Cut and paste string cleaning function with diacritics and allowed characters

string-cleaner-function.js

const get_string_sanitizer = (user_maps = {}, user_allowed ) => {
    const build_map = (map_str) => map_str.split(',').map(c => c.split('|')).reduce((dm, [c, r])=> c.split('').reduce((dm, c) => ({...dm, [c]: r.replace('\\u002C', ',')}), {...dm}), {})
    const map_strings = {
        diacritics: '⁰₀⓪⓿０|0,¹₁①⓵❶➀➊１|1,²₂②⓶❷➁➋２|2,³₃③⓷❸➂➌３|3,⁴₄④⓸❹➃➍４|4,⁵₅⑤⓹❺➄➎５|5,⁶₆⑥⓺❻➅➏６|6,⁷₇⑦⓻❼➆➐７|7,⁸₈⑧⓼❽➇➑８|8,⁹₉⑨⓽❾➈➒９|9,⑩⓾❿➉➓|10,⑪⓫|11,⑫⓬|12,⑬⓭|13,⑭⓮|14,⑮⓯|15,⑯⓰|16,⑰⓱|17,⑱⓲|18,⑲⓳|19,⑳⓴|20,ΑÀÁÂÃÄÅĀĂĄƏǍǞǠǺȀȂȦȺᴀḀẠẢẤẦẨẪẬẮẰẲẴẶⒶＡ|A,ÆǢǼᴁ|AE,ΒƁƂɃʙᴃḂḄḆⒷＢ|B,ÇĆĈĊČƇȻʗᴄḈⒸＣ|C,ÈΕÉÊËĒĔĖĘĚƎƐȄȆȨɆᴇḔḖḘḚḜẸẺẼẾỀỂỄỆⒺⱻＥ|E,ÌÍÎÏĨĪĬĮİƖƗǏȈȊɪᵻḬḮỈỊⒾꟾＩΙ|I,ÐĎĐƉƊƋᴅᴆḊḌḎḐḒⒹꝹＤ|D,ÑŃŅŇŊƝǸȠɴᴎṄṆṈṊⓃＮΝ|N,ÒÓÔÕÖØŌŎŐƆƟƠǑǪǬǾȌȎȪȬȮȰᴏᴐṌṎṐṒỌỎỐỒỔỖỘỚỜỞỠỢⓄꝊꝌＯΟ|O,ÙÚÛÜŨŪŬŮŰŲƯǓǕǗǙǛȔȖɄᴜᵾṲṴṶṸṺỤỦỨỪỬỮỰⓊＵ|U,ÝŶŸƳȲɎʏẎỲỴỶỸỾⓎＹΥ|Y,ÞꝦ|TH,ß|ss,àáâãäåāăąǎǟǡǻȁȃȧɐəɚᶏᶕḁẚạảấầẩẫậắằẳẵặₐₔⓐⱥⱯａ|a,æǣǽᴂ|ae,çćĉċčƈȼɕḉↄⓒꜾꜿｃ|c,èéêëēĕėęěǝȅȇȩɇɘɛɜɝɞʚᴈᶒᶓᶔḕḗḙḛḝẹẻẽếềểễệₑⓔⱸｅ|e,ìíîïĩīĭįıǐȉȋɨᴉᵢᵼᶖḭḯỉịⁱⓘｉ|i,ðďđƌȡɖɗᵭᶁᶑḋḍḏḑḓⓓꝺｄ|d,ñńņňʼnŋƞǹȵɲɳᵰᶇṅṇṉṋⁿⓝｎ|n,òóôõöøōŏőơǒǫǭǿȍȏȫȭȯȱɔɵᴖᴗᶗṍṏṑṓọỏốồổỗộớờởỡợₒⓞⱺꝋꝍｏ|o,ùúûüũūŭůűųưǔǖǘǚǜȕȗʉᵤᶙṳṵṷṹṻụủứừửữựⓤｕ|u,ýÿŷƴȳɏʎẏẙỳỵỷỹỿⓨｙ|y,þᵺꝧ|th,ĜĞĠĢƓǤǥǦǧǴɢʛḠⒼꝽꝾＧ|G,ĝğġģǵɠɡᵷᵹᶃḡⓖꝿｇ|g,ĤĦȞʜḢḤḦḨḪⒽⱧⱵＨΗ|H,ĥħȟɥɦʮʯḣḥḧḩḫẖⓗⱨⱶｈ|h,IJ|IJ,ij|ij,ĴɈᴊⒿＪ|J,ĵǰȷɉɟʄʝⓙⱼｊ|j,ĶƘǨᴋḰḲḴⓀⱩꝀꝂꝄＫΚ|K,ķƙǩʞᶄḱḳḵⓚⱪꝁꝃꝅｋ|k,ĸɋʠⓠꝗꝙｑ|q,ĹĻĽĿŁȽʟᴌḶḸḺḼⓁⱠⱢꝆꝈꞀＬ|L,ĺļľŀłƚȴɫɬɭᶅḷḹḻḽⓛⱡꝇꝉꞁｌ|l,Œɶ|OE,œᴔ|oe,ŔŖŘȐȒɌʀʁᴙᴚṘṚṜṞⓇⱤꝚꞂＲ|R,ŕŗřȑȓɍɼɽɾɿᵣᵲᵳᶉṙṛṝṟⓡꝛꞃｒ|r,ŚŜŞŠȘṠṢṤṦṨⓈꜱꞅＳ|S,śŝşšſșȿʂᵴᶊṡṣṥṧṩẜẝⓢꞄｓ|s,ŢŤŦƬƮȚȾᴛṪṬṮṰⓉꞆＴΤ|T,ţťŧƫƭțȶʇʈᵵṫṭṯṱẗⓣⱦｔ|t,ŴǷᴡẀẂẄẆẈⓌⱲＷ|W,ŵƿʍẁẃẅẇẉẘⓦⱳｗ|w,ŹŻŽƵȜȤᴢẐẒẔⓏⱫꝢＺΖ|Z,źżžƶȝȥɀʐʑᵶᶎẑẓẕⓩⱬꝣｚ|z,ƀƃɓᵬᶀḃḅḇⓑｂ|b,ƑḞⒻꜰꝻꟻＦ|F,ƒᵮᶂḟẛⓕꝼｆ|f,ƕ|hv,ƜᴍḾṀṂⓂⱮꟽꟿＭΜ|M,ƤᴘṔṖⓅⱣꝐꝒꝔＰΡ|P,ƥᵱᵽᶈṕṗⓟꝑꝓꝕꟼｐ|p,ƲɅᴠṼṾỼⓋꝞꝨＶ|V,DŽDZ|DZ,DžDz|Dz,dždzʣʥ|dz,LJ|LJ,Lj|Lj,lj|lj,NJ|NJ,Nj|Nj,nj|nj,Ƕ|HV,Ȣᴕ|OU,ȣ|ou,ȸ|db,ȹ|qp,ɊⓆꝖꝘＱ|Q,ɯɰɱᵯᶆḿṁṃⓜｍ|m,ʋʌᵥᶌṽṿⓥⱱⱴꝟｖ|v,ʦ|ts,ʨ|tc,ʪ|ls,ʫ|lz,ᵫ|ue,ᶍẋẍₓⓧｘ|x,ẊẌⓍＸΧ|X,ẞ|SS,Ỻ|LL,ỻ|ll,⑴|(1),⑵|(2),⑶|(3),⑷|(4),⑸|(5),⑹|(6),⑺|(7),⑻|(8),⑼|(9),⑽|(10),⑾|(11),⑿|(12),⒀|(13),⒁|(14),⒂|(15),⒃|(16),⒄|(17),⒅|(18),⒆|(19),⒇|(20),⒈|1.,⒉|2.,⒊|3.,⒋|4.,⒌|5.,⒍|6.,⒎|7.,⒏|8.,⒐|9.,⒑|10.,⒒|11.,⒓|12.,⒔|13.,⒕|14.,⒖|15.,⒗|16.,⒘|17.,⒙|18.,⒚|19.,⒛|20.,⒜|(a),⒝|(b),⒞|(c),⒟|(d),⒠|(e),⒡|(f),⒢|(g),⒣|(h),⒤|(i),⒥|(j),⒦|(k),⒧|(l),⒨|(m),⒩|(n),⒪|(o),⒫|(p),⒬|(q),⒭|(r),⒮|(s),⒯|(t),⒰|(u),⒱|(v),⒲|(w),⒳|(x),⒴|(y),⒵|(z),Ꜩ|TZ,ꜩ|tz,Ꜳ|AA,ꜳ|aa,Ꜵ|AO,ꜵ|ao,Ꜷ|AU,ꜷ|au,ꜸꜺ|AV,ꜹꜻ|av,Ꜽ|AY,ꜽ|ay,Ꝏ|OO,ꝏ|oo,Ꝡ|VY,ꝡ|vy,ff|ff,fi|fi,fl|fl,ffi|ffi,ffl|ffl,st|st',
        alt: '«»“”„″‶❝❞❮❯＂|",‐‑‒–—⁻₋－|-,‘’‚‛′‵‹›❛´❜＇|\u0027,‸＾|^,‼|!!,⁄／|/,⁅❲［|[,⁆❳］|],⁇|??,⁈|?!,⁉|!?,⁎＊|*,⁏；|;,⁒％|%,⁓～˜|~,⁺₊＋|+,⁼₌＝|=,⁽₍❨❪（⟨|(,⁾₎❩❫）⟩|),❬❰＜|<,❭❱＞|>,❴｛|{,❵｝|},⸨|((,⸩|)),！|!,＃|#,＄|$,＆|&,，|\\u002C,¸．|.,：|:,？|?,＠|@,＼|\u005C,＿|_',
    };
    
    const allowed_strings = [
        '!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~', // querty
        '¢£¤¥¦§©¬­®°±µ·º¼½¾×÷ΓΔΘΛΞΠΣΦΨΩαβθπφ†‡•…‰€™←↑→↓↔⇐⇒⇔−∗∝∼≅≈≠≡≤≥⋅◊', // other alt characters
    ]
    const allowed_map = (user_allowed && user_allowed.length)
    ? user_allowed.split('').reduce((map, s) => ({ ...map, [s]: s }), {})
    : allowed_strings.flat().join('').split('').reduce((map, s) => ({ ...map, [s]: s }), {});

    const maps = { ...Object.keys(map_strings).reduce((maps, key) => ({...maps, [key]: build_map(map_strings[key])}), {}), ...user_maps };
    const  api = (string) => { return {
        string,
        diacritics: () => api(string.replace(/[^u0000-u007E]/g, (a) => (maps.diacritics[a] || a))),
        alt: () => api(string.replace(/[^u0000-u007E]/g, (a) => (maps.alt[a] || a))),
        include: () => api(string.replace(/[^u0000-u007E]/g, (a) => (allowed_map[a]) ? allowed_map[a] : (a.match(/[\w\s]+/)) ? a : '')),
    } }
    return api;
}

This could possibly be replaced in many circumstances and simplified using string.normalize , which takes the following "modes" demonstrated below where const junk = "ŔŖŘȐȒɌʀʁᴙᴚṘṚṜṞⓇⱤꝚꞂＲ";

NFC -- Canonical Decomposition, followed by Canonical Composition.
junk.normalize('NFC').replace(/[\u0300-\u036f]/g, ''; // ŔŖŘȐȒɌʀʁᴙᴚṘṚṜṞⓇⱤꝚꞂＲ none replaced

NFKC -- Compatibility Decomposition, followed by Canonical Composition.
junk.normalize('NFKC').replace(/[\u0300-\u036f]/g, ''; // ŔŖŘȐȒɌʀʁᴙᴚṘṚṜṞRⱤꝚꞂR 2 replaced, where ŔŖŘȐȒɌʀʁᴙᴚṘṚṜṞⱤꝚꞂ are unmatched

NFD -- Canonical Decomposition.
junk.normalize('NFD').replace(/[\u0300-\u036f]/g, ''; // RRRRRɌʀʁᴙᴚRRRRⓇⱤꝚꞂＲ 9 replaced, where ɌʀʁᴙᴚⓇⱤꝚꞂＲ are unmatched

NFKD -- Compatibility Decomposition.
junk.normalize('NFKD').replace(/[\u0300-\u036f]/g, ''; // RRRRRɌʀʁᴙᴚRRRRRⱤꝚꞂR with 8 replaced, where ɌʀʁᴙᴚⱤꝚꞂ are unmatched

It looks like there are many attempts in this stackoverflow post, where there may be a better solution for the diacritics portion