Support unicode emojis and remove emojify.js (#11032)

* Support unicode emojis and remove emojify.js

This PR replaces all use of emojify.js and adds unicode emoji support to various areas of gitea.

This works in a few ways:

First it adds emoji parsing support into gitea itself. This allows us to

 * Render emojis from valid alias (😄)
 * Detect unicode emojis and let us put them in their own class with proper aria-labels and styling
 * Easily allow for custom "emoji"
 * Support all emoji rendering and features without javascript
 * Uses plain unicode and lets the system render in appropriate emoji font
 * Doesn't leave us relying on external sources for updates/fixes/features

That same list of emoji is also used to create a json file which replaces the part of emojify.js that populates the emoji search tribute. This file is about 35KB with GZIP turned on and I've set it to load after the page renders to not hinder page load time (and this removes loading emojify.js also)

For custom "emoji" it uses a pretty simple scheme of just looking for /emojis/img/name.png where name is something a user has put in the "allowed reactions" setting we already have. The gitea reaction that was previously hard coded into a forked copy of emojify.js is included and works as a custom reaction under this method.

The emoji data sourced here is from https://github.com/github/gemoji which is the gem library Github uses for their emoji rendering (and a data source for other sites). So we should be able to easily render any emoji and :alias: that Github can, removing any errors from migrated content. They also update it as well, so we can sync when there are new unicode emoji lists released.

I've included a slimmed down and slightly modified forked copy of https://github.com/knq/emoji to make up our own emoji module. The code is pretty straight forward and again allows us to have a lot of flexibility in what happens.

I had seen a few comments about performance in some of the other threads if we render this ourselves, but there doesn't seem to be any issue here. In a test it can parse, convert, and render 1,000 emojis inside of a large markdown table in about 100ms on my laptop (which is many more emojis than will ever be in any normal issue). This also prevents any flickering and other weirdness from using javascript to render some things while using go for others.

Not included here are image fall back URLS. I don't really think they are necessary for anything new being written in 2020. However, managing the emoji ourselves would allow us to add these as a feature later on if it seems necessary.

Fixes: https://github.com/go-gitea/gitea/issues/9182
Fixes: https://github.com/go-gitea/gitea/issues/8974
Fixes: https://github.com/go-gitea/gitea/issues/8953
Fixes: https://github.com/go-gitea/gitea/issues/6628
Fixes: https://github.com/go-gitea/gitea/issues/5130

* add new shared function emojiHTML

* don't increase emoji size in issue title

* Update templates/repo/issue/view_content/add_reaction.tmpl

Co-Authored-By: 6543 <6543@obermui.de>

* Support for emoji rendering in various templates

* Render code and review comments as they should be

* Better way to handle mail subjects

* insert unicode from tribute selection

* Add template helper for plain text when needed

* Use existing replace function I forgot about

* Don't include emoji greater than Unicode Version 12

Only include emoji and aliases in JSON

* Update build/generate-emoji.go

* Tweak regex slightly to really match everything including random invisible characters. Run tests for every emoji we have

* final updates

* code review

* code review

* hard code gitea custom emoji to match previous behavior

* Update .eslintrc

Co-Authored-By: silverwind <me@silverwind.io>

* disable preempt

Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: 6543 <6543@obermui.de>
Co-authored-by: Lauris BH <lauris@nix.lv>
Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com>

modules/emoji/emoji.go

@@ -0,0 +1,119 @@
+// Copyright 2020 The Gitea Authors. All rights reserved.
+// Copyright 2015 Kenneth Shaw
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package emoji
+
+import (
+	"strings"
+	"sync"
+)
+
+// Gemoji is a set of emoji data.
+type Gemoji []Emoji
+
+// Emoji represents a single emoji and associated data.
+type Emoji struct {
+	Emoji          string
+	Description    string
+	Aliases        []string
+	UnicodeVersion string
+}
+
+var (
+	// codeMap provides a map of the emoji unicode code to its emoji data.
+	codeMap map[string]int
+
+	// aliasMap provides a map of the alias to its emoji data.
+	aliasMap map[string]int
+
+	// codeReplacer is the string replacer for emoji codes.
+	codeReplacer *strings.Replacer
+
+	// aliasReplacer is the string replacer for emoji aliases.
+	aliasReplacer *strings.Replacer
+
+	once sync.Once
+)
+
+func loadMap() {
+
+	once.Do(func() {
+
+		// initialize
+		codeMap = make(map[string]int, len(GemojiData))
+		aliasMap = make(map[string]int, len(GemojiData))
+
+		// process emoji codes and aliases
+		codePairs := make([]string, 0)
+		aliasPairs := make([]string, 0)
+		for i, e := range GemojiData {
+			if e.Emoji == "" || len(e.Aliases) == 0 {
+				continue
+			}
+
+			// setup codes
+			codeMap[e.Emoji] = i
+			codePairs = append(codePairs, e.Emoji, ":"+e.Aliases[0]+":")
+
+			// setup aliases
+			for _, a := range e.Aliases {
+				if a == "" {
+					continue
+				}
+
+				aliasMap[a] = i
+				aliasPairs = append(aliasPairs, ":"+a+":", e.Emoji)
+			}
+		}
+
+		// create replacers
+		codeReplacer = strings.NewReplacer(codePairs...)
+		aliasReplacer = strings.NewReplacer(aliasPairs...)
+	})
+}
+
+// FromCode retrieves the emoji data based on the provided unicode code (ie,
+// "\u2618" will return the Gemoji data for "shamrock").
+func FromCode(code string) *Emoji {
+	loadMap()
+	i, ok := codeMap[code]
+	if !ok {
+		return nil
+	}
+
+	return &GemojiData[i]
+}
+
+// FromAlias retrieves the emoji data based on the provided alias in the form
+// "alias" or ":alias:" (ie, "shamrock" or ":shamrock:" will return the Gemoji
+// data for "shamrock").
+func FromAlias(alias string) *Emoji {
+	loadMap()
+	if strings.HasPrefix(alias, ":") && strings.HasSuffix(alias, ":") {
+		alias = alias[1 : len(alias)-1]
+	}
+
+	i, ok := aliasMap[alias]
+	if !ok {
+		return nil
+	}
+
+	return &GemojiData[i]
+}
+
+// ReplaceCodes replaces all emoji codes with the first corresponding emoji
+// alias (in the form of ":alias:") (ie, "\u2618" will be converted to
+// ":shamrock:").
+func ReplaceCodes(s string) string {
+	loadMap()
+	return codeReplacer.Replace(s)
+}
+
+// ReplaceAliases replaces all aliases of the form ":alias:" with its
+// corresponding unicode value.
+func ReplaceAliases(s string) string {
+	loadMap()
+	return aliasReplacer.Replace(s)
+}

modules/emoji/emoji_test.go

@@ -0,0 +1,67 @@
+// Copyright 2020 The Gitea Authors. All rights reserved.
+// Copyright 2015 Kenneth Shaw
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package emoji
+
+import (
+	"reflect"
+	"testing"
+)
+
+func TestDumpInfo(t *testing.T) {
+	t.Logf("codes: %d", len(codeMap))
+	t.Logf("aliases: %d", len(aliasMap))
+}
+
+func TestLookup(t *testing.T) {
+	a := FromCode("\U0001f37a")
+	b := FromCode("🍺")
+	c := FromAlias(":beer:")
+	d := FromAlias("beer")
+
+	if !reflect.DeepEqual(a, b) {
+		t.Errorf("a and b should equal")
+	}
+	if !reflect.DeepEqual(b, c) {
+		t.Errorf("b and c should equal")
+	}
+	if !reflect.DeepEqual(c, d) {
+		t.Errorf("c and d should equal")
+	}
+	if !reflect.DeepEqual(a, d) {
+		t.Errorf("a and d should equal")
+	}
+
+	m := FromCode("\U0001f44d")
+	n := FromAlias(":thumbsup:")
+	o := FromAlias("+1")
+
+	if !reflect.DeepEqual(m, n) {
+		t.Errorf("m and n should equal")
+	}
+	if !reflect.DeepEqual(n, o) {
+		t.Errorf("n and o should equal")
+	}
+	if !reflect.DeepEqual(m, o) {
+		t.Errorf("m and o should equal")
+	}
+}
+
+func TestReplacers(t *testing.T) {
+	tests := []struct {
+		f      func(string) string
+		v, exp string
+	}{
+		{ReplaceCodes, ":thumbsup: +1 for \U0001f37a! 🍺 \U0001f44d", ":thumbsup: +1 for :beer:! :beer: :+1:"},
+		{ReplaceAliases, ":thumbsup: +1 :+1: :beer:", "\U0001f44d +1 \U0001f44d \U0001f37a"},
+	}
+
+	for i, x := range tests {
+		s := x.f(x.v)
+		if s != x.exp {
+			t.Errorf("test %d `%s` expected `%s`, got: `%s`", i, x.v, x.exp, s)
+		}
+	}
+}

modules/git/command.go

@@ -98,7 +98,6 @@ func (c *Command) RunInDirTimeoutEnvFullPipeline(env []string, timeout time.Dura
 // RunInDirTimeoutEnvFullPipelineFunc executes the command in given directory with given timeout,
 // it pipes stdout and stderr to given io.Writer and passes in an io.Reader as stdin. Between cmd.Start and cmd.Wait the passed in function is run.
 func (c *Command) RunInDirTimeoutEnvFullPipelineFunc(env []string, timeout time.Duration, dir string, stdout, stderr io.Writer, stdin io.Reader, fn func(context.Context, context.CancelFunc) error) error {
-
 	if timeout == -1 {
 		timeout = DefaultCommandExecutionTimeout
 	}

modules/markup/html.go

@@ -6,6 +6,7 @@ package markup
 
 import (
 	"bytes"
+	"fmt"
 	"net/url"
 	"path"
 	"path/filepath"
@@ -13,6 +14,7 @@ import (
 	"strings"
 
 	"code.gitea.io/gitea/modules/base"
+	"code.gitea.io/gitea/modules/emoji"
 	"code.gitea.io/gitea/modules/git"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/markup/common"
@@ -60,6 +62,13 @@ var (
 
 	// blackfriday extensions create IDs like fn:user-content-footnote
 	blackfridayExtRegex = regexp.MustCompile(`[^:]*:user-content-`)
+
+	// EmojiShortCodeRegex find emoji by alias like :smile:
+	EmojiShortCodeRegex = regexp.MustCompile(`\:[\w\+\-]+\:{1}`)
+
+	// find emoji literal: search all emoji hex range as many times as they appear as
+	// some emojis (skin color etc..) are just two or more chained together
+	emojiRegex = regexp.MustCompile(`[\x{1F000}-\x{1FFFF}|\x{2000}-\x{32ff}|\x{fe4e5}-\x{fe4ee}|\x{200D}|\x{FE0F}|\x{e0000}-\x{e007f}]+`)
 )
 
 // CSS class for action keywords (e.g. "closes: #1")
@@ -154,6 +163,8 @@ var defaultProcessors = []processor{
 	issueIndexPatternProcessor,
 	sha1CurrentPatternProcessor,
 	emailAddressProcessor,
+	emojiProcessor,
+	emojiShortCodeProcessor,
 }
 
 type postProcessCtx struct {
@@ -194,6 +205,8 @@ var commitMessageProcessors = []processor{
 	issueIndexPatternProcessor,
 	sha1CurrentPatternProcessor,
 	emailAddressProcessor,
+	emojiProcessor,
+	emojiShortCodeProcessor,
 }
 
 // RenderCommitMessage will use the same logic as PostProcess, but will disable
@@ -226,6 +239,13 @@ var commitMessageSubjectProcessors = []processor{
 	mentionProcessor,
 	issueIndexPatternProcessor,
 	sha1CurrentPatternProcessor,
+	emojiShortCodeProcessor,
+	emojiProcessor,
+}
+
+var emojiProcessors = []processor{
+	emojiShortCodeProcessor,
+	emojiProcessor,
 }
 
 // RenderCommitMessageSubject will use the same logic as PostProcess and
@@ -269,6 +289,17 @@ func RenderDescriptionHTML(
 	return ctx.postProcess(rawHTML)
 }
 
+// RenderEmoji for when we want to just process emoji and shortcodes
+// in various places it isn't already run through the normal markdown procesor
+func RenderEmoji(
+	rawHTML []byte,
+) ([]byte, error) {
+	ctx := &postProcessCtx{
+		procs: emojiProcessors,
+	}
+	return ctx.postProcess(rawHTML)
+}
+
 var byteBodyTag = []byte("<body>")
 var byteBodyTagClosing = []byte("</body>")
 
@@ -319,7 +350,12 @@ func (ctx *postProcessCtx) visitNode(node *html.Node, visitText bool) {
 		if attr.Key == "id" && !(strings.HasPrefix(attr.Val, "user-content-") || blackfridayExtRegex.MatchString(attr.Val)) {
 			node.Attr[idx].Val = "user-content-" + attr.Val
 		}
+
+		if attr.Key == "class" && attr.Val == "emoji" {
+			visitText = false
+		}
 	}
+
 	// We ignore code, pre and already generated links.
 	switch node.Type {
 	case html.TextNode:
@@ -406,6 +442,54 @@ func createKeyword(content string) *html.Node {
 	return span
 }
 
+func createEmoji(content, class, name string) *html.Node {
+	span := &html.Node{
+		Type: html.ElementNode,
+		Data: atom.Span.String(),
+		Attr: []html.Attribute{},
+	}
+	if class != "" {
+		span.Attr = append(span.Attr, html.Attribute{Key: "class", Val: class})
+	}
+	if name != "" {
+		span.Attr = append(span.Attr, html.Attribute{Key: "aria-label", Val: name})
+	}
+
+	text := &html.Node{
+		Type: html.TextNode,
+		Data: content,
+	}
+
+	span.AppendChild(text)
+	return span
+}
+
+func createCustomEmoji(alias, class string) *html.Node {
+
+	span := &html.Node{
+		Type: html.ElementNode,
+		Data: atom.Span.String(),
+		Attr: []html.Attribute{},
+	}
+	if class != "" {
+		span.Attr = append(span.Attr, html.Attribute{Key: "class", Val: class})
+		span.Attr = append(span.Attr, html.Attribute{Key: "aria-label", Val: alias})
+	}
+
+	img := &html.Node{
+		Type:     html.ElementNode,
+		DataAtom: atom.Img,
+		Data:     "img",
+		Attr:     []html.Attribute{},
+	}
+	if class != "" {
+		img.Attr = append(img.Attr, html.Attribute{Key: "src", Val: fmt.Sprintf(`%s/img/emoji/%s.png`, setting.StaticURLPrefix, alias)})
+	}
+
+	span.AppendChild(img)
+	return span
+}
+
 func createLink(href, content, class string) *html.Node {
 	a := &html.Node{
 		Type: html.ElementNode,
@@ -810,6 +894,45 @@ func fullSha1PatternProcessor(ctx *postProcessCtx, node *html.Node) {
 	replaceContent(node, start, end, createCodeLink(urlFull, text, "commit"))
 }
 
+// emojiShortCodeProcessor for rendering text like :smile: into emoji
+func emojiShortCodeProcessor(ctx *postProcessCtx, node *html.Node) {
+
+	m := EmojiShortCodeRegex.FindStringSubmatchIndex(node.Data)
+	if m == nil {
+		return
+	}
+
+	alias := node.Data[m[0]:m[1]]
+	alias = strings.Replace(alias, ":", "", -1)
+	converted := emoji.FromAlias(alias)
+	if converted == nil {
+		// check if this is a custom reaction
+		s := strings.Join(setting.UI.Reactions, " ") + "gitea"
+		if strings.Contains(s, alias) {
+			replaceContent(node, m[0], m[1], createCustomEmoji(alias, "emoji"))
+			return
+		}
+		return
+	}
+
+	replaceContent(node, m[0], m[1], createEmoji(converted.Emoji, "emoji", converted.Description))
+}
+
+// emoji processor to match emoji and add emoji class
+func emojiProcessor(ctx *postProcessCtx, node *html.Node) {
+	m := emojiRegex.FindStringSubmatchIndex(node.Data)
+
+	if m == nil {
+		return
+	}
+
+	codepoint := node.Data[m[0]:m[1]]
+	val := emoji.FromCode(codepoint)
+	if val != nil {
+		replaceContent(node, m[0], m[1], createEmoji(codepoint, "emoji", val.Description))
+	}
+}
+
 // sha1CurrentPatternProcessor renders SHA1 strings to corresponding links that
 // are assumed to be in the same repository.
 func sha1CurrentPatternProcessor(ctx *postProcessCtx, node *html.Node) {

modules/markup/html_test.go

@@ -8,6 +8,7 @@ import (
 	"strings"
 	"testing"
 
+	"code.gitea.io/gitea/modules/emoji"
 	. "code.gitea.io/gitea/modules/markup"
 	"code.gitea.io/gitea/modules/markup/markdown"
 	"code.gitea.io/gitea/modules/setting"
@@ -228,6 +229,50 @@ func TestRender_email(t *testing.T) {
 		`<p>email@domain..com</p>`)
 }
 
+func TestRender_emoji(t *testing.T) {
+	setting.AppURL = AppURL
+	setting.AppSubURL = AppSubURL
+	setting.StaticURLPrefix = AppURL
+
+	test := func(input, expected string) {
+		expected = strings.Replace(expected, "&", "&amp;", -1)
+		buffer := RenderString("a.md", input, setting.AppSubURL, nil)
+		assert.Equal(t, strings.TrimSpace(expected), strings.TrimSpace(buffer))
+	}
+
+	// Make sure we can successfully match every emoji in our dataset with regex
+	for i := range emoji.GemojiData {
+		test(
+			emoji.GemojiData[i].Emoji,
+			`<p><span class="emoji" aria-label="`+emoji.GemojiData[i].Description+`">`+emoji.GemojiData[i].Emoji+`</span></p>`)
+	}
+	for i := range emoji.GemojiData {
+		test(
+			":"+emoji.GemojiData[i].Aliases[0]+":",
+			`<p><span class="emoji" aria-label="`+emoji.GemojiData[i].Description+`">`+emoji.GemojiData[i].Emoji+`</span></p>`)
+	}
+
+	//Text that should be turned into or recognized as emoji
+	test(
+		":gitea:",
+		`<p><span class="emoji" aria-label="gitea"><img src="`+setting.StaticURLPrefix+`/img/emoji/gitea.png"/></span></p>`)
+
+	test(
+		"Some text with 😄 in the middle",
+		`<p>Some text with <span class="emoji" aria-label="grinning face with smiling eyes">😄</span> in the middle</p>`)
+	test(
+		"Some text with :smile: in the middle",
+		`<p>Some text with <span class="emoji" aria-label="grinning face with smiling eyes">😄</span> in the middle</p>`)
+
+	// should match nothing
+	test(
+		"2001:0db8:85a3:0000:0000:8a2e:0370:7334",
+		`<p>2001:0db8:85a3:0000:0000:8a2e:0370:7334</p>`)
+	test(
+		":not exist:",
+		`<p>:not exist:</p>`)
+}
+
 func TestRender_ShortLinks(t *testing.T) {
 	setting.AppURL = AppURL
 	setting.AppSubURL = AppSubURL

modules/markup/sanitizer.go

@@ -63,6 +63,10 @@ func ReplaceSanitizer() {
 	// Allow unlabelled labels
 	sanitizer.policy.AllowNoAttrs().OnElements("label")
 
+	// Allow classes for emojis
+	sanitizer.policy.AllowAttrs("class").Matching(regexp.MustCompile(`emoji`)).OnElements("span")
+	sanitizer.policy.AllowAttrs("class").Matching(regexp.MustCompile(`emoji`)).OnElements("img")
+
 	// Allow generally safe attributes
 	generalSafeAttrs := []string{"abbr", "accept", "accept-charset",
 		"accesskey", "action", "align", "alt",

modules/templates/helper.go

@@ -25,6 +25,7 @@ import (
 
 	"code.gitea.io/gitea/models"
 	"code.gitea.io/gitea/modules/base"
+	"code.gitea.io/gitea/modules/emoji"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/markup"
 	"code.gitea.io/gitea/modules/repository"
@@ -139,6 +140,9 @@ func NewFuncMap() []template.FuncMap {
 		"RenderCommitMessageLink":        RenderCommitMessageLink,
 		"RenderCommitMessageLinkSubject": RenderCommitMessageLinkSubject,
 		"RenderCommitBody":               RenderCommitBody,
+		"RenderEmoji":                    RenderEmoji,
+		"RenderEmojiPlain":               emoji.ReplaceAliases,
+		"ReactionToEmoji":                ReactionToEmoji,
 		"RenderNote":                     RenderNote,
 		"IsMultilineCommitMessage":       IsMultilineCommitMessage,
 		"ThemeColorMetaTag": func() string {
@@ -512,6 +516,29 @@ func RenderCommitBody(msg, urlPrefix string, metas map[string]string) template.H
 	return template.HTML(renderedMessage)
 }
 
+// RenderEmoji renders html text with emoji post processors
+func RenderEmoji(text string) template.HTML {
+	renderedText, err := markup.RenderEmoji([]byte(template.HTMLEscapeString(text)))
+	if err != nil {
+		log.Error("RenderEmoji: %v", err)
+		return template.HTML("")
+	}
+	return template.HTML(renderedText)
+}
+
+//ReactionToEmoji renders emoji for use in reactions
+func ReactionToEmoji(reaction string) template.HTML {
+	val := emoji.FromCode(reaction)
+	if val != nil {
+		return template.HTML(val.Emoji)
+	}
+	val = emoji.FromAlias(reaction)
+	if val != nil {
+		return template.HTML(val.Emoji)
+	}
+	return template.HTML(fmt.Sprintf(`<img src=%s/img/emoji/%s.png></img>`, setting.StaticURLPrefix, reaction))
+}
+
 // RenderNote renders the contents of a git-notes file as a commit message.
 func RenderNote(msg, urlPrefix string, metas map[string]string) template.HTML {
 	cleanMsg := template.HTMLEscapeString(msg)