Spaces:
Sleeping
Sleeping
File size: 8,424 Bytes
47bc13b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 |
"""Hindi transliteration and script handling for multilingual addresses."""
import re
class HindiTransliterator:
"""
Handles Hindi (Devanagari) to Latin transliteration and script detection.
Supports:
- Devanagari to Latin conversion
- Common Hindi address terms
- Mixed script (code-switched) addresses
"""
# Devanagari Unicode range
DEVANAGARI_START = 0x0900
DEVANAGARI_END = 0x097F
# Common Hindi address terms with transliterations
HINDI_TERMS = {
# Devanagari -> Latin
'गली': 'GALI',
'गलि': 'GALI',
'मोहल्ला': 'MOHALLA',
'नगर': 'NAGAR',
'विहार': 'VIHAR',
'पुरी': 'PURI',
'पुर': 'PUR',
'बाग': 'BAGH',
'मार्ग': 'MARG',
'रोड': 'ROAD',
'मंजिल': 'FLOOR',
'पहली': 'FIRST',
'दूसरी': 'SECOND',
'तीसरी': 'THIRD',
'चौथी': 'FOURTH',
'भूतल': 'GROUND FLOOR',
'तहखाना': 'BASEMENT',
'मकान': 'HOUSE',
'प्लॉट': 'PLOT',
'खसरा': 'KHASRA',
'ब्लॉक': 'BLOCK',
'सेक्टर': 'SECTOR',
'कॉलोनी': 'COLONY',
'इलाका': 'AREA',
'क्षेत्र': 'AREA',
'दिल्ली': 'DELHI',
'नई दिल्ली': 'NEW DELHI',
'नम्बर': 'NUMBER',
'नंबर': 'NUMBER',
'संख्या': 'NUMBER',
'पास': 'NEAR',
'सामने': 'OPPOSITE',
'पीछे': 'BEHIND',
'के पास': 'NEAR',
'के सामने': 'OPPOSITE',
'चौक': 'CHOWK',
'बाजार': 'BAZAAR',
'बस्ती': 'BASTI',
'पार्क': 'PARK',
'एक्सटेंशन': 'EXTENSION',
'फेज': 'PHASE',
'वार्ड': 'WARD',
'जोन': 'ZONE',
}
# Devanagari consonants to Latin (basic ITRANS-like mapping)
CONSONANT_MAP = {
'क': 'k', 'ख': 'kh', 'ग': 'g', 'घ': 'gh', 'ङ': 'ng',
'च': 'ch', 'छ': 'chh', 'ज': 'j', 'झ': 'jh', 'ञ': 'ny',
'ट': 't', 'ठ': 'th', 'ड': 'd', 'ढ': 'dh', 'ण': 'n',
'त': 't', 'थ': 'th', 'द': 'd', 'ध': 'dh', 'न': 'n',
'प': 'p', 'फ': 'ph', 'ब': 'b', 'भ': 'bh', 'म': 'm',
'य': 'y', 'र': 'r', 'ल': 'l', 'व': 'v', 'श': 'sh',
'ष': 'sh', 'स': 's', 'ह': 'h',
'क़': 'q', 'ख़': 'kh', 'ग़': 'g', 'ज़': 'z', 'ड़': 'd',
'ढ़': 'dh', 'फ़': 'f', 'य़': 'y',
}
# Devanagari vowels/matras
VOWEL_MAP = {
'अ': 'a', 'आ': 'aa', 'इ': 'i', 'ई': 'ee', 'उ': 'u', 'ऊ': 'oo',
'ए': 'e', 'ऐ': 'ai', 'ओ': 'o', 'औ': 'au', 'अं': 'an', 'अः': 'ah',
'ा': 'a', 'ि': 'i', 'ी': 'ee', 'ु': 'u', 'ू': 'oo',
'े': 'e', 'ै': 'ai', 'ो': 'o', 'ौ': 'au',
'ं': 'n', 'ः': 'h', '्': '', # Halant (vowel killer)
'ँ': 'n', # Chandrabindu
}
# Devanagari digits
DIGIT_MAP = {
'०': '0', '१': '1', '२': '2', '३': '3', '४': '4',
'५': '5', '६': '6', '७': '7', '८': '8', '९': '9',
}
def __init__(self, use_known_terms: bool = True):
"""
Initialize transliterator.
Args:
use_known_terms: Use dictionary of known Hindi address terms
"""
self.use_known_terms = use_known_terms
def contains_devanagari(self, text: str) -> bool:
"""Check if text contains Devanagari script."""
for char in text:
code = ord(char)
if self.DEVANAGARI_START <= code <= self.DEVANAGARI_END:
return True
return False
def get_script_ratio(self, text: str) -> dict[str, float]:
"""
Get ratio of different scripts in text.
Returns dict with 'latin', 'devanagari', 'numeric', 'other' ratios.
"""
if not text:
return {'latin': 0.0, 'devanagari': 0.0, 'numeric': 0.0, 'other': 0.0}
counts: dict[str, float] = {'latin': 0, 'devanagari': 0, 'numeric': 0, 'other': 0}
total = 0
for char in text:
if char.isspace():
continue
total += 1
code = ord(char)
if self.DEVANAGARI_START <= code <= self.DEVANAGARI_END:
counts['devanagari'] += 1
elif char.isascii() and char.isalpha():
counts['latin'] += 1
elif char.isdigit():
counts['numeric'] += 1
else:
counts['other'] += 1
if total == 0:
return counts
return {k: v / total for k, v in counts.items()}
def transliterate(self, text: str) -> str:
"""
Transliterate Devanagari text to Latin script.
Args:
text: Input text (may be mixed script)
Returns:
Transliterated text in Latin script
"""
if not self.contains_devanagari(text):
return text
# First, try to match known terms
if self.use_known_terms:
for hindi, latin in sorted(self.HINDI_TERMS.items(), key=lambda x: -len(x[0])):
text = text.replace(hindi, f' {latin} ')
# Then transliterate remaining Devanagari
result = []
i = 0
while i < len(text):
char = text[i]
code = ord(char)
if self.DEVANAGARI_START <= code <= self.DEVANAGARI_END:
# Check digits first
if char in self.DIGIT_MAP:
result.append(self.DIGIT_MAP[char])
# Check vowels
elif char in self.VOWEL_MAP:
result.append(self.VOWEL_MAP[char])
# Check consonants
elif char in self.CONSONANT_MAP:
result.append(self.CONSONANT_MAP[char])
# Add implicit 'a' unless followed by matra or halant
if i + 1 < len(text):
next_char = text[i + 1]
next_code = ord(next_char)
# If next is a matra (0x093E-0x094D) or halant, don't add 'a'
if not (0x093E <= next_code <= 0x094D):
result.append('a')
else:
result.append('a')
else:
# Unknown Devanagari character
result.append(char)
else:
result.append(char)
i += 1
# Clean up
output = ''.join(result)
output = re.sub(r'\s+', ' ', output)
return output.strip().upper()
def normalize_mixed_script(self, text: str) -> str:
"""
Handle code-mixed (Hindi + English) addresses.
Transliterates Hindi portions while preserving English.
"""
# Split on whitespace to handle word by word
words = text.split()
result = []
for word in words:
if self.contains_devanagari(word):
# Check if it's a known term first
if self.use_known_terms and word in self.HINDI_TERMS:
result.append(self.HINDI_TERMS[word])
else:
result.append(self.transliterate(word))
else:
result.append(word.upper())
return ' '.join(result)
def detect_language(text: str) -> str:
"""
Simple language detection for address text.
Returns: 'hindi', 'english', or 'mixed'
"""
transliterator = HindiTransliterator()
ratios = transliterator.get_script_ratio(text)
if ratios['devanagari'] > 0.5:
return 'hindi'
elif ratios['latin'] > 0.5:
return 'english'
elif ratios['devanagari'] > 0 and ratios['latin'] > 0:
return 'mixed'
else:
return 'english'
|