.0.2
 	go.uber.org/atomic v1.11.0
 	go.uber.org/automaxprocs v1.5.3
 	gocloud.dev v0.34.0

go.sum

@@ -444,9 +444,12 @@ github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9de
 github.com/yuin/goldmark v1.1.32/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.3.5/go.mod h1:mwnBkeHKe2W/ZEtQ+71ViKU8L12m81fl3OWwC1Zlc8k=
+github.com/yuin/goldmark v1.3.7/go.mod h1:mwnBkeHKe2W/ZEtQ+71ViKU8L12m81fl3OWwC1Zlc8k=
 github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
 github.com/yuin/goldmark v1.5.6 h1:COmQAWTCcGetChm3Ig7G/t8AFAN00t+o8Mt4cf7JpwA=
 github.com/yuin/goldmark v1.5.6/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
+github.com/yuin/goldmark-emoji v1.0.2 h1:c/RgTShNgHTtc6xdz2KKI74jJr6rWi7FPgnP9GAsO5s=
+github.com/yuin/goldmark-emoji v1.0.2/go.mod h1:RhP/RWpexdp+KHs7ghKnifRoIs/Bq4nDS7tRbCkOwKY=
 go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
 go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
 go.opencensus.io v0.22.2/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=

helpers/emoji_test.go

@@ -13,13 +13,8 @@
 package helpers
 
 import (
-	"math"
 	"reflect"
-	"strings"
 	"testing"
-
-	"github.com/gohugoio/hugo/bufferpool"
-	"github.com/kyokomi/emoji/v2"
 )
 
 func TestEmojiCustom(t *testing.T) {
@@ -68,76 +63,3 @@ func TestEmojiCustom(t *testing.T) {
 
 	}
 }
-
-// The Emoji benchmarks below are heavily skewed in Hugo's direction:
-//
-// Hugo have a byte slice, wants a byte slice and doesn't mind if the original is modified.
-
-func BenchmarkEmojiKyokomiFprint(b *testing.B) {
-	f := func(in []byte) []byte {
-		buff := bufferpool.GetBuffer()
-		defer bufferpool.PutBuffer(buff)
-		emoji.Fprint(buff, string(in))
-
-		bc := make([]byte, buff.Len())
-		copy(bc, buff.Bytes())
-		return bc
-	}
-
-	doBenchmarkEmoji(b, f)
-}
-
-func BenchmarkEmojiKyokomiSprint(b *testing.B) {
-	f := func(in []byte) []byte {
-		return []byte(emoji.Sprint(string(in)))
-	}
-
-	doBenchmarkEmoji(b, f)
-}
-
-func BenchmarkHugoEmoji(b *testing.B) {
-	doBenchmarkEmoji(b, Emojify)
-}
-
-func doBenchmarkEmoji(b *testing.B, f func(in []byte) []byte) {
-	type input struct {
-		in     []byte
-		expect []byte
-	}
-
-	data := []struct {
-		input  string
-		expect string
-	}{
-		{"A :smile: a day", emoji.Sprint("A :smile: a day")},
-		{"A :smile: and a :beer: day keeps the doctor away", emoji.Sprint("A :smile: and a :beer: day keeps the doctor away")},
-		{"A :smile: a day and 10 " + strings.Repeat(":beer: ", 10), emoji.Sprint("A :smile: a day and 10 " + strings.Repeat(":beer: ", 10))},
-		{"No smiles today.", "No smiles today."},
-		{"No smiles for you or " + strings.Repeat("you ", 1000), "No smiles for you or " + strings.Repeat("you ", 1000)},
-	}
-
-	in := make([]input, b.N*len(data))
-	cnt := 0
-	for i := 0; i < b.N; i++ {
-		for _, this := range data {
-			in[cnt] = input{[]byte(this.input), []byte(this.expect)}
-			cnt++
-		}
-	}
-
-	b.ResetTimer()
-	cnt = 0
-	for i := 0; i < b.N; i++ {
-		for j := range data {
-			currIn := in[cnt]
-			cnt++
-			result := f(currIn.in)
-			// The Emoji implementations gives slightly different output.
-			diffLen := len(result) - len(currIn.expect)
-			diffLen = int(math.Abs(float64(diffLen)))
-			if diffLen > 30 {
-				b.Fatalf("[%d] emoji std, got \n%q but expected \n%q", j, result, currIn.expect)
-			}
-		}
-	}
-}

hugolib/content_map_page.go

@@ -150,7 +150,7 @@ func (m *pageMap) newPageFromContentNode(n *contentNode, parentBucket *pagesMapB
 
 	parseResult, err := pageparser.Parse(
 		r,
-		pageparser.Config{EnableEmoji: s.conf.EnableEmoji},
+		pageparser.Config{},
 	)
 	if err != nil {
 		return nil, err

hugolib/page.go

@@ -794,11 +794,7 @@ Loop:
 			rn.AddShortcode(currShortcode)
 
 		case it.Type == pageparser.TypeEmoji:
-			if emoji := helpers.Emoji(it.ValStr(result.Input())); emoji != nil {
-				rn.AddReplacement(emoji, it)
-			} else {
 			rn.AddBytes(it)
-			}
 		case it.IsEOF():
 			break Loop
 		case it.IsError():

hugolib/page_test.go

@@ -1234,60 +1234,6 @@ func TestChompBOM(t *testing.T) {
 	checkPageTitle(t, p, "Simple")
 }
 
-func TestPageWithEmoji(t *testing.T) {
-	for _, enableEmoji := range []bool{true, false} {
-		v := config.New()
-		v.Set("enableEmoji", enableEmoji)
-
-		b := newTestSitesBuilder(t).WithViper(v)
-
-		b.WithContent("page-emoji.md", `---
-title: "Hugo Smile"
----
-This is a :smile:.
-<!--more--> 
-
-Another :smile: This is :not: :an: :emoji:.
-
-O :christmas_tree:
-
-Write me an :e-mail: or :email:?
-
-Too many colons: :: ::: :::: :?: :!: :.:
-
-If you dislike this video, you can hit that :-1: button :stuck_out_tongue_winking_eye:,
-but if you like it, hit :+1: and get subscribed!
-`)
-
-		b.CreateSites().Build(BuildCfg{})
-
-		if enableEmoji {
-			b.AssertFileContent("public/page-emoji/index.html",
-				"This is a 😄",
-				"Another 😄",
-				"This is :not: :an: :emoji:.",
-				"O 🎄",
-				"Write me an 📧 or ✉️?",
-				"Too many colons: :: ::: :::: :?: :!: :.:",
-				"you can hit that 👎 button 😜,",
-				"hit 👍 and get subscribed!",
-			)
-		} else {
-			b.AssertFileContent("public/page-emoji/index.html",
-				"This is a :smile:",
-				"Another :smile:",
-				"This is :not: :an: :emoji:.",
-				"O :christmas_tree:",
-				"Write me an :e-mail: or :email:?",
-				"Too many colons: :: ::: :::: :?: :!: :.:",
-				"you can hit that :-1: button :stuck_out_tongue_winking_eye:,",
-				"hit :+1: and get subscribed!",
-			)
-		}
-
-	}
-}
-
 func TestPageHTMLContent(t *testing.T) {
 	b := newTestSitesBuilder(t)
 	b.WithSimpleConfigFile()

hugolib/shortcode.go

@@ -26,8 +26,6 @@ import (
 	"strings"
 	"sync"
 
-	"github.com/gohugoio/hugo/helpers"
-
 	"errors"
 
 	"github.com/gohugoio/hugo/common/herrors"
@@ -632,14 +630,7 @@ Loop:
 		case currItem.IsText():
 			sc.inner = append(sc.inner, currItem.ValStr(source))
 		case currItem.Type == pageparser.TypeEmoji:
-			// TODO(bep) avoid the duplication of these "text cases", to prevent
-			// more of #6504 in the future.
-			val := currItem.ValStr(source)
-			if emoji := helpers.Emoji(val); emoji != nil {
-				sc.inner = append(sc.inner, string(emoji))
-			} else {
-				sc.inner = append(sc.inner, val)
-			}
+			sc.inner = append(sc.inner, currItem.ValStr(source))
 		case currItem.IsShortcodeName():
 
 			sc.name = currItem.ValStr(source)

hugolib/shortcode_test.go

@@ -754,33 +754,6 @@ title: "Hugo Rocks!"
 	)
 }
 
-// https://github.com/gohugoio/hugo/issues/6504
-func TestShortcodeEmoji(t *testing.T) {
-	t.Parallel()
-
-	v := config.New()
-	v.Set("enableEmoji", true)
-
-	builder := newTestSitesBuilder(t).WithViper(v)
-
-	builder.WithContent("page.md", `---
-title: "Hugo Rocks!"
----
-
-# doc
-
-{{< event >}}10:30-11:00 My :smile: Event {{< /event >}}
-
-
-`).WithTemplatesAdded(
-		"layouts/shortcodes/event.html", `<div>{{ "\u29BE" }} {{ .Inner }} </div>`)
-
-	builder.Build(BuildCfg{})
-	builder.AssertFileContent("public/page/index.html",
-		"⦾ 10:30-11:00 My 😄 Event",
-	)
-}
-
 func TestShortcodeParams(t *testing.T) {
 	t.Parallel()
 	c := qt.New(t)

hugolib/site.go

@@ -94,7 +94,6 @@ type siteConfigHolder struct {
 	taxonomiesConfig taxonomiesConfig
 	timeout          time.Duration
 	hasCJKLanguage   bool
-	enableEmoji      bool
 }
 
 // Lazily loaded site dependencies.

markup/goldmark/convert.go

@@ -28,6 +28,7 @@ import (
 	"github.com/gohugoio/hugo/markup/converter"
 	"github.com/gohugoio/hugo/markup/tableofcontents"
 	"github.com/yuin/goldmark"
+	emoji "github.com/yuin/goldmark-emoji"
 	"github.com/yuin/goldmark/ast"
 	"github.com/yuin/goldmark/extension"
 	"github.com/yuin/goldmark/parser"
@@ -149,6 +150,10 @@ func newMarkdown(pcfg converter.ProviderConfig) goldmark.Markdown {
 		extensions = append(extensions, c)
 	}
 
+	if pcfg.Conf.EnableEmoji() {
+		extensions = append(extensions, emoji.Emoji)
+	}
+
 	if cfg.Parser.AutoHeadingID {
 		parserOptions = append(parserOptions, parser.WithAutoHeadingID())
 	}
@@ -156,6 +161,7 @@ func newMarkdown(pcfg converter.ProviderConfig) goldmark.Markdown {
 	if cfg.Parser.Attribute.Title {
 		parserOptions = append(parserOptions, parser.WithAttribute())
 	}
+
 	if cfg.Parser.Attribute.Block {
 		extensions = append(extensions, attributes.New())
 	}

markup/goldmark/integration_test.go

@@ -578,3 +578,114 @@ a <!-- b --> c
 		"<li>This is a list item <!-- Comment: an innocent-looking comment --></li>",
 	)
 }
+
+// Issue #7332
+// Issue #11587
+func TestGoldmarkEmojiExtension(t *testing.T) {
+	t.Parallel()
+
+	files := `
+-- config.toml --
+enableEmoji = true
+-- content/p1.md --
+---
+title: "p1"
+---
+~~~text
+:x:
+~~~
+
+{{% include "/p2" %}}
+
+{{< sc1 >}}:smiley:{{< /sc1 >}}
+
+{{< sc2 >}}:+1:{{< /sc2 >}}
+
+{{% sc3 %}}:-1:{{% /sc3 %}}
+
+-- content/p2.md --
+---
+title: "p2"
+---
+:heavy_check_mark:
+-- layouts/shortcodes/include.html --
+{{ $p := site.GetPage (.Get 0) }}
+{{ $p.RenderShortcodes }}
+-- layouts/shortcodes/sc1.html --
+sc1_begin|{{ .Inner }}|sc1_end
+-- layouts/shortcodes/sc2.html --
+sc2_begin|{{ .Inner | .Page.RenderString }}|sc2_end
+-- layouts/shortcodes/sc3.html --
+sc3_begin|{{ .Inner }}|sc3_end
+-- layouts/_default/single.html --
+{{ .Content }}
+`
+
+	b := hugolib.NewIntegrationTestBuilder(
+		hugolib.IntegrationTestConfig{
+			T:           t,
+			TxtarString: files,
+		},
+	).Build()
+
+	b.AssertFileContentExact("public/p1/index.html",
+		// Issue #7332
+		"<span>:x:\n</span>",
+		// Issue #11587
+		"<p>&#x2714;&#xfe0f;</p>",
+		// Should not be converted to emoji
+		"sc1_begin|:smiley:|sc1_end",
+		// Should be converted to emoji
+		"sc2_begin|&#x1f44d;|sc2_end",
+		// Should be converted to emoji
+		"sc3_begin|&#x1f44e;|sc3_end",
+	)
+}
+
+func TestEmojiDisabled(t *testing.T) {
+	t.Parallel()
+
+	files := `
+-- config.toml --
+enableEmoji = false
+-- content/p1.md --
+---
+title: "p1"
+---
+:x:
+-- layouts/_default/single.html --
+{{ .Content }}
+`
+
+	b := hugolib.NewIntegrationTestBuilder(
+		hugolib.IntegrationTestConfig{
+			T:           t,
+			TxtarString: files,
+		},
+	).Build()
+
+	b.AssertFileContentExact("public/p1/index.html", "<p>:x:</p>")
+}
+
+func TestEmojiDefaultConfig(t *testing.T) {
+	t.Parallel()
+
+	files := `
+-- content/p1.md --
+---
+title: "p1"
+---
+:x:
+-- layouts/_default/single.html --
+{{ .Content }}
+`
+
+	b := hugolib.NewIntegrationTestBuilder(
+		hugolib.IntegrationTestConfig{
+			T:           t,
+			TxtarString: files,
+		},
+	).Build()
+
+	b.AssertFileContentExact("public/p1/index.html", "<p>:x:</p>")
+}

parser/pageparser/pagelexer.go

@@ -61,9 +61,7 @@ func (l *pageLexer) Input() []byte {
 	return l.input
 }
 
-type Config struct {
-	EnableEmoji bool
-}
+type Config struct{}
 
 // note: the input position here is normally 0 (start), but
 // can be set if position of first shortcode is known
@@ -103,8 +101,6 @@ var (
 	delimOrg          = []byte("#+")
 	htmlCommentStart  = []byte("<!--")
 	htmlCommentEnd    = []byte("-->")
-
-	emojiDelim = byte(':')
 )
 
 func (l *pageLexer) next() rune {
@@ -276,34 +272,6 @@ func (l *pageLexer) consumeSpace() {
 	}
 }
 
-// lex a string starting at ":"
-func lexEmoji(l *pageLexer) stateFunc {
-	pos := l.pos + 1
-	valid := false
-
-	for i := pos; i < len(l.input); i++ {
-		if i > pos && l.input[i] == emojiDelim {
-			pos = i + 1
-			valid = true
-			break
-		}
-		r, _ := utf8.DecodeRune(l.input[i:])
-		if !(isAlphaNumericOrHyphen(r) || r == '+') {
-			break
-		}
-	}
-
-	if valid {
-		l.pos = pos
-		l.emit(TypeEmoji)
-	} else {
-		l.pos++
-		l.emit(tText)
-	}
-
-	return lexMainSection
-}
-
 type sectionHandlers struct {
 	l *pageLexer
 
@@ -399,20 +367,6 @@ func createSectionHandlers(l *pageLexer) *sectionHandlers {
 
 	handlers := []*sectionHandler{shortCodeHandler, summaryDividerHandler}
 
-	if l.cfg.EnableEmoji {
-		emojiHandler := &sectionHandler{
-			l: l,
-			skipFunc: func(l *pageLexer) int {
-				return l.indexByte(emojiDelim)
-			},
-			lexFunc: func(origin stateFunc, l *pageLexer) (stateFunc, bool) {
-				return lexEmoji, true
-			},
-		}
-
-		handlers = append(handlers, emojiHandler)
-	}
-
 	return &sectionHandlers{
 		l:           l,
 		handlers:    handlers,

parser/pageparser/pageparser_main_test.go

@@ -1,42 +0,0 @@
-// Copyright 2018 The Hugo Authors. All rights reserved.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-// http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-package pageparser
-
-import (
-	"testing"
-
-	qt "github.com/frankban/quicktest"
-)
-
-func TestMain(t *testing.T) {
-	t.Parallel()
-	c := qt.New(t)
-
-	mainTests := []lexerTest{
-		{"emoji #1", "Some text with :emoji:", []typeText{nti(tText, "Some text with "), nti(TypeEmoji, ":emoji:"), tstEOF}},
-		{"emoji #2", "Some text with :emoji: and some text.", []typeText{nti(tText, "Some text with "), nti(TypeEmoji, ":emoji:"), nti(tText, " and some text."), tstEOF}},
-		{"looks like an emoji #1", "Some text and then :emoji", []typeText{nti(tText, "Some text and then "), nti(tText, ":"), nti(tText, "emoji"), tstEOF}},
-		{"looks like an emoji #2", "Some text and then ::", []typeText{nti(tText, "Some text and then "), nti(tText, ":"), nti(tText, ":"), tstEOF}},
-		{"looks like an emoji #3", ":Some :text", []typeText{nti(tText, ":"), nti(tText, "Some "), nti(tText, ":"), nti(tText, "text"), tstEOF}},
-	}
-
-	for i, test := range mainTests {
-		items := collectWithConfig([]byte(test.input), false, lexMainSection, Config{EnableEmoji: true})
-		if !equal(test.input, items, test.items) {
-			got := itemsToString(items, []byte(test.input))
-			expected := testItemsToString(test.items)
-			c.Assert(got, qt.Equals, expected, qt.Commentf("Test %d: %s", i, test.name))
-		}
-	}
-}

parser/pageparser/pageparser_test.go

@@ -38,33 +38,7 @@ This is some summary. This is some summary. This is some summary. This is some s
 
 `
 	input := []byte(start + strings.Repeat(strings.Repeat("this is text", 30)+"{{< myshortcode >}}This is some inner content.{{< /myshortcode >}}", 10))
-	cfg := Config{EnableEmoji: false}
-
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		if _, err := parseBytes(input, cfg, lexIntroSection); err != nil {
-			b.Fatal(err)
-		}
-	}
-}
-
-func BenchmarkParseWithEmoji(b *testing.B) {
-	start := `
-	
-
----
-title: "Front Matters"
-description: "It really does"
----
-
-This is some summary. This is some summary. This is some summary. This is some summary.
-
- <!--more-->
-
-
-`
-	input := []byte(start + strings.Repeat("this is not emoji: ", 50) + strings.Repeat("some text ", 70) + strings.Repeat("this is not: ", 50) + strings.Repeat("but this is a :smile: ", 3) + strings.Repeat("some text ", 70))
-	cfg := Config{EnableEmoji: true}
+	cfg := Config{}
 
 	b.ResetTimer()
 	for i := 0; i < b.N; i++ {