UNPKG

deep-profanity-filter

Version:

A thorough profanity filter that considers most common circumventions. Works with your custom list of blocked and whitelisted words and phrases. Identifies and/or replaces bad words. Works with *wildcards* at *start and/or end* of words.

github.com/Zariem/deep-profanity-filter

Zariem/deep-profanity-filter

202 lines â€Ē 14.1 kB
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
"use strict"; Object.defineProperty(exports, "__esModule", { value: true }); exports.reduceRepeatCharacters = exports.textToLatin = exports.removeTextAccents = exports.unEmoji = void 0; /** * Replaces all emojis in a text that feature a letter with * normal latin characters. * Example: "🇎⭕ 🔛" turns into "go on" or * "ðŸ…ŋ🇚ÂŪïļðŸ’°ðŸ‡Š" turns into "purse". * Note: This does NOT replace random emojis used to represent * or mask letters, such as 'ðŸŒļ' representing an 'O'. * * Suggested usage: `textToLatin(unEmoji(inputText))` * @param {string} inputText - The text from which you would like to remove all text based emojis. * @returns the input text, with all letter based emojis transformed to become text. */ function unEmoji(inputText) { return inputText .replace(/🆔/g, 'id') .replace(/🆚/g, 'vs') .replace(/ðŸ”Ī/g, 'abc') .replace(/🆎/g, 'ab') .replace(/🆑/g, 'cl') .replace(/🆘/g, 'sos') .replace(/ðŸšū/g, 'wc') .replace(/🆖/g, 'ng') .replace(/🆗/g, 'ok') .replace(/🆙/g, 'up') .replace(/🆒/g, 'cool') .replace(/🆕/g, 'new') .replace(/🔟/g, '10') .replace(/🆓/g, 'free') .replace(/🔚/g, 'end') .replace(/🔙/g, 'back') .replace(/🔛/g, 'on') .replace(/🔝/g, 'top') .replace(/🔜/g, 'soon') .replace(/ðŸ‡Ķ|🅰/g, 'a') .replace(/🇧|ðŸ…ą/g, 'b') .replace(/ðŸ‡Ļ|ÂĐïļ/g, 'c') .replace(/ðŸ‡Đ/g, 'd') .replace(/🇊/g, 'e') .replace(/ðŸ‡Ŧ/g, 'f') .replace(/🇎/g, 'g') .replace(/🇭/g, 'h') .replace(/ðŸ‡Ū/g, 'i') .replace(/ðŸ‡Ŋ/g, 'j') .replace(/🇰/g, 'k') .replace(/ðŸ‡ą/g, 'l') .replace(/ðŸ‡ē/g, 'm') .replace(/ðŸ‡ģ/g, 'n') .replace(/ðŸ‡ī|ðŸ…ū|⭕/g, 'o') .replace(/ðŸ‡ĩ|ðŸ…ŋ/g, 'p') .replace(/ðŸ‡ķ/g, 'q') .replace(/🇷|ÂŪïļ/g, 'r') .replace(/ðŸ‡ļ|💰/g, 's') .replace(/ðŸ‡đ/g, 't') .replace(/🇚/g, 'u') .replace(/ðŸ‡ŧ/g, 'v') .replace(/🇞/g, 'w') .replace(/ðŸ‡―|❎|❌/g, 'x') .replace(/ðŸ‡ū/g, 'y') .replace(/ðŸ‡ŋ/g, 'z'); } exports.unEmoji = unEmoji; /** * Removes most common accents from characters. * Example: The text "ZĖĩ͝Ė­ĖĄaĖķĖ‰Í„ĖŽĖ˜lĖķ͗ĖœgĖĩ͒́ĖœĖēoĖķĖ…ĖŠĖž" becomes "Zalgo", * the text "à-cÃītÃĐs" becomes "a-cotes", * non-latin characters stay non-latin, e.g. "ᑕⓞ֑ÎŋĖĪ͕𝕃Ũ…Ü‘" becomes "ᑕⓞÎŋ𝕃". * @param {string} inputText - The text for which you wish to have all * accents removed. * @returns the input text, stripped of all accents. */ function removeTextAccents(inputText) { return inputText .normalize('NFD') .replace(/[\u0300-\u036f]|[\u0591-\u05bd]|\u05bf|\u05c1|\u05c2|\u05c4|\u05c5|\u05c7|\u0711|[\u0730-\u074a]/g, ''); } exports.removeTextAccents = removeTextAccents; /** * Converts a text of fancy unicode font to latin alphabet characters. * This translation happens based on "visual appearance" of the letters, * so if you do this to text that is written in a language of non-latin * alphabet, you will get weird outputs. * * Disclaimer: This may at times mistranslate messages, and the list of * characters that get converted is most likely not complete, although * it is very thoroughly assembled. It will remove most common accents, * and returns a latin string in lower case letters. Any characters that * could not be mapped to latin characters will still appear in the string. * * Example: * "áĩšâ“˜ęŊą â‚ŪᏋäđ‚Õ‡" would turn into "this text" or "ZĖĩ͝Ė­ĖĄaĖķĖ‰Í„ĖŽĖ˜lĖķ͗ĖœgĖĩ͒́ĖœĖēoĖķĖ…ĖŠĖž" turns into "zalgo", or * "ᑕⓞ֑ÎŋĖĪ͕𝕃Ũ…Ü‘" turns into "cool". * * Suggested usage: `textToLatin(unEmoji(inputText))` * @param {string} inputText - The text which you would like to convert to latin * @returns the input text, with foreign or special alphabet letters translated * to latin lower case characters */ function textToLatin(inputText) { inputText = removeTextAccents(inputText); inputText = unEmoji(inputText); // replace upper case letters that look different in lower case inputText = inputText .replace(/Ð/g, 'd') .replace(/ÎĢ/g, 'e') .replace(/Ⴙ/g, 'g') .replace(/Η/g, 'h') .replace(/Ꮰ/g, 'j') .replace(/Μ/g, 'm') .replace(/áŽĪ/g, 'q') .replace(/Շ/g, 't') .replace(/Ôą/g, 'u') .replace(/Հ/g, 'z'); // convert to lower case and replace the rest inputText = inputText.toLowerCase(); return inputText .replace(/áī€|ₐ|áīŽ|áĩƒ|Îą|âąĨ|ɐ|áķ|Ṛ|𝐚|𝐀|𝑎|ðī|ð‘Ļ|𝒂|𝖚|𝖠|ⓐ|𝔞|𝔄|𝖆|𝕎|𝓊|𝓐|ð’ķ|𝒜|𝕒|ð”ļ|ï―|🄰|ɑ|āļ„|ð—Ū|𝗔|ð˜Ē|𝘈|𝙖|𝘞|𝚊|𝙰|Îŧ|â‚ģ|卂|ïū‘|á—Đ/g, 'a') .replace(/áī|ÃĶ|áī‚/g, 'ae') .replace(/ęœĩ/g, 'ao') .replace(/ęœ·/g, 'au') .replace(/ęœđ/g, 'av') .replace(/ęœŧ|ęœģ/g, 'aa') .replace(/ęœ―/g, 'ay') .replace(/ʙ|áīƒ|Ðē|áĩĶ|áīŪ|áĩ‡|ɓ|ƀ|ƃ|áĩŽ|áķ€|𝐛|𝐁|𝒃|ð‘Đ|ð–ŧ|ð–Ą|ⓑ|𝔟|𝔅|𝖇|𝕭|ð“Ŧ|𝓑|𝒷|𝑏|ðĩ|𝕓|ð”đ|ï―‚|ðŸ„ą|℮|ϐ|āđ’|âī†|ÉŪ|áļ|჊|āđ–|ð—Ŋ|𝗕|ð˜Ģ|𝘉|𝙗|ð˜―|𝚋|ð™ą|ß|āļŋ|äđƒ|Ō|Îē|ᗷ/g, 'b') .replace(/áī„|áķœ|ÂĒ|ƈ|Čž|ꞇ|ð’ļ|ɕ|áķ—|ɔ|ↄ|áī|𝐜|𝐂|𝑐|ðķ|𝒄|𝑊|𝖞|ð–Ē|ⓒ|𝔠|ℭ|𝖈|ð•Ū|𝓎|𝓒|𝒞|𝕔|ℂ|ï―ƒ|ðŸ„ē|áĨī|ðŸ…ē|ς|ፈ|𝗰|𝗖|ð˜Ī|𝘊|𝙘|ð˜ū|𝚌|ð™ē|ᄃ|â‚ĩ|匚|â†ŧ|ᑕ|á‘Ē|℃/g, 'c') .replace(/áī…|áī°|áĩˆ|∂|Ꝺ|ɗ|đ|ƌ|ð’đ|ČĄ|áķ‘|áĩ­|áķ|ɖ|𝐝|𝐃|𝒅|ð‘Ŧ|ð–―|ð–Ģ|𝑑|𝐷|ⓓ|ð”Ą|𝔇|𝖉|ð•Ŋ|𝓭|𝓓|𝒟|𝕕|ð”ŧ|ï―„|ðŸ„ģ|ᗩ|áĶ”|ÕŠ|ðŸ…ģ|āđ”|ԃ|ęŪ„|āŧ“|ð—ą|𝗗|ð˜Ĩ|𝘋|𝙙|ð˜ŋ|𝚍|ð™ģ|り|ę­°|á•ē/g, 'd') .replace(/Įģ|Į†/g, 'dz') .replace(/áī‡|ₑ|áīą|áĩ‰|âąŧ|ɛ|Ņ”|ɇ|Éĩ|ęœŋ|Į|âąļ|áķ’|áķ“|ɘ|𝐞|𝐄|𝒆|𝑎|ð–ū|ð–Ī|ⓔ|ð”Ē|𝔈|𝖊|𝕰|ð“Ū|𝓔|𝑒|ðļ|𝕖|𝔞|ï―…|ðŸ„ī|ðŸ…ī|Ō―|ęŪ›|ð—ē|𝗘|ð˜Ķ|𝘌|𝙚|𝙀|𝚎|ð™ī|äđ‡|Îū|á˜ŋ|₮|á—ī/g, 'e') .replace(/ꜰ|áķ |ƒ|ʄ|ꝭ|ð’ŧ|Ꝿ|áĩŪ|áķ‚|Ṝ|᚝|Ō“|áĩģ|𝐟|𝐅|𝒇|𝑭|ð–ŋ|ð–Ĩ|ęŸŧ|ⓕ|ð”Ģ|𝔉|𝖋|ð•ą|ð“Ŋ|𝓕|𝑓|ðđ|𝕗|ð”―|ï―†|ðŸ„ĩ|ℱ|𝒇|á ŧ|âĻ|ðŸ…ĩ|ϝ|ð—ģ|𝗙|𝘧|𝘍|𝙛|𝙁|𝚏|ð™ĩ|ÂĢ|â‚Ģ|ę­ķ|千|ï―·|á–ī/g, 'f') .replace(//g, 'fi') .replace(//g, 'fl') .replace(//g, 'ff') .replace(//g, 'ffi') .replace(//g, 'ffl') .replace(/ÉĒ|ʛ|áīģ|áĩ|É |ĮĨ|áķƒ|ÉĄ|áĩ·|𝓰|𝐠|𝐆|𝐚|𝒈|ð‘Ū|𝗀|ð–Ķ|ⓖ|ð”Ī|𝔊|𝖌|ð•ē|𝓖|𝑔|ð’Ē|𝕘|ð”ū|ï―‡|ðŸ„ķ|Ų‚|ᧁ|ðŸ…ķ|ïŧŪ|ęŪ†|āš‡|ð—ī|𝗚|ð˜Ļ|𝘎|𝙜|𝙂|𝚐|ð™ķ|â‚ē|ム|ց|ᘜ/g, 'g') .replace(/ʜ|Ð―|ₕ|áīī|Ę°|âąĻ|ħ|ÉĶ|ÉĨ|ĘŪ|ĘŊ|ðĄ|𝐇|𝒉|ð‘Ŋ|𝗁|𝖧|ⓗ|ð”Ĩ|ℌ|𝖍|ð•ģ|ð“ą|𝓗|ð’―|ðŧ|𝕙|ℍ|ï―ˆ|🄷|ℋ|🅷|Ņ’|ԋ|ęŪ’|ɧ|ð—ĩ|𝗛|ð˜Đ|𝘏|𝙝|𝙃|𝚑|𝙷|卄|ん|Õ°|ᕾ/g, 'h') .replace(/ƕ/g, 'hu') .replace(/ÉŠ|áĩĒ|áīĩ|áķĶ|⁹|Îđ|Äą|ÉĻ|áķ–|áī‰|ðĒ|𝐈|𝒊|𝑰|𝗂|ð–Ļ|ðšĪ|ⓘ|ð”Ķ|ℑ|𝖎|ð•ī|ð“ē|𝓘|ð’ū|𝑖|𝐞|𝕚|𝕀|ï―‰|ðŸ„ļ|ę ļ|ðŸ…ļ|āđ€|ę­ĩ|ð—ķ|𝗜|𝘊|𝘐|𝙞|𝙄|𝚒|ð™ļ|äļĻ|ïū‰|ᓰ|áķĪ/g, 'i') .replace(/Äģ/g, 'ij') .replace(/áīŠ|âąž|áīķ|Ęē|ʝ|ɉ|Č·|ɟ|ðĢ|𝐉|𝑗|ð―|𝒋|ð‘ą|𝔧|𝗃|ð–Đ|ðšĨ|ⓙ|𝔧|𝔍|𝖏|ð•ĩ|ð“ģ|𝓙|ð’ŋ|ð’Ĩ|𝕛|𝕁|ï―Š|ðŸ„đ|ℐ|Ņ˜|ę đ|Õĩ|ðŸ…đ|āļ§|𝗷|𝗝|ð˜Ŧ|𝘑|𝙟|𝙅|𝚓|ð™đ|Ũ |ïūŒ|Ų„|ᒍ|ᒚ|Úķ|ᒎ/g, 'j') .replace(/áī‹|К|ₖ|áī·|áĩ|⹊|ꝃ|ƙ|ꝁ|ꝅ|áķ„|ʞ|ðĪ|𝐊|𝑘|ðū|𝒌|ð‘ē|𝗄|𝖊|ⓚ|ð”Ļ|𝔎|𝖐|ð•ķ|ð“ī|𝓚|𝓀|ð’Ķ|𝕜|𝕂|ï―‹|🄚|Ō|ᛕ|🅚|ӄ|ęŪķ|ð—ļ|𝗞|𝘎|𝘒|𝙠|𝙆|𝚔|𝙚|₭|ã‚đ|ŌŸ|á–―áļ|ŌĄ/g, 'k') .replace(/ʟ|áīŒ|ₗ|áīļ|ËĄ|ℓ|⹥|ꝉ|ŀ|ł|ꞁ|ÉŽ|Čī|áķ…|É­|Åŋ|Éŋ|ŨŸ|ðĨ|𝐋|𝒍|ð‘ģ|𝗅|ð–Ŧ|⅃|ⓛ|ð”Đ|𝔏|𝖑|𝕷|ð“ĩ|𝓛|𝓁|𝑙|ðŋ|𝕝|𝕃|ï―Œ|ðŸ„ŧ|ℒ|áĨĢ|ðŸ…ŧ|ʅ|ęŪ­|ÉĐ|ð—đ|𝗟|𝘭|𝘓|ð™Ą|𝙇|𝚕|ð™ŧ|ᄂ|ã„Ĩ|ïūš|ӏ|ᒩ/g, 'l') .replace(/Į‰/g, 'lj') .replace(/áī|О|ₘ|áīđ|áĩ|Éą|áĩŊ|áķ†|É°|ðĶ|𝐌|𝒎|ð‘ī|𝗆|𝖎|ⓜ|𝔊|𝔐|𝖒|ð•ļ|ð“ķ|𝓜|𝓂|𝑚|𝑀|𝕞|𝕄|ï―|🄞|â„ģ|ᗰ|🅞|āđ“|ęŪ‡|𝗚|𝗠|ð˜Ū|𝘔|ð™Ē|𝙈|𝚖|𝙞|â‚Ĩ|įˆŠ|ïūķ|á˜ŧ|Ïŧ/g, 'm') .replace(/Éī|ₙ|áīš|âŋ|áīŽ|Ðļ|ꞃ|Éē|ƞ|Čĩ|áĩ°|áķ‡|Éģ|áī|𝐧|𝐍|𝑛|𝑁|𝒏|ð‘ĩ|𝗇|𝖭|ⓝ|ð”Ŧ|𝔑|𝖓|ð•đ|𝓷|𝓝|𝓃|ð’Đ|𝕟|ℕ|ï―Ž|ðŸ„―|ðŸ…―|āļ |Õž|ęŪ‘|ŋ|āš–|ð—ŧ|ð—Ą|ð˜Ŋ|𝘕|ð™Ģ|𝙉|𝚗|ð™―|Ðŋ|η|â‚Ķ|几|刀|Õē|ហ|ᑎ|ᘉ|áķ°/g, 'n') .replace(/ĮŒ/g, 'nj') .replace(/áī|ₒ|áīž|áĩ’|σ|ꝋ|ꝍ|Ãļ|ČĢ|âąš|áī‘|áī“|ðĻ|𝐎|𝑂|𝒐|ð‘ķ|𝗈|ð–Ū|ⓞ|𝔎|𝔒|𝖔|𝕚|ð“ļ|𝓞|𝑜|𝒊|𝕠|𝕆|ï―|ę­ī|ðŸ„ū|❀|āđ|օ|ę­·|āŧ|𝗞|ð—Ē|𝘰|𝘖|ð™Ī|𝙊|𝚘|ð™ū|ÓĐ|ð|ㄖ|ãŪ|â™Ē|āķ§|ᓍ|♡|ðŸĩ|ᗝ|Îŋ|❁|â˜Ŋ/g, 'o') .replace(/ꝏ/g, 'oo') .replace(/Éķ|œ|áī”/g, 'oe') .replace(/ÆĢ/g, 'on') .replace(/áī˜|ₚ|áīū|áĩ–|ρ|ꝓ|ÆĨ|ꝕ|áĩ―|ꝑ|áĩą|áķˆ|ðĐ|𝐏|𝑝|𝑃|ꟾ|𝒑|𝑷|𝗉|ð–Ŋ|ⓟ|𝔭|𝔓|𝖕|ð•ŧ|ð“đ|𝓟|𝓅|ð’Ŧ|ð•Ą|ℙ|ï―|ðŸ„ŋ|℘|Ũ§|ք|ę­ū|ð—―|ð—Ģ|ð˜ą|𝘗|ð™Ĩ|𝙋|𝚙|ð™ŋ|Ãū|₱|åĐ|ï―ą|φ|ᑭ|á•ĩ/g, 'p') .replace(/ĮŦ|áĩĐ|áĩ |ꝙ|ꝗ|Ę |ɋ|𝐊|𝐐|𝒒|ð‘ļ|𝗊|𝖰|ⓠ|ð”Ū|𝔔|𝖖|𝕞|𝓚|𝓠|𝓆|𝒎|ð•Ē|ℚ|ï―‘|🅀|𝑞|𝑄|🆀|ϙ|ÕĶ|āđ‘|ð—ū|ð—Ī|ð˜ē|𝘘|ð™Ķ|𝙌|𝚚|𝚀|ゐ|ŌĐ|á‘Ŧ|á•ī/g, 'q') .replace(/ʀ|áĩĢ|áīŋ|Ęģ|áī•|áīš|ʁ|áī™|Ņ|ꞅ|ɍ|É―|Éū|Éž|áĩē|áķ‰|Éđ|Éŧ|Éš|âąđ|ęŪ§|ðŦ|𝐑|𝒓|ð‘đ|𝗋|ð–ą|ⓡ|ð”Ŋ|ℜ|𝖗|ð•―|ð“ŧ|ð“Ą|𝓇|𝑟|𝑅|ð•Ģ|ℝ|ï―’|🅁|ℛ|ꭈ|🆁|Ðģ|ęŪĒ|ā―ž|ð—ŋ|ð—Ĩ|ð˜ģ|𝘙|𝙧|𝙍|𝚛|𝚁|å°š|ā― |ᖇ/g, 'r') .replace(/ꜱ|ₛ|ËĒ|ʂ|áĩī|áķŠ|Čŋ|ęļ|𝐎|𝐒|𝑠|𝑆|𝒔|𝑚|𝗌|ð–ē|â“Ē|ꙅ|𝔰|𝔖|𝖘|ð•ū|𝓞|ð“Ē|𝓈|ð’Ū|ð•Ī|𝕊|ï―“|🅂|áĶ“|ęŊą|🆂|āļĢ|ֆ|ęŪĨ|𝘀|ð—Ķ|ð˜ī|𝘚|ð™Ļ|𝙎|𝚜|𝚂|ÆĻ|Ņ•|§|â‚ī|äļ‚|ϛ|ᔕ/g, 's') .replace(//g, 'st') .replace(/áī›|Ņ‚|ₜ|áĩ€|áĩ—|ƚ|ÉŦ|âąĶ|Æ­|ʈ|ŧ|Čķ|áĩĩ|ÆŦ|ʇ|𝐭|𝐓|𝒕|ð‘ŧ|𝗍|ð–ģ|â“Ģ|ð”ą|𝔗|𝖙|ð•ŋ|ð“―|ð“Ģ|𝓉|ð’Ŋ|ð•Ĩ|𝕋|ï―”|🅃|𝑇|🆃|ęŪĶ|𝘁|𝗧|ð˜ĩ|ð‘Ą|𝘛|ð™Đ|𝙏|𝚝|𝚃|†|â‚Ū|ㄒ|ï―ē|Õ§|Íģ|á–ķ|äļ…/g, 't') .replace(/áĩš/g, 'th') .replace(/ęœĐ/g, 'tz') .replace(/áīœ|áĩĪ|áĩ|áĩ˜|ʋ|áķ™|ðŪ|𝐔|ð‘Ē|𝑈|𝒖|𝑞|𝗎|ð–ī|â“Ī|ð”ē|𝔘|𝖚|𝖀|ð“ū|ð“Ī|𝓊|𝒰|ð•Ķ|𝕌|ï―•|🅄|υ|🆄|āļĒ|ʊ|ęŪž|ęŪ°|āļ™|𝘂|ð—Ļ|ð˜ķ|𝘜|𝙊|𝙐|𝚞|𝚄|Ņ†|Âĩ|Ξ|ʉ|ã„Đ|ãē|Õī|ᑌ|ᑘ/g, 'u') .replace(/áĩŦ/g, 'ue') .replace(/áī |áĩĨ|âą―|áĩ›|Î―|ʌ|ꝟ|âąī|áķŒ|âąą|ðŊ|𝐕|ð‘Ģ|𝑉|𝒗|ð‘―|𝗏|ð–ĩ|â“Ĩ|ð”ģ|𝔙|𝖛|𝖁|ð“ŋ|ð“Ĩ|𝓋|ð’ą|𝕧|𝕍|ï―–|🅅|🆅|ŨĐ|ęŪ™|Û·|āļ‡|𝘃|ð—Đ|𝘷|𝘝|ð™Ŧ|𝙑|𝚟|𝚅|áŊ|√|Ņĩ|áŧ―|ᐚ/g, 'v') .replace(/ꝡ/g, 'vy') .replace(/áīĄ|áĩ‚|Ę·|ω|âąģ|ʍ|ÉŊ|𝐰|𝐖|ð‘Ī|𝑊|𝒘|ð‘ū|𝗐|ð–ķ|â“Ķ|ð”ī|𝔚|𝖜|𝖂|𝔀|ð“Ķ|𝓌|ð’ē|ð•Ļ|𝕎|ï―—|🅆|āļŽ|🆆|ÕĄ|ęŪ—|āšŸ|𝘄|𝗊|ð˜ļ|𝘞|𝙎|𝙒|𝚠|𝚆|Ņ‰|â‚Đ|åąą|ā°š|á—Ŋ|ᘚ/g, 'w') .replace(/χ|ₓ|ËĢ|áķ|ðą|𝐗|ð‘Ĩ|𝑋|𝒙|ð‘ŋ|𝗑|𝖷|ⓧ|ð”ĩ|𝔛|𝖝|𝖃|𝔁|𝓧|𝓍|ð’ģ|ð•Đ|𝕏|ï―˜|🅇|🆇|Ũ|Ó―|Ōģ|𝘅|ð—Ŧ|ð˜đ|𝘟|𝙭|𝙓|ðšĄ|𝚇|×|Óŋ|äđ‚|ïū’|áƒŊ|᙭|Ðķ|ጀ/g, 'x') .replace(/ʏ|áĩ§|Ęļ|Ņ‡|Æī|áŧŋ|ɏ|ʎ|ðē|𝐘|ð‘Ķ|𝑌|𝒚|𝒀|𝗒|ð–ļ|â“Ļ|ð”ķ|𝔜|𝖞|𝖄|𝔂|ð“Ļ|𝓎|ð’ī|𝕊|𝕐|ï―™|🅈|âī˜|ŌŊ|🆈|ŨĨ|ყ|ę­đ|āļŊ|𝘆|𝗎|𝘚|𝘠|ð™Ū|𝙔|ðšĒ|𝚈|Ņƒ|ÂĨ|ㄚ|ïū˜|Õū|ӌ|á–ŧ/g, 'y') .replace(/áīĒ|áķŧ|ęŦ|áĩđ|⹎|ČĨ|Æķ|ʑ|áĩķ|áķŽ|ʐ|ɀ|ðģ|𝐙|𝑧|𝑍|𝒛|𝒁|𝗓|ð–đ|â“Đ|𝔷|â„Ļ|𝖟|𝖅|𝔃|ð“Đ|𝓏|ð’ĩ|ð•Ŧ|â„Ī|ï―š|🅉|Æš|🆉|Õđ|ፚ|āšŠ|𝘇|𝗭|ð˜ŧ|ð˜Ą|ð™Ŋ|𝙕|ðšĢ|𝚉|äđ™|ᘔ|ᗱ|Õ·/g, 'z'); } exports.textToLatin = textToLatin; /** * For any given input text, reduce any repeating characters to a given maximum amount of repetitions. * * As an example, the input string: `"heeellllooooo"` becomes: `"heelloo"` if that number is 2, or * `"heeelllooo"` if that number is 3, or `helo` if that number is 1. * * For English, it is recommended to not use values lower than 2. If this preprocessing is used, make * sure that the bad words and allowed terms also feature at most the same number of repeated characters. * I.e. if using this with the number "2", there is no use of putting words like "helllo" on the lists. * @param inputText - The text from which to remove repeat characters. * @param maxAllowedCharsInSequence - The maximum number of characters in sequence (such as "aaa", "bbb", * ...) that are allowed to remain in the input string. `Recommended: 2 or 3`, depending on the language * of your input text. * @returns The input text with all repeat characters that occur more than the max amount in sequence * removed. * @throws If `maxAllowedCharsInSequence` is not an integer (such as 1.5) or if it is 0 or less. */ function reduceRepeatCharacters(inputText, maxAllowedCharsInSequence) { if (maxAllowedCharsInSequence === void 0) { maxAllowedCharsInSequence = 2; } if (!Number.isInteger(maxAllowedCharsInSequence) || maxAllowedCharsInSequence < 1) { throw new Error('reduceRepeatCharacters - maxAllowedCharsInSequence needs to be an integer that is larger than 0'); } // match any word character `(\w)`, then refer to the last match `\1` and check if it appears // n or more times `{n,}` (where n is maxAllowedCharsInSequence) // replace this with our matched character '$1' repeated n times // in normal regexp terms this would be `string.replace(/(\w)\1{2,}/g, '$1$1')` if `n = 2` var regexp = new RegExp('(\\w)\\1{' + maxAllowedCharsInSequence + ',}', 'g'); var replacementStr = '$1'.repeat(maxAllowedCharsInSequence); return inputText.replace(regexp, replacementStr); } exports.reduceRepeatCharacters = reduceRepeatCharacters; //# sourceMappingURL=input_preprocessor.js.map