Update idn_phonemes.py

idn_phonemes.py

@@ -1,155 +1,244 @@
 import re
 
+# --- IPA map ---
 ipa_map = {
-    "ng": "ŋ",
-    "ny": "ɲ",
-    "sy": "ʃ",
-    "kh": "x",
-    "c": "tʃ",
-    "j": "dʒ",
-    "y": "j",
-    "r": "r",
-    "x": "ks",
-    "a": "a",
-    "i": "i",
-    "u": "u",
-    "e": "ə",
-    "o": "o",
-    "b": "b",
-    "d": "d̪",
-    "t": "t̪",
-    "g": "ɡ",
-    "k": "k",
-    "p": "p",
-    "m": "m",
-    "n": "n",
-    "l": "l",
-    "s": "s",
-    "h": "h",
-    "w": "w",
+    "ng": "ŋ", "ny": "ɲ", "sy": "ʃ", "kh": "x", "c": "tʃ", "j": "dʒ",
+    "b": "b", "d": "d̪", "t": "t̪", "g": "ɡ", "k": "k", "p": "p",
+    "m": "m", "n": "n", "l": "l", "s": "s", "h": "h", "r": "r", "w": "w", "y": "j",
+    "a": "a", "i": "i", "u": "u", "o": "o", "e": "ə"
 }
 
-num_words = {
-    0: "nol",
-    1: "satu",
-    2: "dua",
-    3: "tiga",
-    4: "empat",
-    5: "lima",
-    6: "enam",
-    7: "tujuh",
-    8: "delapan",
-    9: "sembilan",
-    10: "sepuluh",
-    11: "sebelas"
+# Sebutan huruf
+letter_words = {
+    "a":"a","b":"be","c":"ce","d":"de","e":"e","f":"ef","g":"ge","h":"ha",
+    "i":"i","j":"je","k":"ka","l":"el","m":"em","n":"en","o":"o","p":"pe",
+    "q":"ki","r":"er","s":"es","t":"te","u":"u","v":"fe","w":"we","x":"eks",
+    "y":"ye","z":"zet"
+}
+
+digit_words = {
+    "0":"nol","1":"satu","2":"dua","3":"tiga","4":"empat",
+    "5":"lima","6":"enam","7":"tujuh","8":"delapan","9":"sembilan"
 }
 
+# --- Number to words (hingga triliun) ---
 def number_to_words(n: int) -> str:
-    """Konversi angka 0–9999 ke kata dalam bahasa Indonesia"""
-    if n < 12:
-        return num_words[n]
-    elif n < 20:
-        return number_to_words(n-10) + " belas"
-    elif n < 100:
-        puluhan, sisa = divmod(n, 10)
-        result = number_to_words(puluhan) + " puluh"
-        if sisa:
-            result += " " + number_to_words(sisa)
-        return result
-    elif n < 200:
-        return "seratus" + (" " + number_to_words(n-100) if n > 100 else "")
-    elif n < 1000:
-        ratusan, sisa = divmod(n, 100)
-        result = number_to_words(ratusan) + " ratus"
-        if sisa:
-            result += " " + number_to_words(sisa)
-        return result
-    elif n < 2000:
-        return "seribu" + (" " + number_to_words(n-1000) if n > 1000 else "")
-    elif n < 10000:
-        ribuan, sisa = divmod(n, 1000)
-        result = number_to_words(ribuan) + " ribu"
-        if sisa:
-            result += " " + number_to_words(sisa)
-        return result
-    else:
-        return str(n)  # fallback
-
-
-
-def expand_abbreviation(word: str) -> str:
-    """Ubah singkatan (huruf kapital) jadi ucapan Indonesia"""
-    if word.isupper() and len(word) > 1:  # contoh: KTP, DPR, RI
-        return " ".join(letter_words.get(ch.lower(), ch) for ch in word)
-    return word
+    n = int(n)
+    if n == 0:
+        return "nol"
+    def _below_thousand(x):
+        words = ["nol","satu","dua","tiga","empat","lima","enam","tujuh","delapan","sembilan","sepuluh","sebelas"]
+        if x < 12: return words[x]
+        if x < 20: return _below_thousand(x-10)+" belas"
+        if x < 100:
+            q,r=divmod(x,10); return _below_thousand(q)+" puluh"+((" "+_below_thousand(r)) if r else "")
+        if x < 200: return "seratus"+((" "+_below_thousand(x-100)) if x>100 else "")
+        if x < 1000:
+            q,r=divmod(x,100); return _below_thousand(q)+" ratus"+((" "+_below_thousand(r)) if r else "")
+    scales=[(1_000_000_000_000,"triliun"),(1_000_000_000,"miliar"),(1_000_000,"juta"),(1000,"ribu")]
+    parts=[]; remaining=n
+    for v,nm in scales:
+        if remaining>=v:
+            q,remaining=divmod(remaining,v)
+            if v==1000 and q==1: parts.append("seribu")
+            else: parts.append(number_to_words(q)+" "+nm)
+    if remaining: parts.append(_below_thousand(remaining))
+    return " ".join(parts)
 
+# --- Nomor HP ---
+phone_pattern=re.compile(r'(?<!\w)(?:\+62|\d)\d{7,}(?!\w)')
+def expand_phones(text:str)->str:
+    def repl(m):
+        digits=re.findall(r'\d',m.group(0))
+        return " ".join(digit_words[d] for d in digits)
+    return phone_pattern.sub(repl,text)
 
-letter_words = {
-    "a": "a",
-    "b": "be",
-    "c": "ce",
-    "d": "de",
-    "e": "e",
-    "f": "ef",
-    "g": "ge",
-    "h": "ha",
-    "i": "i",
-    "j": "je",
-    "k": "ka",
-    "l": "el",
-    "m": "em",
-    "n": "en",
-    "o": "o",
-    "p": "pe",
-    "q": "ki",
-    "r": "er",
-    "s": "es",
-    "t": "te",
-    "u": "u",
-    "v": "fe",
-    "w": "we",
-    "x": "eks",
-    "y": "ye",
-    "z": "zet",
-}
+# --- Angka umum ---
+def expand_numbers(text:str)->str:
+    def repl(m):
+        return number_to_words(int(m.group()))
+    return re.sub(r'\d+',repl,text)
 
+# --- Singkatan ---
+abbr_pattern=re.compile(r'(?<!\w)([A-Z]{2,})(?!\w)')
+def expand_abbreviations(text:str)->str:
+    def repl(m):
+        token=m.group(1)
+        if token=="HP":  # <-- jangan expand 'HP' kalau berdiri sendiri
+            return "ha pe"
+        return " ".join(letter_words[ch.lower()] for ch in token)
+    return abbr_pattern.sub(repl,text)
+
+# --- IPA ---
+def apply_ipa_map(text:str)->str:
+    t=text.lower()
+    for k in sorted(ipa_map,key=len,reverse=True):
+        t=re.sub(re.escape(k),ipa_map[k],t)
+    return "["+re.sub(r'\s+',' ',t).strip()+"]"
+
+# --- Pipeline ---
+def indo_to_ipa(text:str)->str:
+    # 1. nomor HP
+    step1=expand_phones(text)
+    # 2. angka biasa
+    step2=expand_numbers(step1)
+    # 3. singkatan
+    step3=expand_abbreviations(step2)
+    # 4. mapping IPA
+    return apply_ipa_map(step3)
+
+# import re
+
+# ipa_map = {
+#     "ng": "ŋ",
+#     "ny": "ɲ",
+#     "sy": "ʃ",
+#     "kh": "x",
+#     "c": "tʃ",
+#     "j": "dʒ",
+#     "y": "j",
+#     "r": "r",
+#     "x": "ks",
+#     "a": "a",
+#     "i": "i",
+#     "u": "u",
+#     "e": "ə",
+#     "o": "o",
+#     "b": "b",
+#     "d": "d̪",
+#     "t": "t̪",
+#     "g": "ɡ",
+#     "k": "k",
+#     "p": "p",
+#     "m": "m",
+#     "n": "n",
+#     "l": "l",
+#     "s": "s",
+#     "h": "h",
+#     "w": "w",
+# }
+
+# num_words = {
+#     0: "nol",
+#     1: "satu",
+#     2: "dua",
+#     3: "tiga",
+#     4: "empat",
+#     5: "lima",
+#     6: "enam",
+#     7: "tujuh",
+#     8: "delapan",
+#     9: "sembilan",
+#     10: "sepuluh",
+#     11: "sebelas"
+# }
+
+# def number_to_words(n: int) -> str:
+#     """Konversi angka 0–9999 ke kata dalam bahasa Indonesia"""
+#     if n < 12:
+#         return num_words[n]
+#     elif n < 20:
+#         return number_to_words(n-10) + " belas"
+#     elif n < 100:
+#         puluhan, sisa = divmod(n, 10)
+#         result = number_to_words(puluhan) + " puluh"
+#         if sisa:
+#             result += " " + number_to_words(sisa)
+#         return result
+#     elif n < 200:
+#         return "seratus" + (" " + number_to_words(n-100) if n > 100 else "")
+#     elif n < 1000:
+#         ratusan, sisa = divmod(n, 100)
+#         result = number_to_words(ratusan) + " ratus"
+#         if sisa:
+#             result += " " + number_to_words(sisa)
+#         return result
+#     elif n < 2000:
+#         return "seribu" + (" " + number_to_words(n-1000) if n > 1000 else "")
+#     elif n < 10000:
+#         ribuan, sisa = divmod(n, 1000)
+#         result = number_to_words(ribuan) + " ribu"
+#         if sisa:
+#             result += " " + number_to_words(sisa)
+#         return result
+#     else:
+#         return str(n)  # fallback
+
+
+
+# def expand_abbreviation(word: str) -> str:
+#     """Ubah singkatan (huruf kapital) jadi ucapan Indonesia"""
+#     if word.isupper() and len(word) > 1:  # contoh: KTP, DPR, RI
+#         return " ".join(letter_words.get(ch.lower(), ch) for ch in word)
+#     return word
+
+
+# letter_words = {
+#     "a": "a",
+#     "b": "be",
+#     "c": "ce",
+#     "d": "de",
+#     "e": "e",
+#     "f": "ef",
+#     "g": "ge",
+#     "h": "ha",
+#     "i": "i",
+#     "j": "je",
+#     "k": "ka",
+#     "l": "el",
+#     "m": "em",
+#     "n": "en",
+#     "o": "o",
+#     "p": "pe",
+#     "q": "ki",
+#     "r": "er",
+#     "s": "es",
+#     "t": "te",
+#     "u": "u",
+#     "v": "fe",
+#     "w": "we",
+#     "x": "eks",
+#     "y": "ye",
+#     "z": "zet",
+# }
 
-def indo_to_ipa(text: str) -> str:
-    text = text.lower()
-    
-    # Tangani singkatan (huruf kapital semua)
-    words = []
-    for w in text.split():
-        if w.isupper() and len(w) > 1:
-            words.append(expand_abbreviation(w))
-        else:
-            words.append(w)
-    text = " ".join(words)
-    
-    # Tangani angka → kata
-    def replace_number(match):
-        num = int(match.group())
-        return number_to_words(num)
-    text = re.sub(r"\d+", replace_number, text)
-    
-    # Konversi huruf → IPA
-    for k in sorted(ipa_map.keys(), key=lambda x: -len(x)):
-        text = re.sub(k, ipa_map[k], text)
-    
-    return text
 
 # def indo_to_ipa(text: str) -> str:
 #     text = text.lower()
     
-#     # Cari semua angka dalam teks dan ubah ke kata
+#     # Tangani singkatan (huruf kapital semua)
+#     words = []
+#     for w in text.split():
+#         if w.isupper() and len(w) > 1:
+#             words.append(expand_abbreviation(w))
+#         else:
+#             words.append(w)
+#     text = " ".join(words)
+    
+#     # Tangani angka → kata
 #     def replace_number(match):
 #         num = int(match.group())
 #         return number_to_words(num)
-    
 #     text = re.sub(r"\d+", replace_number, text)
     
 #     # Konversi huruf → IPA
 #     for k in sorted(ipa_map.keys(), key=lambda x: -len(x)):
 #         text = re.sub(k, ipa_map[k], text)
     
-#     return text
+#     return text
+
+# # def indo_to_ipa(text: str) -> str:
+# #     text = text.lower()
+    
+# #     # Cari semua angka dalam teks dan ubah ke kata
+# #     def replace_number(match):
+# #         num = int(match.group())
+# #         return number_to_words(num)
+    
+# #     text = re.sub(r"\d+", replace_number, text)
+    
+# #     # Konversi huruf → IPA
+# #     for k in sorted(ipa_map.keys(), key=lambda x: -len(x)):
+# #         text = re.sub(k, ipa_map[k], text)
+    
+# #     return text