BlueGit

timfa bluegit

Files

bluegit / internal / i18n / cuneiform.go

cuneiform.go

feat: nieuw-formulier, groepsoverzicht, logo's en talen
timfa project_87923614_bot_3aff574f6f98414b945a8799374f9092 · pushed by timfa · 4 weeks ago
Failed d1203d3b History
cuneiform.go 5.7 KiB Download Raw
1package i18n
2
3import "strings"
4
5// Transliterate writes English phonetically in Sumero-Akkadian
6// cuneiform (Unicode U+12000). Digits and punctuation stay, so a
7// sentence keeps its shape. The spelling is syllabic: consonants
8// without a following vowel take an inherent a, o is written u,
9// and there is no separate sign for f, v or o.
10func Transliterate(s string) string {
11 var b strings.Builder
12 word := strings.Builder{}
13 flush := func() {
14 if word.Len() == 0 {
15 return
16 }
17 b.WriteString(encodeWord(word.String()))
18 word.Reset()
19 }
20 for _, r := range s {
21 if isLetter(r) || r == '\'' {
22 word.WriteRune(r)
23 continue
24 }
25 flush()
26 b.WriteRune(r)
27 }
28 flush()
29 return b.String()
30}
31
32func isLetter(r rune) bool {
33 return (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z')
34}
35
36func encodeWord(raw string) string {
37 ph := phonemes(raw)
38 if len(ph) == 0 {
39 return ""
40 }
41 var out []rune
42 i := 0
43 for i < len(ph) {
44 if i+2 < len(ph) && isC(ph[i]) && isV(ph[i+1]) && isC(ph[i+2]) && (i+3 == len(ph) || isC(ph[i+3])) {
45 if sign, ok := signs[ph[i]+ph[i+1]+ph[i+2]]; ok {
46 out = append(out, sign)
47 i += 3
48 continue
49 }
50 }
51 if i+1 < len(ph) && isC(ph[i]) && isV(ph[i+1]) {
52 if sign, ok := cv(ph[i], ph[i+1]); ok {
53 out = append(out, sign)
54 i += 2
55 continue
56 }
57 }
58 if i+1 < len(ph) && isV(ph[i]) && isC(ph[i+1]) && (i+2 == len(ph) || isC(ph[i+2])) {
59 if sign, ok := signs[ph[i]+ph[i+1]]; ok {
60 out = append(out, sign)
61 i += 2
62 continue
63 }
64 }
65 if isV(ph[i]) {
66 if sign, ok := signs[ph[i]]; ok {
67 out = append(out, sign)
68 }
69 i++
70 continue
71 }
72 if sign, ok := cv(ph[i], "a"); ok {
73 out = append(out, sign)
74 }
75 i++
76 }
77 return string(out)
78}
79
80func cv(c, v string) (rune, bool) {
81 if v == "e" {
82 if sign, ok := signs[c+"e"]; ok {
83 return sign, true
84 }
85 if sign, ok := signs[c+"i"]; ok {
86 return sign, true
87 }
88 }
89 sign, ok := signs[c+v]
90 return sign, ok
91}
92
93func isC(s string) bool { return s != "a" && s != "e" && s != "i" && s != "u" }
94func isV(s string) bool { return !isC(s) }
95
96func phonemes(raw string) []string {
97 s := strings.ToLower(raw)
98 var out []string
99 for i := 0; i < len(s); i++ {
100 if i+1 < len(s) {
101 switch s[i : i+2] {
102 case "qu":
103 out = append(out, "k", "w")
104 i++
105 continue
106 case "ck":
107 out = append(out, "k")
108 i++
109 continue
110 case "ph":
111 out = append(out, "p")
112 i++
113 continue
114 case "th":
115 out = append(out, "t")
116 i++
117 continue
118 case "sh":
119 out = append(out, "š")
120 i++
121 continue
122 case "ch":
123 out = append(out, "š")
124 i++
125 continue
126 case "wh":
127 out = append(out, "w")
128 i++
129 continue
130 case "ng":
131 out = append(out, "n")
132 i++
133 continue
134 case "ee", "ea":
135 out = append(out, "i")
136 i++
137 continue
138 case "oo", "ou":
139 out = append(out, "u")
140 i++
141 continue
142 case "ai", "ay":
143 out = append(out, "e")
144 i++
145 continue
146 }
147 }
148 c := s[i]
149 switch c {
150 case 'a', 'e', 'i', 'u':
151 out = append(out, string(c))
152 case 'o':
153 out = append(out, "u")
154 case 'y':
155 if i == 0 || isVowelByte(s[i-1]) {
156 out = append(out, "y")
157 } else {
158 out = append(out, "i")
159 }
160 case 'c':
161 if i+1 < len(s) && (s[i+1] == 'e' || s[i+1] == 'i' || s[i+1] == 'y') {
162 out = append(out, "s")
163 } else {
164 out = append(out, "k")
165 }
166 case 'f':
167 out = append(out, "p")
168 case 'v':
169 out = append(out, "b")
170 case 'j':
171 out = append(out, "y")
172 case 'q':
173 out = append(out, "k")
174 case 'x':
175 out = append(out, "k", "s")
176 case 'b', 'd', 'g', 'h', 'k', 'l', 'm', 'n', 'p', 'r', 's', 't', 'w', 'z':
177 out = append(out, string(c))
178 }
179 }
180 compact := out[:0]
181 for _, p := range out {
182 if len(compact) > 0 && compact[len(compact)-1] == p && isC(p) {
183 continue
184 }
185 compact = append(compact, p)
186 }
187 return compact
188}
189
190func isVowelByte(c byte) bool {
191 return c == 'a' || c == 'e' || c == 'i' || c == 'o' || c == 'u' || c == 'y'
192}
193
194// signs maps a conventional Akkadian reading to one sign.
195// pu is written with BU, and w+vowel with PI, which are the usual values.
196var signs = map[string]rune{
197 "a": '𒀀', "e": '𒂊', "i": '𒄿', "u": '𒌋',
198 "ba": '𒁀', "bi": '𒁉', "bu": '𒁍', "pu": '𒁍',
199 "da": '𒁕', "di": '𒁲', "du": '𒁺',
200 "ga": '𒂵', "gi": '𒄀', "gu": '𒄖',
201 "ha": '𒄩', "hi": '𒄭', "hu": '𒄷',
202 "ka": '𒅗', "ki": '𒆠', "ku": '𒆪',
203 "la": '𒆷', "li": '𒇷', "lu": '𒇻',
204 "ma": '𒈠', "me": '𒈨', "mi": '𒈪', "mu": '𒈬',
205 "na": '𒈾', "ne": '𒉈', "ni": '𒉌', "nu": '𒉡',
206 "pa": '𒉺', "pi": '𒉿',
207 "ra": '𒊏', "ri": '𒊑', "ru": '𒊒',
208 "sa": '𒊓', "si": '𒋛', "su": '𒋢',
209 "ša": '𒊭', "še": '𒊺', "ši": '𒅆', "šu": '𒋗',
210 "ta": '𒋫', "te": '𒋼', "ti": '𒋾', "tu": '𒌅',
211 "wa": '𒉿', "we": '𒉿', "wi": '𒉿', "wu": '𒉿',
212 "ya": '𒅀', "za": '𒍝', "zi": '𒍣', "zu": '𒍪',
213 "ab": '𒀊', "ad": '𒀜', "ak": '𒀝', "al": '𒀠', "an": '𒀭', "aš": '𒀸',
214 "el": '𒂖', "en": '𒂗', "eš": '𒂠',
215 "ib": '𒅁', "ig": '𒅅', "il": '𒅋', "im": '𒅎', "in": '𒅔', "ir": '𒅕', "iš": '𒅖',
216 "ub": '𒌒', "ud": '𒌓', "um": '𒌝', "un": '𒌦', "ur": '𒌨', "uš": '𒍑',
217 "bad": '𒁁', "bal": '𒁄', "bar": '𒁇',
218 "dag": '𒁖', "dam": '𒁮', "dar": '𒁯',
219 "dib": '𒁳', "dim": '𒁴', "din": '𒁷',
220 "dub": '𒁾', "dug": '𒂁', "dun": '𒂄',
221 "gal": '𒃲', "gam": '𒃵', "gan": '𒃶', "gar": '𒃻',
222 "gul": '𒄢', "gur": '𒄥',
223 "hal": '𒄬',
224 "kal": '𒆗', "kin": '𒆥', "kul": '𒆰', "kur": '𒆳',
225 "lal": '𒇲', "lam": '𒇴', "lum": '𒈝',
226 "mar": '𒈥',
227 "nam": '𒉆', "nun": '𒉣',
228 "pad": '𒉻', "pap": '𒉽',
229 "rab": '𒊐',
230 "sag": '𒊕', "sar": '𒊬',
231 "tab": '𒋰', "tag": '𒋳', "tar": '𒋻',
232 "tum": '𒌈', "tur": '𒌉',
233 "zag": '𒍠',
234}