nepali_unicoder 0.2.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- nepali_unicoder/__init__.py +3 -0
- nepali_unicoder/__main__.py +41 -0
- nepali_unicoder/convert.py +13 -0
- nepali_unicoder/data/preeti_rules.json +226 -0
- nepali_unicoder/data/roman_rules.json +112 -0
- nepali_unicoder/data/word_maps.json +67 -0
- nepali_unicoder/engine.py +107 -0
- nepali_unicoder/loader.py +120 -0
- nepali_unicoder/rules.py +2 -0
- nepali_unicoder/tokenizer.py +93 -0
- nepali_unicoder/trie.py +53 -0
- nepali_unicoder-0.2.0.dist-info/METADATA +210 -0
- nepali_unicoder-0.2.0.dist-info/RECORD +17 -0
- nepali_unicoder-0.2.0.dist-info/WHEEL +5 -0
- nepali_unicoder-0.2.0.dist-info/entry_points.txt +2 -0
- nepali_unicoder-0.2.0.dist-info/licenses/LICENSE +21 -0
- nepali_unicoder-0.2.0.dist-info/top_level.txt +1 -0
|
@@ -0,0 +1,41 @@
|
|
|
1
|
+
import argparse
|
|
2
|
+
import sys
|
|
3
|
+
|
|
4
|
+
from nepali_unicoder.convert import Converter
|
|
5
|
+
|
|
6
|
+
|
|
7
|
+
def main():
|
|
8
|
+
parser = argparse.ArgumentParser(
|
|
9
|
+
prog="python -m nepali_unicoder",
|
|
10
|
+
description="Convert Romanized Nepali or Preeti font text to Unicode Devanagari.",
|
|
11
|
+
)
|
|
12
|
+
parser.add_argument(
|
|
13
|
+
"text",
|
|
14
|
+
nargs="*",
|
|
15
|
+
help="The text to convert. If omitted, reads from stdin.",
|
|
16
|
+
)
|
|
17
|
+
parser.add_argument(
|
|
18
|
+
"--preeti",
|
|
19
|
+
action="store_true",
|
|
20
|
+
help="Enable Preeti to Unicode conversion mode.",
|
|
21
|
+
)
|
|
22
|
+
|
|
23
|
+
args = parser.parse_args()
|
|
24
|
+
|
|
25
|
+
# Determine input text
|
|
26
|
+
if args.text:
|
|
27
|
+
input_text = " ".join(args.text)
|
|
28
|
+
elif not sys.stdin.isatty():
|
|
29
|
+
input_text = sys.stdin.read().strip()
|
|
30
|
+
else:
|
|
31
|
+
parser.print_help()
|
|
32
|
+
return
|
|
33
|
+
|
|
34
|
+
mode = "preeti" if args.preeti else "roman"
|
|
35
|
+
converter = Converter(mode=mode)
|
|
36
|
+
result = converter.convert(input_text)
|
|
37
|
+
print(result)
|
|
38
|
+
|
|
39
|
+
|
|
40
|
+
if __name__ == "__main__":
|
|
41
|
+
main()
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
from nepali_unicoder.engine import Engine
|
|
2
|
+
|
|
3
|
+
|
|
4
|
+
class Converter:
|
|
5
|
+
"""
|
|
6
|
+
Wrapper around Engine for backward compatibility.
|
|
7
|
+
"""
|
|
8
|
+
|
|
9
|
+
def __init__(self, mode: str = "roman"):
|
|
10
|
+
self.engine = Engine(mode=mode)
|
|
11
|
+
|
|
12
|
+
def convert(self, text: str) -> str:
|
|
13
|
+
return self.engine.transliterate(text)
|
|
@@ -0,0 +1,226 @@
|
|
|
1
|
+
{
|
|
2
|
+
"mappings": {
|
|
3
|
+
"a": "ब",
|
|
4
|
+
"b": "द",
|
|
5
|
+
"c": "अ",
|
|
6
|
+
"d": "म",
|
|
7
|
+
"e": "भ",
|
|
8
|
+
"f": "ा",
|
|
9
|
+
"g": "न",
|
|
10
|
+
"h": "ज",
|
|
11
|
+
"i": "ष्",
|
|
12
|
+
"j": "व",
|
|
13
|
+
"k": "प",
|
|
14
|
+
"l": "ि",
|
|
15
|
+
"m": "m",
|
|
16
|
+
"n": "ल",
|
|
17
|
+
"o": "य",
|
|
18
|
+
"p": "उ",
|
|
19
|
+
"q": "त्र",
|
|
20
|
+
"r": "च",
|
|
21
|
+
"s": "क",
|
|
22
|
+
"t": "त",
|
|
23
|
+
"u": "ग",
|
|
24
|
+
"v": "ख",
|
|
25
|
+
"w": "ध",
|
|
26
|
+
"x": "ह",
|
|
27
|
+
"y": "थ",
|
|
28
|
+
"z": "श",
|
|
29
|
+
"A": "ब्",
|
|
30
|
+
"B": "द्य",
|
|
31
|
+
"C": "ऋ",
|
|
32
|
+
"D": "म्",
|
|
33
|
+
"E": "भ्",
|
|
34
|
+
"F": "ँ",
|
|
35
|
+
"G": "न्",
|
|
36
|
+
"H": "ज्",
|
|
37
|
+
"I": "क्ष्",
|
|
38
|
+
"J": "व्",
|
|
39
|
+
"K": "प्",
|
|
40
|
+
"L": "ी",
|
|
41
|
+
"M": "ः",
|
|
42
|
+
"N": "ल्",
|
|
43
|
+
"O": "इ",
|
|
44
|
+
"P": "ए",
|
|
45
|
+
"Q": "त्त",
|
|
46
|
+
"R": "च्",
|
|
47
|
+
"S": "क्",
|
|
48
|
+
"T": "त्",
|
|
49
|
+
"U": "ग्",
|
|
50
|
+
"V": "ख्",
|
|
51
|
+
"W": "ध्",
|
|
52
|
+
"X": "ह्",
|
|
53
|
+
"Y": "थ्",
|
|
54
|
+
"Z": "श्",
|
|
55
|
+
"0": "ण्",
|
|
56
|
+
"1": "ज्ञ",
|
|
57
|
+
"2": "द्द",
|
|
58
|
+
"3": "घ",
|
|
59
|
+
"4": "द्ध",
|
|
60
|
+
"5": "छ",
|
|
61
|
+
"6": "ट",
|
|
62
|
+
"7": "ठ",
|
|
63
|
+
"8": "ड",
|
|
64
|
+
"9": "ढ",
|
|
65
|
+
"!": "१",
|
|
66
|
+
"@": "२",
|
|
67
|
+
"#": "३",
|
|
68
|
+
"$": "४",
|
|
69
|
+
"%": "५",
|
|
70
|
+
"^": "६",
|
|
71
|
+
"&": "७",
|
|
72
|
+
"*": "८",
|
|
73
|
+
"(": "९",
|
|
74
|
+
")": "०",
|
|
75
|
+
"[": "ृ",
|
|
76
|
+
"]": "े",
|
|
77
|
+
"{": "{",
|
|
78
|
+
"}": "ै",
|
|
79
|
+
"|": "्र",
|
|
80
|
+
";": "स",
|
|
81
|
+
":": "स्",
|
|
82
|
+
"'": "ु",
|
|
83
|
+
"\"": "ू",
|
|
84
|
+
",": ",",
|
|
85
|
+
".": "।",
|
|
86
|
+
"/": "र",
|
|
87
|
+
"<": "?",
|
|
88
|
+
">": "श्र",
|
|
89
|
+
"?": "रु",
|
|
90
|
+
"\\": "्",
|
|
91
|
+
"`": "ञ",
|
|
92
|
+
"~": "ञ्",
|
|
93
|
+
"_": ")",
|
|
94
|
+
"-": "(",
|
|
95
|
+
"=": ".",
|
|
96
|
+
"...": "...",
|
|
97
|
+
"+": "ं",
|
|
98
|
+
"Ù": ";",
|
|
99
|
+
"Ú": ":",
|
|
100
|
+
"«": "्र",
|
|
101
|
+
"»": "्र",
|
|
102
|
+
"¿": "रू",
|
|
103
|
+
"å": "द्व",
|
|
104
|
+
"ˆ": "फ्",
|
|
105
|
+
"ª": "ङ",
|
|
106
|
+
"æ": "“",
|
|
107
|
+
"Æ": "”",
|
|
108
|
+
"¥": "र्",
|
|
109
|
+
"¶": "ठ्ठ",
|
|
110
|
+
"§": "ट्ट",
|
|
111
|
+
"£": "घ्",
|
|
112
|
+
"Ë": "ङ्ग",
|
|
113
|
+
"Í": "ङ्क",
|
|
114
|
+
"‰": "झ्"
|
|
115
|
+
},
|
|
116
|
+
"post_rules": [
|
|
117
|
+
[
|
|
118
|
+
"्ा",
|
|
119
|
+
""
|
|
120
|
+
],
|
|
121
|
+
[
|
|
122
|
+
"(त्र|त्त)([^उभप]+?)m",
|
|
123
|
+
"\\1m\\2"
|
|
124
|
+
],
|
|
125
|
+
[
|
|
126
|
+
"त्रm",
|
|
127
|
+
"क्र"
|
|
128
|
+
],
|
|
129
|
+
[
|
|
130
|
+
"त्तm",
|
|
131
|
+
"क्त"
|
|
132
|
+
],
|
|
133
|
+
[
|
|
134
|
+
"([^उभप]+?)m",
|
|
135
|
+
"m\\1"
|
|
136
|
+
],
|
|
137
|
+
[
|
|
138
|
+
"उm",
|
|
139
|
+
"ऊ"
|
|
140
|
+
],
|
|
141
|
+
[
|
|
142
|
+
"भm",
|
|
143
|
+
"झ"
|
|
144
|
+
],
|
|
145
|
+
[
|
|
146
|
+
"पm",
|
|
147
|
+
"फ"
|
|
148
|
+
],
|
|
149
|
+
[
|
|
150
|
+
"इ{",
|
|
151
|
+
"ई"
|
|
152
|
+
],
|
|
153
|
+
[
|
|
154
|
+
"ि((.्)*[^्])",
|
|
155
|
+
"\\1ि"
|
|
156
|
+
],
|
|
157
|
+
[
|
|
158
|
+
"(.[ािीुूृेैोौंःँ]*?){",
|
|
159
|
+
"{\\1"
|
|
160
|
+
],
|
|
161
|
+
[
|
|
162
|
+
"((.्)*){",
|
|
163
|
+
"{\\1"
|
|
164
|
+
],
|
|
165
|
+
[
|
|
166
|
+
"{",
|
|
167
|
+
"र्"
|
|
168
|
+
],
|
|
169
|
+
[
|
|
170
|
+
"([ाीुूृेैोौंःँ]+?)(्(.्)*[^्])",
|
|
171
|
+
"\\2\\1"
|
|
172
|
+
],
|
|
173
|
+
[
|
|
174
|
+
"्([ाीुूृेैोौंःँ]+?)((.्)*[^्])",
|
|
175
|
+
"्\\2\\1"
|
|
176
|
+
],
|
|
177
|
+
[
|
|
178
|
+
"([ंँ])([ािीुूृेैोौः]*)",
|
|
179
|
+
"\\2\\1"
|
|
180
|
+
],
|
|
181
|
+
[
|
|
182
|
+
"([ँंेैुू])\\1",
|
|
183
|
+
"\\1"
|
|
184
|
+
],
|
|
185
|
+
[
|
|
186
|
+
"^ः",
|
|
187
|
+
":"
|
|
188
|
+
],
|
|
189
|
+
[
|
|
190
|
+
"टृ",
|
|
191
|
+
"ट्ट"
|
|
192
|
+
],
|
|
193
|
+
[
|
|
194
|
+
"ेा",
|
|
195
|
+
"ाे"
|
|
196
|
+
],
|
|
197
|
+
[
|
|
198
|
+
"ैा",
|
|
199
|
+
"ाै"
|
|
200
|
+
],
|
|
201
|
+
[
|
|
202
|
+
"अाे",
|
|
203
|
+
"ओ"
|
|
204
|
+
],
|
|
205
|
+
[
|
|
206
|
+
"अाै",
|
|
207
|
+
"औ"
|
|
208
|
+
],
|
|
209
|
+
[
|
|
210
|
+
"अा",
|
|
211
|
+
"आ"
|
|
212
|
+
],
|
|
213
|
+
[
|
|
214
|
+
"एे",
|
|
215
|
+
"ऐ"
|
|
216
|
+
],
|
|
217
|
+
[
|
|
218
|
+
"ाे",
|
|
219
|
+
"ो"
|
|
220
|
+
],
|
|
221
|
+
[
|
|
222
|
+
"ाै",
|
|
223
|
+
"ौ"
|
|
224
|
+
]
|
|
225
|
+
]
|
|
226
|
+
}
|
|
@@ -0,0 +1,112 @@
|
|
|
1
|
+
{
|
|
2
|
+
"consonants": {
|
|
3
|
+
"k": "क",
|
|
4
|
+
"kh": "ख",
|
|
5
|
+
"g": "ग",
|
|
6
|
+
"gh": "घ",
|
|
7
|
+
"ng": "ङ",
|
|
8
|
+
"ch": "च",
|
|
9
|
+
"chh": "छ",
|
|
10
|
+
"j": "ज",
|
|
11
|
+
"jh": "झ",
|
|
12
|
+
"ny": "ञ",
|
|
13
|
+
"T": "ट",
|
|
14
|
+
"Th": "ठ",
|
|
15
|
+
"D": "ड",
|
|
16
|
+
"Dh": "ढ",
|
|
17
|
+
"N": "ण",
|
|
18
|
+
"t": "त",
|
|
19
|
+
"th": "थ",
|
|
20
|
+
"d": "द",
|
|
21
|
+
"dh": "ध",
|
|
22
|
+
"n": "न",
|
|
23
|
+
"p": "प",
|
|
24
|
+
"ph": "फ",
|
|
25
|
+
"f": "फ",
|
|
26
|
+
"b": "ब",
|
|
27
|
+
"bh": "भ",
|
|
28
|
+
"m": "म",
|
|
29
|
+
"y": "य",
|
|
30
|
+
"r": "र",
|
|
31
|
+
"l": "ल",
|
|
32
|
+
"v": "व",
|
|
33
|
+
"w": "व",
|
|
34
|
+
"sh": "श",
|
|
35
|
+
"Sh": "ष",
|
|
36
|
+
"s": "स",
|
|
37
|
+
"h": "ह",
|
|
38
|
+
"ksh": "क्ष",
|
|
39
|
+
"gy": "ज्ञ",
|
|
40
|
+
"shr": "श्र",
|
|
41
|
+
"tr": "त्र"
|
|
42
|
+
|
|
43
|
+
},
|
|
44
|
+
"vowels": {
|
|
45
|
+
"a": "अ",
|
|
46
|
+
"aa": "आ",
|
|
47
|
+
"A": "आ",
|
|
48
|
+
"i": "इ",
|
|
49
|
+
"I": "इ",
|
|
50
|
+
"ee": "ई",
|
|
51
|
+
"ii": "ई",
|
|
52
|
+
"u": "उ",
|
|
53
|
+
"U": "उ",
|
|
54
|
+
"oo": "ऊ",
|
|
55
|
+
"uu": "ऊ",
|
|
56
|
+
"Ri": "ऋ",
|
|
57
|
+
"R": "ऋ",
|
|
58
|
+
"Ree": "ॠ",
|
|
59
|
+
"lRi": "ऌ",
|
|
60
|
+
"e": "ए",
|
|
61
|
+
"ai": "ऐ",
|
|
62
|
+
"o": "ओ",
|
|
63
|
+
"au": "औ"
|
|
64
|
+
},
|
|
65
|
+
"matras": {
|
|
66
|
+
"a": "",
|
|
67
|
+
"aa": "ा",
|
|
68
|
+
"A": "ा",
|
|
69
|
+
"i": "ि",
|
|
70
|
+
"I": "ि",
|
|
71
|
+
"ee": "ी",
|
|
72
|
+
"ii": "ी",
|
|
73
|
+
"u": "ु",
|
|
74
|
+
"U": "ु",
|
|
75
|
+
"oo": "ू",
|
|
76
|
+
"uu": "ू",
|
|
77
|
+
"Ri": "ृ",
|
|
78
|
+
"R": "ृ",
|
|
79
|
+
"Ree": "ॄ",
|
|
80
|
+
"e": "े",
|
|
81
|
+
"ai": "ै",
|
|
82
|
+
"o": "ो",
|
|
83
|
+
"au": "ौ"
|
|
84
|
+
},
|
|
85
|
+
"special": {
|
|
86
|
+
"om": "ॐ",
|
|
87
|
+
"Om": "ॐ",
|
|
88
|
+
"aum": "ॐ",
|
|
89
|
+
".": "।",
|
|
90
|
+
"..": "॥",
|
|
91
|
+
"|": "।",
|
|
92
|
+
"||": "॥",
|
|
93
|
+
"M": "ं",
|
|
94
|
+
"H": "ः",
|
|
95
|
+
"~": "ँ"
|
|
96
|
+
},
|
|
97
|
+
"compound_consonants": {
|
|
98
|
+
|
|
99
|
+
},
|
|
100
|
+
"digits": {
|
|
101
|
+
"0": "०",
|
|
102
|
+
"1": "१",
|
|
103
|
+
"2": "२",
|
|
104
|
+
"3": "३",
|
|
105
|
+
"4": "४",
|
|
106
|
+
"5": "५",
|
|
107
|
+
"6": "६",
|
|
108
|
+
"7": "७",
|
|
109
|
+
"8": "८",
|
|
110
|
+
"9": "९"
|
|
111
|
+
}
|
|
112
|
+
}
|
|
@@ -0,0 +1,67 @@
|
|
|
1
|
+
{
|
|
2
|
+
"aum": "ॐ",
|
|
3
|
+
"bibek": "विवेक",
|
|
4
|
+
"call": "कल",
|
|
5
|
+
"code": "कोड",
|
|
6
|
+
"cool": "कूल",
|
|
7
|
+
"facebook": "फेसबुक",
|
|
8
|
+
"fun": "फन",
|
|
9
|
+
"garage": "ग्यारेज",
|
|
10
|
+
"gas": "ग्यास",
|
|
11
|
+
"google": "गूगल",
|
|
12
|
+
"hola": "होला",
|
|
13
|
+
"hudai": "हुँदै",
|
|
14
|
+
"instagram": "इन्स्टाग्राम",
|
|
15
|
+
"kathmandu": "काठमाडौं",
|
|
16
|
+
"Kathmandu": "काठमाडौं",
|
|
17
|
+
"laptop": "ल्याप्टप",
|
|
18
|
+
"message": "मेसेज",
|
|
19
|
+
"messenger": "मेसेन्जर",
|
|
20
|
+
"nepal": "नेपाल",
|
|
21
|
+
"nepali": "नेपाली",
|
|
22
|
+
"number": "नम्बर",
|
|
23
|
+
"pandey": "पाण्डे",
|
|
24
|
+
"phone": "फोन",
|
|
25
|
+
"plan": "प्लान",
|
|
26
|
+
"radio": "रेडियो",
|
|
27
|
+
"sms": "एसएमएस",
|
|
28
|
+
"SMS": "एसएमएस",
|
|
29
|
+
"telecom": "टेलिकम",
|
|
30
|
+
"twitter": "ट्विटर",
|
|
31
|
+
"vivek": "विवेक",
|
|
32
|
+
"about": "अबाउट",
|
|
33
|
+
"address": "ठेगाना",
|
|
34
|
+
"bank": "बैंक",
|
|
35
|
+
"battery": "ब्याट्री",
|
|
36
|
+
"bus": "बस",
|
|
37
|
+
"camera": "क्यामेरा",
|
|
38
|
+
"chat": "च्याट",
|
|
39
|
+
"city": "सिटी",
|
|
40
|
+
"coffee": "कफी",
|
|
41
|
+
"computer": "कम्प्युटर",
|
|
42
|
+
"contact": "कन्ट्याक्ट",
|
|
43
|
+
"country": "देश",
|
|
44
|
+
"date": "मिति",
|
|
45
|
+
"email": "इमेल",
|
|
46
|
+
"file": "फाइल",
|
|
47
|
+
"food": "फुड",
|
|
48
|
+
"friend": "फ्रेन्ड",
|
|
49
|
+
"game": "गेम",
|
|
50
|
+
"hello": "हेलो",
|
|
51
|
+
"internet": "इन्टरनेट",
|
|
52
|
+
"location": "लोकेशन",
|
|
53
|
+
"market": "मार्केट",
|
|
54
|
+
"meeting": "मिटिङ",
|
|
55
|
+
"mobile": "मोबाइल",
|
|
56
|
+
"music": "संगीत",
|
|
57
|
+
"name": "नाम",
|
|
58
|
+
"photo": "फोटो",
|
|
59
|
+
"school": "स्कूल",
|
|
60
|
+
"shop": "पसल",
|
|
61
|
+
"station": "स्टेसन",
|
|
62
|
+
"taxi": "ट्याक्सी",
|
|
63
|
+
"time": "समय",
|
|
64
|
+
"video": "भिडियो",
|
|
65
|
+
"wifi": "वाईफाई",
|
|
66
|
+
"work": "काम"
|
|
67
|
+
}
|
|
@@ -0,0 +1,107 @@
|
|
|
1
|
+
import re
|
|
2
|
+
from typing import Optional
|
|
3
|
+
|
|
4
|
+
from nepali_unicoder.loader import PreetiLoader, RuleLoader
|
|
5
|
+
from nepali_unicoder.tokenizer import Tokenizer
|
|
6
|
+
from nepali_unicoder.trie import Trie
|
|
7
|
+
|
|
8
|
+
|
|
9
|
+
class Engine:
|
|
10
|
+
VALID_MODES = {"roman", "preeti"}
|
|
11
|
+
|
|
12
|
+
def __init__(
|
|
13
|
+
self,
|
|
14
|
+
trie: Optional[Trie] = None,
|
|
15
|
+
tokenizer: Optional[Tokenizer] = None,
|
|
16
|
+
mode: str = "roman",
|
|
17
|
+
):
|
|
18
|
+
if mode not in self.VALID_MODES:
|
|
19
|
+
raise ValueError(
|
|
20
|
+
f"Invalid mode {mode!r}. Expected one of {sorted(self.VALID_MODES)}"
|
|
21
|
+
)
|
|
22
|
+
self.mode = mode
|
|
23
|
+
self.post_rules = []
|
|
24
|
+
|
|
25
|
+
if trie is None:
|
|
26
|
+
if mode == "preeti":
|
|
27
|
+
loader = PreetiLoader()
|
|
28
|
+
# Load post-processing rules for Preeti mode
|
|
29
|
+
raw_rules = loader.get_post_rules()
|
|
30
|
+
self.post_rules = [
|
|
31
|
+
(re.compile(pattern), replacement)
|
|
32
|
+
for pattern, replacement in raw_rules
|
|
33
|
+
]
|
|
34
|
+
else:
|
|
35
|
+
loader = RuleLoader()
|
|
36
|
+
self.trie = loader.load()
|
|
37
|
+
else:
|
|
38
|
+
self.trie = trie
|
|
39
|
+
|
|
40
|
+
if tokenizer is None:
|
|
41
|
+
self.tokenizer = Tokenizer()
|
|
42
|
+
else:
|
|
43
|
+
self.tokenizer = tokenizer
|
|
44
|
+
|
|
45
|
+
def transliterate(self, text: str) -> str:
|
|
46
|
+
"""
|
|
47
|
+
Convert Roman text to Devanagari using the Trie.
|
|
48
|
+
"""
|
|
49
|
+
if not text:
|
|
50
|
+
return ""
|
|
51
|
+
|
|
52
|
+
use_blocks = self.mode != "preeti"
|
|
53
|
+
tokens = self.tokenizer.tokenize(text, use_blocks=use_blocks)
|
|
54
|
+
result = []
|
|
55
|
+
|
|
56
|
+
for token in tokens:
|
|
57
|
+
if token.type == "BLOCK":
|
|
58
|
+
result.append(token.value)
|
|
59
|
+
elif token.type == "LITERAL":
|
|
60
|
+
result.append(token.value)
|
|
61
|
+
elif token.type == "NUMBER":
|
|
62
|
+
# Process Number chunk: transliterate digits, keep others (like .) as is
|
|
63
|
+
for char in token.value:
|
|
64
|
+
if char.isdigit():
|
|
65
|
+
# Use Trie to find digit mapping (digits are single chars in rules)
|
|
66
|
+
match_val, _ = self.trie.longest_match(char)
|
|
67
|
+
if match_val:
|
|
68
|
+
result.append(match_val)
|
|
69
|
+
else:
|
|
70
|
+
result.append(char)
|
|
71
|
+
else:
|
|
72
|
+
result.append(char)
|
|
73
|
+
elif token.type == "ROMAN":
|
|
74
|
+
# Process Roman chunk with Trie
|
|
75
|
+
chunk = token.value
|
|
76
|
+
idx = 0
|
|
77
|
+
chunk_len = len(chunk)
|
|
78
|
+
|
|
79
|
+
while idx < chunk_len:
|
|
80
|
+
# Try to find longest match starting at idx
|
|
81
|
+
# Pass chunk and idx directly to avoid slicing
|
|
82
|
+
match_val, match_len = self.trie.longest_match(chunk, idx)
|
|
83
|
+
|
|
84
|
+
if match_val:
|
|
85
|
+
result.append(match_val)
|
|
86
|
+
idx += match_len
|
|
87
|
+
else:
|
|
88
|
+
# No match, keep character as is
|
|
89
|
+
result.append(chunk[idx])
|
|
90
|
+
idx += 1
|
|
91
|
+
|
|
92
|
+
output = "".join(result)
|
|
93
|
+
|
|
94
|
+
# Apply post-processing rules for Preeti mode
|
|
95
|
+
if self.mode == "preeti" and self.post_rules:
|
|
96
|
+
output = self._apply_post_rules(output)
|
|
97
|
+
|
|
98
|
+
return output
|
|
99
|
+
|
|
100
|
+
def _apply_post_rules(self, text: str) -> str:
|
|
101
|
+
"""
|
|
102
|
+
Apply post-processing rules (regex replacements) to the text.
|
|
103
|
+
Used for Preeti mode to handle contextual transformations.
|
|
104
|
+
"""
|
|
105
|
+
for pattern, replacement in self.post_rules:
|
|
106
|
+
text = pattern.sub(replacement, text)
|
|
107
|
+
return text
|
|
@@ -0,0 +1,120 @@
|
|
|
1
|
+
import json
|
|
2
|
+
import logging
|
|
3
|
+
import os
|
|
4
|
+
|
|
5
|
+
from nepali_unicoder.trie import Trie
|
|
6
|
+
|
|
7
|
+
_logger = logging.getLogger(__name__)
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
def load_json_data(filename):
|
|
11
|
+
path = os.path.join(os.path.dirname(__file__), "data", filename)
|
|
12
|
+
try:
|
|
13
|
+
with open(path, "r", encoding="utf-8") as f:
|
|
14
|
+
return json.load(f)
|
|
15
|
+
except FileNotFoundError:
|
|
16
|
+
_logger.warning("Rule file %s not found.", filename)
|
|
17
|
+
return {}
|
|
18
|
+
|
|
19
|
+
|
|
20
|
+
class RuleLoader:
|
|
21
|
+
def __init__(self):
|
|
22
|
+
self.word_maps_path = os.path.join(
|
|
23
|
+
os.path.dirname(__file__), "data", "word_maps.json"
|
|
24
|
+
)
|
|
25
|
+
|
|
26
|
+
def load(self) -> Trie:
|
|
27
|
+
"""Load rules and custom mappings into a Trie."""
|
|
28
|
+
trie = Trie()
|
|
29
|
+
self._load_rules(trie)
|
|
30
|
+
self._load_custom_mappings(trie)
|
|
31
|
+
return trie
|
|
32
|
+
|
|
33
|
+
def _load_rules(self, trie: Trie):
|
|
34
|
+
data = load_json_data("roman_rules.json")
|
|
35
|
+
|
|
36
|
+
consonants = data.get("consonants", {})
|
|
37
|
+
vowels = data.get("vowels", {})
|
|
38
|
+
matras = data.get("matras", {})
|
|
39
|
+
special = data.get("special", {})
|
|
40
|
+
digits = data.get("digits", {})
|
|
41
|
+
|
|
42
|
+
# 1. Consonants and Combinations (inserted BEFORE vowels so that
|
|
43
|
+
# independent vowel keys inserted next will override any collision,
|
|
44
|
+
# e.g. 'lRi' consonant+matra combo is overridden by vowel 'lRi' -> ऌ)
|
|
45
|
+
halanta = "्"
|
|
46
|
+
|
|
47
|
+
for rom_cons, dev_cons in consonants.items():
|
|
48
|
+
# Case 1: Consonant alone (halanta form) -> 'k' -> 'क्'
|
|
49
|
+
trie.add(rom_cons, dev_cons + halanta)
|
|
50
|
+
|
|
51
|
+
# Case 2: Consonant + 'a' (Schwa form) -> 'ka' -> 'क'
|
|
52
|
+
trie.add(rom_cons + "a", dev_cons)
|
|
53
|
+
|
|
54
|
+
# Case 3: Consonant + other vowels -> 'ki' -> 'कि'
|
|
55
|
+
for rom_vowel, matra in matras.items():
|
|
56
|
+
if rom_vowel == "a":
|
|
57
|
+
continue # Handled above
|
|
58
|
+
trie.add(rom_cons + rom_vowel, dev_cons + matra)
|
|
59
|
+
|
|
60
|
+
# 1b. Compound consonants whose romanization ends in a vowel-like letter
|
|
61
|
+
# (e.g. 'tra' -> 'त्र') -- these CANNOT go through the normal consonant
|
|
62
|
+
# loop because appending 'a' for the schwa would yield 'traa' not 'tra'.
|
|
63
|
+
# Instead we build their combos explicitly:
|
|
64
|
+
# 'tra' -> 'त्र' (natural: the trailing 'a' IS the schwa)
|
|
65
|
+
# 'tra' + matra -> 'त्र' + matra (e.g. 'trai' -> 'त्रि')
|
|
66
|
+
# 'tra' + halanta -> 'त्र्' (explicit halanta: append 'H' or use '{}')
|
|
67
|
+
compound_consonants = data.get("compound_consonants", {})
|
|
68
|
+
for rom_stem, dev_stem in compound_consonants.items():
|
|
69
|
+
# Schwa form: the romanization itself (e.g. 'tra' -> 'त्र')
|
|
70
|
+
trie.add(rom_stem, dev_stem)
|
|
71
|
+
# With other vowels / matras (e.g. 'trai' -> 'त्रि')
|
|
72
|
+
for rom_vowel, matra in matras.items():
|
|
73
|
+
if rom_vowel == "a":
|
|
74
|
+
continue # schwa already covered above
|
|
75
|
+
trie.add(rom_stem + rom_vowel, dev_stem + matra)
|
|
76
|
+
|
|
77
|
+
# 2. Independent Vowels (inserted AFTER consonant combos so they win
|
|
78
|
+
# on collision, e.g. 'lRi' -> ऌ overrides 'l'+'Ri' -> लृ)
|
|
79
|
+
for rom, dev in vowels.items():
|
|
80
|
+
trie.add(rom, dev)
|
|
81
|
+
|
|
82
|
+
# 3. Special, Digits, Punctuation
|
|
83
|
+
for rom, dev in special.items():
|
|
84
|
+
trie.add(rom, dev)
|
|
85
|
+
|
|
86
|
+
for rom, dev in digits.items():
|
|
87
|
+
trie.add(rom, dev)
|
|
88
|
+
|
|
89
|
+
def _load_custom_mappings(self, trie: Trie):
|
|
90
|
+
if not os.path.exists(self.word_maps_path):
|
|
91
|
+
return
|
|
92
|
+
|
|
93
|
+
with open(self.word_maps_path, "r", encoding="utf-8") as f:
|
|
94
|
+
mappings = json.load(f)
|
|
95
|
+
|
|
96
|
+
for roman, devanagari in mappings.items():
|
|
97
|
+
trie.add(roman, devanagari)
|
|
98
|
+
# Automatically add capitalized version if it doesn't exist
|
|
99
|
+
# (e.g. if 'nepal' is in maps, also add 'Nepal')
|
|
100
|
+
cap_roman = roman.capitalize()
|
|
101
|
+
if cap_roman not in mappings:
|
|
102
|
+
trie.add(cap_roman, devanagari)
|
|
103
|
+
|
|
104
|
+
|
|
105
|
+
class PreetiLoader:
|
|
106
|
+
def __init__(self):
|
|
107
|
+
# Parse the JSON file once; reused by both load() and get_post_rules()
|
|
108
|
+
# so the file is never read more than once per PreetiLoader instance.
|
|
109
|
+
self._data = load_json_data("preeti_rules.json")
|
|
110
|
+
|
|
111
|
+
def load(self) -> Trie:
|
|
112
|
+
"""Load Preeti rules into a Trie."""
|
|
113
|
+
trie = Trie()
|
|
114
|
+
mappings = self._data.get("mappings", {})
|
|
115
|
+
for key, value in mappings.items():
|
|
116
|
+
trie.add(key, value)
|
|
117
|
+
return trie
|
|
118
|
+
|
|
119
|
+
def get_post_rules(self):
|
|
120
|
+
return self._data.get("post_rules", [])
|
nepali_unicoder/rules.py
ADDED
|
@@ -0,0 +1,93 @@
|
|
|
1
|
+
import re
|
|
2
|
+
from dataclasses import dataclass
|
|
3
|
+
from typing import List
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
@dataclass
|
|
7
|
+
class Token:
|
|
8
|
+
value: str
|
|
9
|
+
type: str # 'ROMAN', 'BLOCK', 'PUNCTUATION', 'UNKNOWN', 'LITERAL', 'NUMBER'
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
class Tokenizer:
|
|
13
|
+
# Compiled once at class definition time -- not per tokenize() call.
|
|
14
|
+
_RE_BLOCK = re.compile(r"\{([^}]*)\}")
|
|
15
|
+
_RE_NUMBER = re.compile(r"\d*\.\d+|\d+")
|
|
16
|
+
|
|
17
|
+
def tokenize(self, text: str, use_blocks: bool = True) -> List[Token]:
|
|
18
|
+
"""
|
|
19
|
+
Split text into tokens using regex-based matching for robustness.
|
|
20
|
+
"""
|
|
21
|
+
tokens = []
|
|
22
|
+
i = 0
|
|
23
|
+
n = len(text)
|
|
24
|
+
roman_buffer = []
|
|
25
|
+
|
|
26
|
+
re_block = self._RE_BLOCK
|
|
27
|
+
re_number = self._RE_NUMBER
|
|
28
|
+
|
|
29
|
+
def flush_roman():
|
|
30
|
+
if roman_buffer:
|
|
31
|
+
tokens.append(Token(value="".join(roman_buffer), type="ROMAN"))
|
|
32
|
+
roman_buffer.clear()
|
|
33
|
+
|
|
34
|
+
while i < n:
|
|
35
|
+
# 1. Check for Ellipsis (Priority)
|
|
36
|
+
if text.startswith("...", i):
|
|
37
|
+
flush_roman()
|
|
38
|
+
tokens.append(Token(value="...", type="LITERAL"))
|
|
39
|
+
i += 3
|
|
40
|
+
continue
|
|
41
|
+
|
|
42
|
+
# 2. Check for Blocks and Braces (if enabled)
|
|
43
|
+
if use_blocks:
|
|
44
|
+
# Escaped brace {{
|
|
45
|
+
if text.startswith("{{", i):
|
|
46
|
+
flush_roman()
|
|
47
|
+
tokens.append(Token(value="{", type="LITERAL"))
|
|
48
|
+
i += 2
|
|
49
|
+
continue
|
|
50
|
+
|
|
51
|
+
# Block {content}
|
|
52
|
+
match_block = re_block.match(text, i)
|
|
53
|
+
if match_block:
|
|
54
|
+
flush_roman()
|
|
55
|
+
tokens.append(Token(value=match_block.group(1), type="BLOCK"))
|
|
56
|
+
i = match_block.end()
|
|
57
|
+
continue
|
|
58
|
+
|
|
59
|
+
# Escaped brace }}
|
|
60
|
+
if text.startswith("}}", i):
|
|
61
|
+
flush_roman()
|
|
62
|
+
tokens.append(Token(value="}", type="LITERAL"))
|
|
63
|
+
i += 2
|
|
64
|
+
continue
|
|
65
|
+
|
|
66
|
+
# Unmatched closing brace }
|
|
67
|
+
if text.startswith("}", i):
|
|
68
|
+
flush_roman()
|
|
69
|
+
tokens.append(Token(value="}", type="LITERAL"))
|
|
70
|
+
i += 1
|
|
71
|
+
continue
|
|
72
|
+
|
|
73
|
+
# Unmatched opening brace {
|
|
74
|
+
if text.startswith("{", i):
|
|
75
|
+
flush_roman()
|
|
76
|
+
tokens.append(Token(value="{", type="LITERAL"))
|
|
77
|
+
i += 1
|
|
78
|
+
continue
|
|
79
|
+
|
|
80
|
+
# 3. Check for Numbers
|
|
81
|
+
match_number = re_number.match(text, i)
|
|
82
|
+
if match_number:
|
|
83
|
+
flush_roman()
|
|
84
|
+
tokens.append(Token(value=match_number.group(0), type="NUMBER"))
|
|
85
|
+
i = match_number.end()
|
|
86
|
+
continue
|
|
87
|
+
|
|
88
|
+
# 4. Fallback: Accumulate Roman
|
|
89
|
+
roman_buffer.append(text[i])
|
|
90
|
+
i += 1
|
|
91
|
+
|
|
92
|
+
flush_roman()
|
|
93
|
+
return tokens
|
nepali_unicoder/trie.py
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
1
|
+
from typing import Dict, Optional, Tuple
|
|
2
|
+
|
|
3
|
+
|
|
4
|
+
class TrieNode:
|
|
5
|
+
__slots__ = ("children", "value", "is_end")
|
|
6
|
+
|
|
7
|
+
def __init__(self):
|
|
8
|
+
self.children: Dict[str, TrieNode] = {}
|
|
9
|
+
self.value: Optional[str] = None
|
|
10
|
+
self.is_end: bool = False
|
|
11
|
+
|
|
12
|
+
|
|
13
|
+
class Trie:
|
|
14
|
+
def __init__(self):
|
|
15
|
+
self.root = TrieNode()
|
|
16
|
+
|
|
17
|
+
def add(self, key: str, value: str) -> None:
|
|
18
|
+
"""Insert a key-value pair into the Trie."""
|
|
19
|
+
node = self.root
|
|
20
|
+
for char in key:
|
|
21
|
+
if char not in node.children:
|
|
22
|
+
node.children[char] = TrieNode()
|
|
23
|
+
node = node.children[char]
|
|
24
|
+
node.is_end = True
|
|
25
|
+
node.value = value
|
|
26
|
+
|
|
27
|
+
def longest_match(
|
|
28
|
+
self, text: str, start_index: int = 0
|
|
29
|
+
) -> Tuple[Optional[str], int]:
|
|
30
|
+
"""
|
|
31
|
+
Find the longest key in the Trie that matches the start of the text.
|
|
32
|
+
Returns (value, length_of_match).
|
|
33
|
+
If no match found, returns (None, 0).
|
|
34
|
+
"""
|
|
35
|
+
node = self.root
|
|
36
|
+
last_match_value = None
|
|
37
|
+
last_match_len = 0
|
|
38
|
+
current_len = 0
|
|
39
|
+
n = len(text)
|
|
40
|
+
i = start_index
|
|
41
|
+
|
|
42
|
+
while i < n:
|
|
43
|
+
char = text[i]
|
|
44
|
+
if char not in node.children:
|
|
45
|
+
break
|
|
46
|
+
node = node.children[char]
|
|
47
|
+
current_len += 1
|
|
48
|
+
i += 1
|
|
49
|
+
if node.is_end:
|
|
50
|
+
last_match_value = node.value
|
|
51
|
+
last_match_len = current_len
|
|
52
|
+
|
|
53
|
+
return last_match_value, last_match_len
|
|
@@ -0,0 +1,210 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: nepali_unicoder
|
|
3
|
+
Version: 0.2.0
|
|
4
|
+
Summary: A Python package for Roman to Nepali (Devanagari) transliteration
|
|
5
|
+
Author-email: realsanjeev <075bei033.sanjeev@pcampus.edu.np>
|
|
6
|
+
License: MIT
|
|
7
|
+
Project-URL: Homepage, https://github.com/realsanjeev/nepali_unicoder
|
|
8
|
+
Project-URL: Documentation, https://realsanjeev.github.io/nepali_unicoder/
|
|
9
|
+
Project-URL: Bug Tracker, https://github.com/realsanjeev/nepali_unicoder/issues
|
|
10
|
+
Keywords: nepali,unicoder,roman to devanagari,nepali transliteration,preeti to unicode
|
|
11
|
+
Classifier: Programming Language :: Python
|
|
12
|
+
Classifier: Programming Language :: Python :: 3
|
|
13
|
+
Classifier: Operating System :: OS Independent
|
|
14
|
+
Requires-Python: >=3.7
|
|
15
|
+
Description-Content-Type: text/markdown
|
|
16
|
+
License-File: LICENSE
|
|
17
|
+
Provides-Extra: dev
|
|
18
|
+
Requires-Dist: pre-commit; extra == "dev"
|
|
19
|
+
Requires-Dist: mkdocs; extra == "dev"
|
|
20
|
+
Requires-Dist: mkdocs-material; extra == "dev"
|
|
21
|
+
Dynamic: license-file
|
|
22
|
+
|
|
23
|
+
# Nepali Unicoder
|
|
24
|
+
|
|
25
|
+
[](https://realsanjeev.github.io/nepali_unicoder/)
|
|
26
|
+
[](https://pypi.org/project/nepali-unicoder/)
|
|
27
|
+
|
|
28
|
+
A robust Python package for converting Romanized Nepali text and Preeti font text into Unicode Devanagari script. It uses a greedy matching algorithm for Roman transliteration and a two-phase conversion process for Preeti with contextual rules.
|
|
29
|
+
|
|
30
|
+
Read the **[Full Documentation](https://realsanjeev.github.io/nepali_unicoder/)** for detailed usage guides, Preeti mapping references, and API details.
|
|
31
|
+
|
|
32
|
+
## Features
|
|
33
|
+
|
|
34
|
+
- **Accurate Transliteration**: Uses a greedy matching algorithm to prioritize longer phonetic matches (e.g., 'kha' is matched before 'k' and 'h').
|
|
35
|
+
- **Preeti Font Support**: Full support for Preeti to Unicode conversion with 30+ contextual rules for accurate transformation.
|
|
36
|
+
- **Smart Vowel Handling**: Distinguishes between independent vowels (e.g., 'aa' -> 'आ') and vowel signs/matras (e.g., 'ka' -> 'क', 'kaa' -> 'का').
|
|
37
|
+
- **Contextual Rules**: Handles complex Devanagari rules like reph positioning, matra reordering, and special character combinations.
|
|
38
|
+
- **Mixed Content Support**: Allows keeping English words or specific text in Roman script using `{}` blocks.
|
|
39
|
+
- **Customizable**: Supports custom word-level overrides via `word_maps.json`.
|
|
40
|
+
- **CLI Support**: Can be used directly from the command line.
|
|
41
|
+
|
|
42
|
+
## Installation
|
|
43
|
+
|
|
44
|
+
You can install the package locally:
|
|
45
|
+
|
|
46
|
+
```bash
|
|
47
|
+
pip install nepali-unicoder
|
|
48
|
+
```
|
|
49
|
+
|
|
50
|
+
## Usage
|
|
51
|
+
|
|
52
|
+
### Command Line Interface (CLI)
|
|
53
|
+
|
|
54
|
+
You can use the converter directly from the terminal:
|
|
55
|
+
|
|
56
|
+
```bash
|
|
57
|
+
# Using the installed command
|
|
58
|
+
nepali-unicoder "namaste"
|
|
59
|
+
# Output: नमस्ते
|
|
60
|
+
|
|
61
|
+
# Or using python module
|
|
62
|
+
python -m nepali_unicoder "namaste"
|
|
63
|
+
# Output: नमस्ते
|
|
64
|
+
|
|
65
|
+
# Pipe input
|
|
66
|
+
echo "mero naam sanjeev ho" | nepali-unicoder
|
|
67
|
+
# Output: मेरो नाम् सन्जीव् हो
|
|
68
|
+
```
|
|
69
|
+
|
|
70
|
+
### Python API
|
|
71
|
+
|
|
72
|
+
```python
|
|
73
|
+
from nepali_unicoder.convert import Converter
|
|
74
|
+
|
|
75
|
+
converter = Converter()
|
|
76
|
+
|
|
77
|
+
# Basic conversion
|
|
78
|
+
text = "namaste nepal"
|
|
79
|
+
print(converter.convert(text))
|
|
80
|
+
# Output: नमस्ते नेपाल
|
|
81
|
+
|
|
82
|
+
# Using 'as-is' blocks for English text
|
|
83
|
+
mixed_text = "mero naam {Sanjeev} ho"
|
|
84
|
+
print(converter.convert(mixed_text))
|
|
85
|
+
# Output: मेरो नाम् Sanjeev हो
|
|
86
|
+
```
|
|
87
|
+
|
|
88
|
+
### Preeti Mode
|
|
89
|
+
|
|
90
|
+
Convert Preeti font text to Unicode with full support for contextual rules:
|
|
91
|
+
|
|
92
|
+
```python
|
|
93
|
+
from nepali_unicoder.convert import Converter
|
|
94
|
+
|
|
95
|
+
# Create converter in Preeti mode
|
|
96
|
+
preeti_converter = Converter(mode="preeti")
|
|
97
|
+
|
|
98
|
+
# Basic conversion
|
|
99
|
+
preeti_text = "s{sf" # Preeti characters
|
|
100
|
+
print(preeti_converter.convert(preeti_text))
|
|
101
|
+
# Output: र्कका
|
|
102
|
+
|
|
103
|
+
# The converter handles:
|
|
104
|
+
# - Reph positioning: { → र् (moves before consonant)
|
|
105
|
+
# - Matra reordering: l (ि) moves after consonant
|
|
106
|
+
# - Special m transformations
|
|
107
|
+
# - Vowel combinations
|
|
108
|
+
# - Literal brackets: { and } are treated as normal characters in Preeti mode
|
|
109
|
+
```
|
|
110
|
+
|
|
111
|
+
#### Preeti Character Examples
|
|
112
|
+
|
|
113
|
+
| Preeti | Unicode | Description |
|
|
114
|
+
|--------|---------|-------------|
|
|
115
|
+
| `s` | `क` | Consonant ka |
|
|
116
|
+
| `s{` | `र्क` | Reph + ka (contextual) |
|
|
117
|
+
| `sl` | `कि` | ka + short i (reordered) |
|
|
118
|
+
| `qm` | `क्र` | Special m transformation |
|
|
119
|
+
| `!@#` | `१२३` | Nepali numbers |
|
|
120
|
+
| `Ù` / `Ú` | `;` / `:` | Literal punctuation |
|
|
121
|
+
| `«` / `»` | `्र` | Ra-foot (for ट, ठ, ड, ढ) |
|
|
122
|
+
| `¿` | `रू` | Combined ruu |
|
|
123
|
+
| `å` | `द्व` | Combined dva |
|
|
124
|
+
| `ˆ` | `फ्` | Half ph |
|
|
125
|
+
| `ª` | `ङ` | Consonant nga |
|
|
126
|
+
| `æ` / `Æ` | `“` / `”` | Curly quotes |
|
|
127
|
+
| `¥` | `र्` | Half ra |
|
|
128
|
+
| `¶` | `ठ्ठ` | Combined thth |
|
|
129
|
+
| `§` | `ट्ट` | Combined tt |
|
|
130
|
+
| `£` | `घ्` | Half gh |
|
|
131
|
+
| `Ë` / `Í` | `ङ्ग` / `ङ्क` | Combined nga-ga / nga-ka |
|
|
132
|
+
| `‰` | `झ्` | Half jh |
|
|
133
|
+
|
|
134
|
+
#### CLI for Preeti
|
|
135
|
+
|
|
136
|
+
```bash
|
|
137
|
+
python -m nepali_unicoder --preeti "s{sf"
|
|
138
|
+
# Output: र्कका
|
|
139
|
+
```
|
|
140
|
+
|
|
141
|
+
|
|
142
|
+
## Transliteration Rules
|
|
143
|
+
|
|
144
|
+
- **Consonants**: `k` -> `क्`, `ka` -> `क`, `kh` -> `ख्`, `kha` -> `ख`
|
|
145
|
+
- **Vowels**: `a` -> `अ`, `aa` -> `आ`, `i` -> `इ`, `u` -> `उ`
|
|
146
|
+
- **Matras**: `ki` -> `कि`, `ko` -> `को`
|
|
147
|
+
- **Special**: `.` -> `।`, `..` -> `॥`
|
|
148
|
+
- **Numbers**: `0-9` -> `०-९` (Decimal points are preserved: `1.5` -> `१.५`)
|
|
149
|
+
|
|
150
|
+
## Advanced Usage
|
|
151
|
+
|
|
152
|
+
### Handling Complex Text
|
|
153
|
+
|
|
154
|
+
The converter handles mixed content gracefully. You can use `{}` to keep text as-is (e.g., for English words or code snippets).
|
|
155
|
+
|
|
156
|
+
```python
|
|
157
|
+
text = "mero naam {Sanjeev} ho ra ma 12.5 barsa ko bhaye."
|
|
158
|
+
print(converter.convert(text))
|
|
159
|
+
# Output: मेरो नाम् Sanjeev हो र म १२.५ बर्स को भये।
|
|
160
|
+
```
|
|
161
|
+
|
|
162
|
+
### Configuration
|
|
163
|
+
|
|
164
|
+
The package uses `word_maps.json` for custom word-level overrides, located in the `src/nepali_unicoder` directory.
|
|
165
|
+
|
|
166
|
+
1. **`word_maps.json`**: Defines custom word-level overrides. Use this for words that don't follow standard phonetic rules.
|
|
167
|
+
|
|
168
|
+
Example `word_maps.json`:
|
|
169
|
+
```json
|
|
170
|
+
{
|
|
171
|
+
"nepal": "नेपाल",
|
|
172
|
+
"kathamandu": "काठमाडौँ"
|
|
173
|
+
}
|
|
174
|
+
```
|
|
175
|
+
|
|
176
|
+
## Contribution
|
|
177
|
+
|
|
178
|
+
We welcome contributions! Here's how you can help:
|
|
179
|
+
|
|
180
|
+
1. **Clone the repository**:
|
|
181
|
+
```bash
|
|
182
|
+
git clone https://github.com/realsanjeev/nepali_unicoder.git
|
|
183
|
+
cd nepali_unicoder
|
|
184
|
+
```
|
|
185
|
+
|
|
186
|
+
2. **Set up a virtual environment**:
|
|
187
|
+
```bash
|
|
188
|
+
python3 -m venv .venv
|
|
189
|
+
source .venv/bin/activate
|
|
190
|
+
pip install -e .
|
|
191
|
+
```
|
|
192
|
+
|
|
193
|
+
3. **Run tests**:
|
|
194
|
+
```bash
|
|
195
|
+
python -m unittest discover tests
|
|
196
|
+
```
|
|
197
|
+
|
|
198
|
+
4. **Submit a Pull Request**: Create a new branch, make your changes, and submit a PR.
|
|
199
|
+
|
|
200
|
+
## Development
|
|
201
|
+
|
|
202
|
+
To run tests:
|
|
203
|
+
|
|
204
|
+
```bash
|
|
205
|
+
python -m unittest discover tests
|
|
206
|
+
```
|
|
207
|
+
|
|
208
|
+
## License
|
|
209
|
+
|
|
210
|
+
[MIT](./LICENSE)
|
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
nepali_unicoder/__init__.py,sha256=ll1McRUqVExqDwXUNpMhRvfzMYKp0DPevUjKofmgoec,71
|
|
2
|
+
nepali_unicoder/__main__.py,sha256=dsrU6FZQM-4oORWuzteKE897f8Ahr3bg9DmPhg6AzfE,974
|
|
3
|
+
nepali_unicoder/convert.py,sha256=st0O0zc_-UZiGjcHCo8RJUNkjjam1cmecs39VT4zlZ0,306
|
|
4
|
+
nepali_unicoder/engine.py,sha256=hKSI58hOt7Ek5Q98CWBbOHbQm0jPa-solWySVuevvEk,3651
|
|
5
|
+
nepali_unicoder/loader.py,sha256=Qx9l2QkbeDg-0sDuIW_PJcf6-5iaFxn52NyTAOPqK-E,4593
|
|
6
|
+
nepali_unicoder/rules.py,sha256=Ezxpwz0bq55QWzr_qU9HR-spEa5W3zy-xYolqoLPVGE,134
|
|
7
|
+
nepali_unicoder/tokenizer.py,sha256=XtygIM1Pgz908MDQmycevGG80XVfgT_wyPN9v-_cnSI,2934
|
|
8
|
+
nepali_unicoder/trie.py,sha256=7HF1WCSaCUP0x3uEIiXxdyyuHDI_j2B1nbnm976P7Wo,1478
|
|
9
|
+
nepali_unicoder/data/preeti_rules.json,sha256=_0-UXtDY7_mHnzRsUWibh3FU4S448Crw25ImrT_p9YQ,4426
|
|
10
|
+
nepali_unicoder/data/roman_rules.json,sha256=F8FuDfYvA17CpVjgOCrRuev2bDyViT6MA-qaWi1pKlo,2147
|
|
11
|
+
nepali_unicoder/data/word_maps.json,sha256=z6di4YZn_EGvkKvNa9bffLcYw-esTAKYA2PJPIqBqrk,2158
|
|
12
|
+
nepali_unicoder-0.2.0.dist-info/licenses/LICENSE,sha256=nWolli2fvrHF2ActODUDH0QjSKZf8EDb-M-vp74xa9I,1068
|
|
13
|
+
nepali_unicoder-0.2.0.dist-info/METADATA,sha256=re8lsooBQwdjIKn_wgnZSbOlLhVXT6PiHweLDOiDFFM,6626
|
|
14
|
+
nepali_unicoder-0.2.0.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
|
|
15
|
+
nepali_unicoder-0.2.0.dist-info/entry_points.txt,sha256=JtHpymd3LbtuCj1uPl9p2NsACU_fdMSSjdYtJ-VUZGU,66
|
|
16
|
+
nepali_unicoder-0.2.0.dist-info/top_level.txt,sha256=hoBnas3ztt0HTE0NM5jWgBsqDetj9OPg4rMwMPTqM3I,16
|
|
17
|
+
nepali_unicoder-0.2.0.dist-info/RECORD,,
|
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2025 RealSanjeev
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
nepali_unicoder
|