x

2023-06-05 16:05:25 +08:00
parent dab997e54a
commit 0599f15f9e
6 changed files with 47750 additions and 1 deletions
@@ -7,6 +7,8 @@ import (
 	"regexp"
 	"strconv"
 	"strings"
+
+	"github.com/cdle/sillyplus/emoji"
 )

 type Strings struct {
@@ -159,3 +161,14 @@ func convertParams(params map[string]interface{}) map[string]string {
 	}
 	return result
 }
+
+func (sender *Strings) ReplaceEmojis(str string, f func([]string) string) string {
+	return emoji.ReplaceEmojisWithFunc(str, func(e emoji.Emoji) string {
+		return f(e.CodePoint2)
+	})
+}
+
+// `\[emoji=([0-9A-Z]{4})\]`
+func (sender *Strings) ReplaceToEmojis(str string, pattern string) string {
+	return emoji.ReplaceToEmojis(str, pattern)
+}
@@ -0,0 +1,176 @@
+package emoji
+
+import (
+	"fmt"
+	"regexp"
+	"strconv"
+	"strings"
+	"unicode/utf16"
+)
+
+func init() {
+	for i, emoji := range emojiMap {
+		code_points := []string{}
+		for _, v := range strings.Split(emojiMap[i].CodePoint, " ") {
+			switch len(v) {
+			case 4:
+				code_points = append(code_points, v)
+			case 5:
+				code_points = append(code_points, transfer(v)...)
+			default:
+			}
+		}
+		if len(code_points) == 0 {
+			for _, v := range strings.Split(emojiMap[i].CodePoint, "-") {
+				switch len(v) {
+				case 4:
+					code_points = append(code_points, v)
+				case 5:
+					code_points = append(code_points, transfer(v)...)
+				default:
+				}
+			}
+		}
+		if len(code_points) == 0 {
+			delete(emojiMap, i)
+		} else {
+			emoji.CodePoint2 = code_points
+			emojiMap[i] = emoji
+		}
+	}
+}
+
+func ConvertRegex(input string) string {
+	// 定义要转义的字符集合
+	escapeSet := []string{"\\", "[", "]", "(", ")", "{", "}", ".", "*", "+", "?", "^", "$", "|"}
+
+	// 定义正则表达式模式和替换字符串
+	var pattern = `%s`
+	var replaceStr = `([0-9A-Za-z]{4})`
+
+	// 对可能引起正则表达式冲突的字符进行转义
+	for _, esc := range escapeSet {
+		input = strings.ReplaceAll(input, esc, "\\"+esc)
+	}
+
+	// 将子匹配内容替换为指定格式的正则表达式
+	replaced := strings.ReplaceAll(input, pattern, replaceStr)
+
+	// 返回转换结果和是否大写的标记
+	return replaced
+}
+
+// \[emoji=[0-9A-Z]{4}\]
+func ReplaceToEmojis(str string, pattern string) string {
+	format := pattern
+	pattern = ConvertRegex(pattern)
+	str = regexp.MustCompile("[0-9]"+pattern+pattern).ReplaceAllStringFunc(str, func(s string) string {
+		if !strings.EqualFold(strings.ToUpper(s[1:]), strings.ToUpper(fmt.Sprintf(format+format, "FE0F", "20E3"))) { //"[emoji=FE0F][emoji=20E3]"
+			return s
+		}
+		return fmt.Sprintf(format, fmt.Sprintf("%04X", s[0])) + s[1:]
+	})
+	res := str
+	h := func(ssr *[]string) string {
+		chs := []string{}
+		ch := ""
+		l := 100
+		lft := []string{}
+	again:
+		for j := range emojiMap {
+			ok, left := matchPrefix(emojiMap[j].CodePoint2, *ssr)
+			if ok {
+				if ln := len(left); ln < l {
+					ch = emojiMap[j].Character
+					l = ln
+					lft = left
+				}
+			}
+		}
+		if ch != "" {
+			chs = append(chs, ch)
+		}
+		if len(lft) != 0 && ch != "" {
+			ch = ""
+			l = 100
+			*ssr = lft
+			lft = []string{}
+			goto again
+		}
+		return strings.Join(chs, "")
+	}
+	sss := regexp.MustCompile(pattern).FindAllStringSubmatchIndex(str, -1)
+	ssr := []string{}
+	ssrs := []string{}
+	e := -1
+	for i := range sss {
+		ss := sss[i]
+		a, b, c, d := ss[0], ss[1], ss[2], ss[3]
+
+		//结算
+		if e != a && len(ssr) != 0 {
+			rs := h(&ssr)
+			res = strings.Replace(res, strings.Join(ssrs, ""), rs, 1)
+			ssr = []string{}
+			ssrs = []string{}
+			e = -1
+		}
+
+		ssrs = append(ssrs, str[a:b])
+		// if upper {
+		// 	ssr = append(ssr, str[c:d])
+		// } else {
+		ssr = append(ssr, strings.ToUpper(str[c:d]))
+		// }
+		e = b
+	}
+	if len(ssr) != 0 {
+		rs := h(&ssr)
+		res = strings.Replace(res, strings.Join(ssrs, ""), rs, 1)
+		e = -1
+	}
+	return res
+}
+
+func transfer(str string) []string {
+	// 将输入字符串解析为十六进制数
+	unicodeValue, err := strconv.ParseUint(str, 16, 32)
+	if err != nil {
+		panic(err)
+	}
+	// 将 Unicode 编码值转换为 UTF-16 编码单元
+	utf16Units := utf16.Encode([]rune{rune(unicodeValue)})
+
+	// 将 UTF-16 编码单元格式化为字符串，并返回字符串切片
+	return []string{fmt.Sprintf("%04X", utf16Units[0]), fmt.Sprintf("%04X", utf16Units[1])}
+}
+
+func matchPrefix(s1 []string, s2 []string) (bool, []string) {
+	if len(s1) > len(s2) {
+		return false, s2
+	}
+	for i := range s1 {
+		if s1[i] != s2[i] {
+			return false, nil
+		}
+	}
+	return true, s2[len(s1):]
+}
+
+func RuneToHx(a rune) string {
+	return fmt.Sprintf("%04X", a)
+}
+
+func HexToRune(hexStr string) (rune, error) {
+	// 解析十六进制字符串，获取对应的无符号整数
+	u, err := strconv.ParseUint(hexStr, 16, 32)
+	if err != nil {
+		return 0, err
+	}
+
+	// 将无符号整数转换为 rune 类型的 Unicode 编码值
+	r := rune(u)
+
+	// 返回转换结果
+	return r, nil
+}
@@ -0,0 +1,170 @@
+package emoji
+
+import (
+	"bytes"
+	"errors"
+	"strings"
+	"unicode"
+
+	"github.com/rivo/uniseg"
+)
+
+// errors
+var (
+	ErrStrNotEmoji = errors.New("the string is not emoji")
+)
+
+// Emoji is an entity that represents comprehensive emoji info.
+type Emoji struct {
+	Slug        string   `json:"slug"`
+	Character   string   `json:"character"`
+	UnicodeName string   `json:"unicode_name"`
+	CodePoint   string   `json:"code_point"`
+	CodePoint2  []string `json:"code_points"`
+	Group       string   `json:"group"`
+	SubGroup    string   `json:"sub_group"`
+}
+
+// ContainsEmoji checks whether given string contains emoji or not. It uses local emoji list as provider.
+func ContainsEmoji(s string) bool {
+	for _, r := range s {
+		if _, ok := emojiMap[string(r)]; ok {
+			return true
+		}
+	}
+
+	gr := uniseg.NewGraphemes(s)
+	for gr.Next() {
+		if _, ok := emojiMap[gr.Str()]; ok {
+			return true
+		}
+	}
+
+	return false
+}
+
+// AllEmojis gets all emojis from provider.
+func AllEmojis() []Emoji {
+	return emojiMapToSlice(emojiMap)
+}
+
+// RemoveEmojis removes all emojis from the s string and returns a new string.
+func RemoveEmojis(s string) string {
+	return ReplaceEmojisWithFunc(s, nil)
+}
+
+// ReplaceEmojisWith replaces all emojis from the s string with the specified rune and returns a new string.
+func ReplaceEmojisWith(s string, c rune) string {
+	replacerStr := string(c)
+	return ReplaceEmojisWithFunc(s, func(em Emoji) string {
+		return replacerStr
+	})
+}
+
+// ReplaceEmojisWithSlug replaces all emojis from the s string with the emoji's slug and returns a new string.
+func ReplaceEmojisWithSlug(s string) string {
+	return ReplaceEmojisWithFunc(s, func(em Emoji) string {
+		return em.Slug
+	})
+}
+
+type replacerFn func(e Emoji) string
+
+// ReplaceEmojisWithFunc replaces all emojis from the s string with the result of the replacerFn function and returns a new string.
+func ReplaceEmojisWithFunc(s string, replacer replacerFn) string {
+	cleanBuf := bytes.Buffer{}
+
+	gr := uniseg.NewGraphemes(s)
+	for gr.Next() {
+		em, ok := emojiMap[gr.Str()]
+		if !ok {
+			cleanBuf.Write(gr.Bytes())
+			continue
+		}
+
+		if replacer != nil {
+			cleanBuf.WriteString(replacer(em))
+		}
+	}
+
+	res := cleanBuf
+	res.Reset()
+	for _, r := range cleanBuf.String() {
+		em, ok := emojiMap[string(r)]
+		if !ok {
+			res.WriteRune(r)
+			continue
+		}
+
+		if replacer != nil {
+			res.WriteString(replacer(em))
+		}
+	}
+
+	return strings.TrimFunc(res.String(), func(r rune) bool {
+		return unicode.IsSpace(r) || !unicode.IsGraphic(r) || !unicode.IsPrint(r) || unicode.In(r, unicode.Variation_Selector)
+	})
+}
+
+// GetInfo returns a gomoji.Emoji model representation of provided emoji.
+// If the emoji was not found, it returns the gomoji.ErrStrNotEmoji error
+func GetInfo(emoji string) (Emoji, error) {
+	em, ok := emojiMap[emoji]
+	if !ok {
+		return Emoji{}, ErrStrNotEmoji
+	}
+
+	return em, nil
+}
+
+// CollectAll finds all emojis in given string. Unlike FindAll, this does not
+// distinct repeating occurrences of emoji. If there are no emojis it returns a nil-slice.
+func CollectAll(s string) []Emoji {
+	var emojis []Emoji
+
+	gr := uniseg.NewGraphemes(s)
+	for gr.Next() {
+		if em, ok := emojiMap[gr.Str()]; ok {
+			emojis = append(emojis, em)
+			continue
+		}
+
+		start, end := gr.Positions()
+		for _, r := range s[start:end] {
+			if em, ok := emojiMap[string(r)]; ok {
+				emojis = append(emojis, em)
+			}
+		}
+	}
+
+	return emojis
+}
+
+// FindAll finds all emojis in given string. If there are no emojis it returns a nil-slice.
+func FindAll(s string) []Emoji {
+	emojis := make(map[string]Emoji)
+
+	gr := uniseg.NewGraphemes(s)
+	for gr.Next() {
+		if em, ok := emojiMap[gr.Str()]; ok {
+			emojis[gr.Str()] = em
+		}
+	}
+
+	for _, r := range s {
+		if em, ok := emojiMap[string(r)]; ok {
+			emojis[string(r)] = em
+		}
+	}
+
+	return emojiMapToSlice(emojis)
+}
+
+func emojiMapToSlice(em map[string]Emoji) []Emoji {
+	var emojis []Emoji
+	for _, emoji := range em {
+		emojis = append(emojis, emoji)
+	}
+
+	return emojis
+}
@@ -14,6 +14,7 @@ require (
 	github.com/dop251/goja v0.0.0-20230427124612-428fc442ff5f
 	github.com/dop251/goja_nodejs v0.0.0-20230322100729-2550c7b6c124
 	github.com/elastic/go-elasticsearch/v6 v6.8.5
+	github.com/forPelevin/gomoji v1.1.8
 	github.com/gin-gonic/gin v1.8.1
 	github.com/go-redis/redis/v8 v8.11.5
 	github.com/goccy/go-json v0.9.7
@@ -23,7 +24,6 @@ require (
 	github.com/robfig/cron/v3 v3.0.1
 	github.com/stretchr/testify v1.8.0
 	golang.org/x/net v0.4.0
-	gopkg.in/yaml.v2 v2.4.0
 )

 require (
@@ -31,6 +31,8 @@ require (
 	github.com/google/pprof v0.0.0-20230510103437-eeec1cb781c3 // indirect
 	github.com/pelletier/go-toml/v2 v2.0.1 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
+	github.com/rivo/uniseg v0.4.3 // indirect
+	gopkg.in/yaml.v2 v2.4.0 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 )

@@ -62,6 +62,8 @@ github.com/elastic/go-elasticsearch/v6 v6.8.5/go.mod h1:UwaDJsD3rWLM5rKNFzv9hgox
 github.com/elazarl/go-bindata-assetfs v1.0.0/go.mod h1:v+YaWX3bdea5J/mo8dSETolEo7R71Vk1u8bnjau5yw4=
 github.com/envoyproxy/go-control-plane v0.9.1-0.20191026205805-5f8ba28d4473/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
 github.com/envoyproxy/protoc-gen-validate v0.1.0/go.mod h1:iSmxcyjqTsJpI2R4NaDN7+kN2VEUnK/pcBlmesArF7c=
+github.com/forPelevin/gomoji v1.1.8 h1:JElzDdt0TyiUlecy6PfITDL6eGvIaxqYH1V52zrd0qQ=
+github.com/forPelevin/gomoji v1.1.8/go.mod h1:8+Z3KNGkdslmeGZBC3tCrwMrcPy5GRzAD+gL9NAwMXg=
 github.com/fsnotify/fsnotify v1.4.7/go.mod h1:jwhsz4b93w/PPRr/qN1Yymfu8t87LnFCMoQvtojpjFo=
 github.com/fsnotify/fsnotify v1.4.9 h1:hsms1Qyu0jgnwNXIxa+/V/PDsU6CfLf6CNO8H7IWoS4=
 github.com/gin-contrib/gzip v0.0.6 h1:NjcunTcGAj5CO1gn4N8jHOSIeRFHIbn51z6K+xaN4d4=
@@ -199,6 +201,8 @@ github.com/prometheus/common v0.10.0/go.mod h1:Tlit/dnDKsSWFlCLTWaA1cyBgKHSMdTB8
 github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
 github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
 github.com/prometheus/procfs v0.1.3/go.mod h1:lV6e/gmhEcM9IjHGsFOCxxuZ+z1YqCvr4OA4YeYWdaU=
+github.com/rivo/uniseg v0.4.3 h1:utMvzDsuh3suAEnhH0RdHmoPbU648o6CvXxTx4SBMOw=
+github.com/rivo/uniseg v0.4.3/go.mod h1:FN3SvrM+Zdj16jyLfmOkMNblXMcoc8DfTHruCPUcx88=
 github.com/robfig/cron/v3 v3.0.1 h1:WdRxkvbJztn8LMz/QEvLN5sBU+xKpSqwwUO1Pjr4qDs=
 github.com/robfig/cron/v3 v3.0.1/go.mod h1:eQICP3HwyT7UooqI/z+Ov+PtYAWygg1TEWWzGIFLtro=
 github.com/rogpeppe/go-internal v1.3.0/go.mod h1:M8bDsm7K2OlrFYOpmOWEs/qY81heoFRclV5y23lUDJ4=