regex - npm Package Compare versions

Comparing version 1.1.0 to 1.1.1

dist/regex.min.js

@@ -1,2 +0,2 @@
-var Regex=(()=>{var h=Object.defineProperty;var v=Object.getOwnPropertyDescriptor;var B=Object.getOwnPropertyNames;var X=Object.prototype.hasOwnProperty;var Z=(e,t)=>{for(var r in t)h(e,r,{get:t[r],enumerable:!0})},z=(e,t,r,n)=>{if(t&&typeof t=="object"||typeof t=="function")for(let o of B(t))!X.call(e,o)&&o!==r&&h(e,o,{get:()=>t[o],enumerable:!(n=v(t,o))||n.enumerable});return e};var J=e=>z(h({},"__esModule",{value:!0}),e);var se={};Z(se,{default:()=>oe,make:()=>W,partial:()=>_,regex:()=>P});var N=class{#e;constructor(t){this.#e=t}toString(){return String(this.#e)}};function _(e,...t){if(Array.isArray(e?.raw))return new N(e.raw.flatMap((r,n)=>n<e.raw.length-1?[r,t[n]]:r).join(""));if(!t.length)return new N(e??"");throw new Error(`Unexpected arguments: ${JSON.stringify([e,...t])}`)}var s={DEFAULT:"DEFAULT",CHAR_CLASS:"CHAR_CLASS",GROUP_NAME:"GROUP_NAME",ENCLOSED_TOKEN:"ENCLOSED_TOKEN",INTERVAL_QUANTIFIER:"INTERVAL_QUANTIFIER",INVALID_INCOMPLETE_TOKEN:"INVALID_INCOMPLETE_TOKEN"},E={DEFAULT:"CC_DEFAULT",RANGE:"CC_RANGE",ENCLOSED_TOKEN:"CC_ENCLOSED_TOKEN",Q_TOKEN:"CC_Q_TOKEN",INVALID_INCOMPLETE_TOKEN:"CC_INVALID_INCOMPLETE_TOKEN"},g=(()=>{let e=!0;try{new RegExp("(?i-ms:)")}catch{e=!1}return e})(),R="&!#$%*+,.:;<=>?@^`~";function C(e,t){return t===s.CHAR_CLASS?e.replace(new RegExp(String.raw`[()\[\]{}|\\/\-${R}]`,"g"),"\\$&"):e.replace(/[()\[\]{}|\\^$*+?.]/g,"\\$&")}function O(e){return e.replace(new RegExp(String.raw`^([${R}])(?!\1)`),(t,r,n)=>`\\${t}${n+1===e.length?"":t}`)}function x(e){return e.replace(/^\^/,"\\^^")}function L(e,t){return A(e,String.raw`\\0(?!\d)`,"\\u{0}",t)}function K(e,t,r){let n=0;for(let[o]of e.matchAll(new RegExp(`[${C(t+r)}]`,"g")))if(n+=o===t?1:-1,n<0)return r;return n>0?t:""}function G(e,t,r){let n=e.replace(/\\./gsu,"");if(n.at(-1)==="\\")return"\\";if(t===s.DEFAULT)return K(n,"(",")");if(t===s.CHAR_CLASS&&!(r===E.ENCLOSED_TOKEN||r===E.Q_TOKEN))return K(n,"[","]");if(t===s.ENCLOSED_TOKEN||t===s.INTERVAL_QUANTIFIER||r===E.ENCLOSED_TOKEN||r===E.Q_TOKEN){if(n.includes("}"))return"}"}else if(t===s.GROUP_NAME&&n.includes(">"))return">";return""}var m=new RegExp(String.raw`
+var Regex=(()=>{var R=Object.defineProperty;var B=Object.getOwnPropertyDescriptor;var J=Object.getOwnPropertyNames;var X=Object.prototype.hasOwnProperty;var Y=(e,t)=>{for(var r in t)R(e,r,{get:t[r],enumerable:!0})},ee=(e,t,r,n)=>{if(t&&typeof t=="object"||typeof t=="function")for(let o of J(t))!X.call(e,o)&&o!==r&&R(e,o,{get:()=>t[o],enumerable:!(n=B(t,o))||n.enumerable});return e};var te=e=>ee(R({},"__esModule",{value:!0}),e);var le={};Y(le,{default:()=>fe,make:()=>Z,partial:()=>O,regex:()=>b});var A={DEFAULT:"DEFAULT",CHAR_CLASS:"CHAR_CLASS"};function m(e,t,r,n){let o=new RegExp(String.raw`(?<found>${t})|\\?.`,"gsu"),a=0,s="";for(let l of e.matchAll(o)){let{0:u,groups:{found:E}}=l;if(E&&(!n||n===A.DEFAULT==!a)){r instanceof Function?s+=r(l):s+=r;continue}u==="["?a++:u==="]"&&a&&a--,s+=u}return s}function re(e,t,r,n){if(!new RegExp(t,"su").test(e))return!1;let o=new RegExp(String.raw`(?<found>${t})|\\?.`,"gsu"),a=0;for(let s of e.matchAll(o)){let{0:l,groups:{found:u}}=s;if(u&&(!n||n===A.DEFAULT==!a))return r&&r(s),!0;l==="["?a++:l==="]"&&a&&a--}return!1}function K(e,t,r){return re(e,t,null,r)}function k(e){if(!K(e,String.raw`\(\?>`,A.DEFAULT))return e;let t=new RegExp(String.raw`(?<noncapturingStart>\(\?(?:[:=!>A-Za-z\-]|<[=!]))|(?<capturingStart>\((?:\?<[^>]+>)?)|(?<backrefNum>\\[1-9]\d*)|\\?.`,"gsu"),r=3,n,o=0,a=0,s=NaN;do{n=!1;let l=0,u=0,E=!1,p;for(t.lastIndex=Number.isNaN(s)?0:s+r;p=t.exec(e);){let{0:i,index:N,groups:{backrefNum:d,capturingStart:_,noncapturingStart:h}}=p;if(i==="[")l++;else if(l)i==="]"&&l--;else if(i==="(?>"&&!E)s=N,E=!0;else if(E&&h)u++;else if(_)E&&u++,o++;else if(i===")"&&E){if(!u){a++,e=`${e.slice(0,s)}(?:(?=(${e.slice(s+r,N)}))\\k<${a+o}>)${e.slice(N+1)}`,n=!0,o--;break}u--}else if(d)throw new Error(`Invalid decimal escape "${i}" in interpolated regex; cannot be used with atomic group`)}}while(n);return e=m(e,String.raw`\\k<(?<backrefNum>\d+)>`,({groups:{backrefNum:l}})=>`\\${l}`,A.DEFAULT),e}var T=class{#e;constructor(t){this.#e=t}toString(){return String(this.#e)}};function O(e,...t){if(Array.isArray(e?.raw))return new T(e.raw.flatMap((r,n)=>n<e.raw.length-1?[r,t[n]]:r).join(""));if(!t.length)return new T(e??"");throw new Error(`Unexpected arguments: ${JSON.stringify([e,...t])}`)}var f={...A,GROUP_NAME:"GROUP_NAME",ENCLOSED_TOKEN:"ENCLOSED_TOKEN",INTERVAL_QUANTIFIER:"INTERVAL_QUANTIFIER",INVALID_INCOMPLETE_TOKEN:"INVALID_INCOMPLETE_TOKEN"},c={DEFAULT:"CC_DEFAULT",RANGE:"CC_RANGE",ENCLOSED_TOKEN:"CC_ENCLOSED_TOKEN",Q_TOKEN:"CC_Q_TOKEN",INVALID_INCOMPLETE_TOKEN:"CC_INVALID_INCOMPLETE_TOKEN"},S=(()=>{let e=!0;try{new RegExp("(?i-ms:)")}catch{e=!1}return e})(),U="&!#$%*+,.:;<=>?@^`~";function C(e,t){return t===f.CHAR_CLASS?e.replace(new RegExp(String.raw`[()\[\]{}|\\/\-${U}]`,"g"),"\\$&"):e.replace(/[()\[\]{}|\\^$*+?.]/g,"\\$&")}function I(e){return e.replace(new RegExp(String.raw`^([${U}])(?!\1)`),(t,r,n)=>`\\${t}${n+1===e.length?"":t}`)}function V(e){return e.replace(/^\^/,"\\^^")}function w(e,t){return m(e,String.raw`\\0(?!\d)`,"\\u{0}",t)}function G(e,t,r){let n=0;for(let[o]of e.matchAll(new RegExp(`[${C(t+r)}]`,"g")))if(n+=o===t?1:-1,n<0)return r;return n>0?t:""}function M(e,t,r){let n=e.replace(/\\./gsu,"");if(n.at(-1)==="\\")return"\\";if(t===f.DEFAULT)return G(n,"(",")");if(t===f.CHAR_CLASS&&!(r===c.ENCLOSED_TOKEN||r===c.Q_TOKEN))return G(n,"[","]");if(t===f.ENCLOSED_TOKEN||t===f.INTERVAL_QUANTIFIER||r===c.ENCLOSED_TOKEN||r===c.Q_TOKEN){if(n.includes("}"))return"}"}else if(t===f.GROUP_NAME&&n.includes(">"))return">";return""}var g=new RegExp(String.raw`
   (?<groupN> \(\?< (?! [=!] ) | \\k< )
@@ -20,11 +20,11 @@ | (?<enclosedT> \\[pPu]\{ )
 | \[\^
-| \(\?[:=!<>ims\-]
-| (?<dp> [${R}] ) \k<dp>
+| \(\? [:=!<>A-Za-z\-]
+| (?<dp> [${U}] ) \k<dp>
 | \\[1-9]\d*
 | --
 | \\? .
-`.replace(/\s+/g,""),"gsu");function T(e,{regexContext:t=s.DEFAULT,charClassContext:r=E.DEFAULT,charClassDepth:n=0,lastPos:o=0}){m.lastIndex=o;let u;for(;u=m.exec(e);){let{0:i,groups:{groupN:c,enclosedT:l,qT:f,intervalQ:a,incompleteT:p}}=u;i==="["||i==="[^"?(n++,t=s.CHAR_CLASS,r=E.DEFAULT):i==="]"&&t===s.CHAR_CLASS?(n&&n--,n||(t=s.DEFAULT),r=E.DEFAULT):t===s.CHAR_CLASS?p?r=E.INVALID_INCOMPLETE_TOKEN:i==="-"?r=E.RANGE:l?r=E.ENCLOSED_TOKEN:f?r=E.Q_TOKEN:(i==="}"&&(r===E.ENCLOSED_TOKEN||r===E.Q_TOKEN)||r===E.INVALID_INCOMPLETE_TOKEN||r===E.RANGE)&&(r=E.DEFAULT):p?t=s.INVALID_INCOMPLETE_TOKEN:c?t=s.GROUP_NAME:l?t=s.ENCLOSED_TOKEN:a?t=s.INTERVAL_QUANTIFIER:(i===">"&&t===s.GROUP_NAME||i==="}"&&(t===s.ENCLOSED_TOKEN||t===s.INTERVAL_QUANTIFIER)||t===s.INVALID_INCOMPLETE_TOKEN)&&(t=s.DEFAULT)}return{regexContext:t,charClassContext:r,charClassDepth:n,lastPos:e.length}}function A(e,t,r,n){let o=new RegExp(String.raw`(?<found>${t})|\\?.`,"gsu"),u=0,i="";for(let c of e.matchAll(o)){let{0:l,groups:{found:f}}=c;if(f&&(!n||n===s.DEFAULT==!u)){r instanceof Function?i+=r(c):i+=r;continue}l==="["?u++:l==="]"&&u&&u--,i+=l}return i}function k(e,t,r){if(!new RegExp(t,"su").test(e))return!1;let n=new RegExp(String.raw`(?<found>${t})|\\?.`,"gsu"),o=0;for(let u of e.matchAll(n)){let{0:i,groups:{found:c}}=u;if(i==="[")o++;else if(o)i==="]"&&o--;else if(c)return r&&r(u),!0}return!1}function U(e){let t=/(?<capture>\((?:(?!\?)|\?<[^>]+>))|\\?./gsu;return Array.from(e.matchAll(t)).filter(r=>r.groups.capture).length}function V(e,t){return e.replace(/\\([1-9]\d*)|\\?./gsu,(r,n)=>n?"\\"+(Number(n)+t):r)}var Y=["Basic_Emoji","Emoji_Keycap_Sequence","RGI_Emoji_Modifier_Sequence","RGI_Emoji_Flag_Sequence","RGI_Emoji_Tag_Sequence","RGI_Emoji_ZWJ_Sequence","RGI_Emoji"].join("|");function $(e){let t=new RegExp(String.raw`
+`.replace(/\s+/g,""),"gsu");function L(e,{regexContext:t=f.DEFAULT,charClassContext:r=c.DEFAULT,charClassDepth:n=0,lastPos:o=0}){g.lastIndex=o;let a;for(;a=g.exec(e);){let{0:s,groups:{groupN:l,enclosedT:u,qT:E,intervalQ:p,incompleteT:i}}=a;s==="["||s==="[^"?(n++,t=f.CHAR_CLASS,r=c.DEFAULT):s==="]"&&t===f.CHAR_CLASS?(n&&n--,n||(t=f.DEFAULT),r=c.DEFAULT):t===f.CHAR_CLASS?i?r=c.INVALID_INCOMPLETE_TOKEN:s==="-"?r=c.RANGE:u?r=c.ENCLOSED_TOKEN:E?r=c.Q_TOKEN:(s==="}"&&(r===c.ENCLOSED_TOKEN||r===c.Q_TOKEN)||r===c.INVALID_INCOMPLETE_TOKEN||r===c.RANGE)&&(r=c.DEFAULT):i?t=f.INVALID_INCOMPLETE_TOKEN:l?t=f.GROUP_NAME:u?t=f.ENCLOSED_TOKEN:p?t=f.INTERVAL_QUANTIFIER:(s===">"&&t===f.GROUP_NAME||s==="}"&&(t===f.ENCLOSED_TOKEN||t===f.INTERVAL_QUANTIFIER)||t===f.INVALID_INCOMPLETE_TOKEN)&&(t=f.DEFAULT)}return{regexContext:t,charClassContext:r,charClassDepth:n,lastPos:e.length}}function F(e){let t=/(?<capture>\((?:(?!\?)|\?<[^>]+>))|\\?./gsu;return Array.from(e.matchAll(t)).filter(r=>r.groups.capture).length}function y(e,t){return e.replace(/\\([1-9]\d*)|\\?./gsu,(r,n)=>n?"\\"+(Number(n)+t):r)}var ne=["Basic_Emoji","Emoji_Keycap_Sequence","RGI_Emoji_Modifier_Sequence","RGI_Emoji_Flag_Sequence","RGI_Emoji_Tag_Sequence","RGI_Emoji_ZWJ_Sequence","RGI_Emoji"].join("|");function $(e){let t=new RegExp(String.raw`
 \\ (?:
     c [A-Za-z]
-  | p \{ (?<pPropOfStr> ${Y} ) \}
+  | p \{ (?<pPropOfStr> ${ne} ) \}
   | [pP] \{ [^\}]+ \}
@@ -39,5 +39,5 @@ | (?<qPropOfStr> q )
 | .
-  `.replace(/\s+/g,""),"gsu"),r=!1,n;for(let{0:o,groups:u}of e.matchAll(t)){if(u.pPropOfStr||u.qPropOfStr||o==="["&&r)return!0;if(["-","--","&&"].includes(o))r=!1;else if(!["[","]"].includes(o)){if(r||n==="]")return!0;r=!0}n=o}return!1}function D(e,t,r){let n={raw:[]},o=[],u={};return e.raw.forEach((i,c)=>{let l=r(i,{...u,lastPos:0});if(n.raw.push(l.transformed),u=l.runningContext,c<e.raw.length-1){let f=t[c];if(f instanceof N){let a=r(f,{...u,lastPos:0});o.push(_(a.transformed)),u=a.runningContext}else o.push(f)}}),{template:n,values:o}}function M(e){if(!k(e,String.raw`\(\?>`))return e;let t=new RegExp(String.raw`(?<noncapturingStart>\(\?(?:[:=!>]|<[=!]|[ims\-]+:))|(?<capturingStart>\((?:\?<[^>]+>)?)|(?<backrefNum>\\[1-9]\d*)|\\?.`,"gsu"),r=3,n,o=0,u=0,i=NaN;do{n=!1;let c=0,l=0,f=!1,a;for(t.lastIndex=Number.isNaN(i)?0:i+r;a=t.exec(e);){let{0:p,index:d,groups:{backrefNum:I,capturingStart:S,noncapturingStart:w}}=a;if(p==="[")c++;else if(c)p==="]"&&c--;else if(p==="(?>"&&!f)i=d,f=!0;else if(f&&w)l++;else if(S)f&&l++,o++;else if(p===")"&&f){if(!l){u++,e=`${e.slice(0,i)}(?:(?=(${e.slice(i+r,d)}))\\k<${u+o}>)${e.slice(d+1)}`,n=!0,o--;break}l--}else if(I)throw new Error(`Invalid decimal escape "${p}" in interpolated regex; cannot be used with atomic group`)}}while(n);return e=A(e,String.raw`\\k<(?<backrefNum>\d+)>`,({groups:{backrefNum:c}})=>`\\${c}`,s.DEFAULT),e}function y(e,t){e=String(e);let r="",n="";for(let[o]of e.matchAll(m)){r+=o,t=T(r,t);let{regexContext:u}=t;if(u===s.DEFAULT)if(o==="(")n+="(?:";else{if(/^\\[1-9]/.test(o))throw new Error(`Invalid decimal escape "${o}" with implicit flag n; replace with named backreference`);n+=o}else n+=o}return{transformed:n,runningContext:t}}var j=/^\s$/,ee=/^\\[\s#]$/,Q=/^[ \t]$/,te=/^\\[ \t]$/;function H(e,t){e=String(e);let r=!1,n=!1,o=!1,u="",i="",c="",l=!1,f=(a,{prefix:p=!0,postfix:d=!1}={})=>(a=(l&&p?"(?:)":"")+a+(d?"(?:)":""),l=!1,a);for(let[a]of e.matchAll(m)){if(o){a===`
-`&&(o=!1,l=!0);continue}if(r){if(j.test(a))continue;r=!1,l=!0}else if(n){if(Q.test(a))continue;n=!1}u+=a,t=T(u,t);let{regexContext:p,charClassContext:d}=t;if(a==="-"&&p===s.CHAR_CLASS&&c===E.RANGE)throw new Error("Invalid unescaped hyphen as the end value for a range");if(p===s.DEFAULT&&/^[?*+]\??$/.test(a)||p===s.INTERVAL_QUANTIFIER&&a==="{")i+=f(a,{prefix:!1,postfix:a==="?"});else if(p===s.DEFAULT)j.test(a)?r=!0:a.startsWith("#")?o=!0:ee.test(a)?i+=f(a[1],{prefix:!1}):i+=f(a);else if(p===s.CHAR_CLASS&&a!=="["&&a!=="[^")if(Q.test(a)&&(d===E.DEFAULT||d===E.RANGE||d===E.Q_TOKEN))n=!0;else{if(d===E.INVALID_INCOMPLETE_TOKEN)throw new Error(`Invalid incomplete token in character class: "${a}"`);te.test(a)&&(d===E.DEFAULT||d===E.Q_TOKEN)?i+=f(a[1],{prefix:!1}):d===E.DEFAULT?i+=f(O(L(a))):i+=f(a)}else i+=f(a);r||n||o||(c=d)}return{transformed:i,runningContext:t}}function q(e){let t=String.raw`\(\?:\)`;return e=A(e,`${t}(?:${t})+`,"(?:)",s.DEFAULT),e=A(e,String.raw`^${t}(?![?*+{])|${t}$|${t}(?=[()|$\\])|(?<=[()|>^]|\(\?(?:[:=!]|<[=!]))${t}`,"",s.DEFAULT),e}function P(e,...t){let r=this instanceof Function?this:RegExp;if(Array.isArray(e?.raw))return F(r,{flags:""},e,...t);if((typeof e=="string"||e===void 0)&&!t.length)return F.bind(null,r,{flags:e});if({}.toString.call(e)==="[object Object]"&&!t.length)return F.bind(null,r,e);throw new Error(`Unexpected arguments: ${JSON.stringify([e,...t])}`)}function F(e,t,r,...n){let{flags:o="",__flagN:u=!0,__flagX:i=!0,__rake:c=t.__flagX??!0}=t;if(/[vu]/.test(o))throw new Error("Flags v/u cannot be explicitly added since v is always enabled");i&&({template:r,values:n}=D(r,n,H)),u&&({template:r,values:n}=D(r,n,y));let l=0,f="",a={};return r.raw.forEach((p,d)=>{let I=r.raw[d]||r.raw[d+1];l+=U(p),f+=L(p,s.CHAR_CLASS),a=T(f,a);let{regexContext:S,charClassContext:w}=a;if(d<r.raw.length-1){let b=re(n[d],o,S,w,I,l);l+=b.capturesAdded||0,f+=b.value}}),f=M(f),new e(c?q(f):f,`v${o}`)}function re(e,t,r,n,o,u){if(e instanceof RegExp&&r!==s.DEFAULT)throw new Error("Cannot interpolate a RegExp at this position because the syntax context does not match");if(r===s.INVALID_INCOMPLETE_TOKEN||n===E.INVALID_INCOMPLETE_TOKEN)throw new Error("Interpolation preceded by invalid incomplete token");let i=e instanceof N,c;if(!(e instanceof RegExp)){e=String(e),i||(c=C(e,r));let l=G(c||e,r,n);if(l)throw new Error(`Unescaped stray "${l}" in the interpolated value would have side effects outside it`)}if(r===s.ENCLOSED_TOKEN||r===s.INTERVAL_QUANTIFIER||r===s.GROUP_NAME||n===E.ENCLOSED_TOKEN||n===E.Q_TOKEN)return{value:i?e:c};if(r===s.CHAR_CLASS){if(i){if(A(e,"^-|^&&|-$|&&$","")!==e)throw new Error("In character classes, a partial cannot use a range/set operator at its boundary; move the operation into the partial or the operator outside of it");let f=x(O(e));return{value:$(e)?`[${f}]`:L(f)}}return{value:$(c)?`[${c}]`:c}}if(e instanceof RegExp){let l=ne(e,t),f=V(l.value,u);return{value:l.usedModifier?f:`(?:${f})`,capturesAdded:U(e.source)}}return i?{value:`(?:${e})`}:{value:o?`(?:${c})`:c}}function ne(e,t){let r={i:null,m:null,s:null},n="\\n\\r\\u2028\\u2029",o=e.source;if(e.ignoreCase!==t.includes("i"))if(g)r.i=e.ignoreCase;else throw new Error("Pattern modifiers not supported, so the value of flag i on the interpolated RegExp must match the outer regex");if(e.dotAll!==t.includes("s")&&(g?r.s=e.dotAll:o=A(o,"\\.",e.dotAll?"[^]":`[^${n}]`,s.DEFAULT)),e.multiline!==t.includes("m")&&(g?r.m=e.multiline:(o=A(o,"\\^",e.multiline?`(?<=^|[${n}])`:"(?<![^])",s.DEFAULT),o=A(o,"\\$",e.multiline?`(?=$|[${n}])`:"(?![^])",s.DEFAULT))),g){let u=Object.keys(r),i=u.filter(l=>r[l]===!0).join(""),c=u.filter(l=>r[l]===!1).join("");if(c&&(i+=`-${c}`),i)return{value:`(?${i}:${o})`,usedModifier:!0}}return{value:o}}var W=P;var oe={make:W,partial:_,regex:P};return J(se);})();
-//! regex 1.1.0; Steven Levithan; MIT License
+  `.replace(/\s+/g,""),"gsu"),r=!1,n;for(let{0:o,groups:a}of e.matchAll(t)){if(a.pPropOfStr||a.qPropOfStr||o==="["&&r)return!0;if(["-","--","&&"].includes(o))r=!1;else if(!["[","]"].includes(o)){if(r||n==="]")return!0;r=!0}n=o}return!1}function D(e,t,r){let n={raw:[]},o=[],a={};return e.raw.forEach((s,l)=>{let u=r(s,{...a,lastPos:0});if(n.raw.push(u.transformed),a=u.runningContext,l<e.raw.length-1){let E=t[l];if(E instanceof T){let p=r(E,{...a,lastPos:0});o.push(O(p.transformed)),a=p.runningContext}else o.push(E)}}),{template:n,values:o}}function j(e,t){e=String(e);let r="",n="";for(let[o]of e.matchAll(g)){r+=o,t=L(r,t);let{regexContext:a}=t;if(a===f.DEFAULT)if(o==="(")n+="(?:";else{if(/^\\[1-9]/.test(o))throw new Error(`Invalid decimal escape "${o}" with implicit flag n; replace with named backreference`);n+=o}else n+=o}return{transformed:n,runningContext:t}}var Q=/^\s$/,oe=/^\\[\s#]$/,H=/^[ \t]$/,se=/^\\[ \t]$/;function q(e,t){e=String(e);let r=!1,n=!1,o=!1,a="",s="",l="",u="",E=!1,p=(i,{prefix:N=!0,postfix:d=!1}={})=>(i=(E&&N?"(?:)":"")+i+(d?"(?:)":""),E=!1,i);for(let[i]of e.matchAll(g)){if(o){i===`
+`&&(o=!1,E=!0);continue}if(r){if(Q.test(i))continue;r=!1,E=!0}else if(n){if(H.test(i))continue;n=!1}a+=i,t=L(a,t);let{regexContext:N,charClassContext:d}=t;if(i==="-"&&N===f.CHAR_CLASS&&u===c.RANGE)throw new Error("Invalid unescaped hyphen as the end value for a range");if(N===f.DEFAULT&&/^[?*+]\??$/.test(i)||N===f.INTERVAL_QUANTIFIER&&i==="{")s+=p(i,{prefix:!1,postfix:l==="("});else if(N===f.DEFAULT)Q.test(i)?r=!0:i.startsWith("#")?o=!0:oe.test(i)?s+=p(i[1],{prefix:!1}):s+=p(i);else if(N===f.CHAR_CLASS&&i!=="["&&i!=="[^")if(H.test(i)&&(d===c.DEFAULT||d===c.RANGE||d===c.Q_TOKEN))n=!0;else{if(d===c.INVALID_INCOMPLETE_TOKEN)throw new Error(`Invalid incomplete token in character class: "${i}"`);se.test(i)&&(d===c.DEFAULT||d===c.Q_TOKEN)?s+=p(i[1],{prefix:!1}):d===c.DEFAULT?s+=p(I(w(i))):s+=p(i)}else s+=p(i);r||n||o||(l=i,u=d)}return{transformed:s,runningContext:t}}function W(e){let t=String.raw`\(\?:\)`;return e=m(e,`${t}(?:${t})+`,"(?:)",A.DEFAULT),e=m(e,String.raw`^${t}(?![?*+{])|${t}$|${t}(?=[()|$\\])|(?<=[()|>^]|\(\?(?:[:=!]|<[=!]))${t}`,"",A.DEFAULT),e}var b=function(e,...t){let r=this instanceof Function?this:RegExp;if(Array.isArray(e?.raw))return P(r,{flags:""},e,...t);if((typeof e=="string"||e===void 0)&&!t.length)return P.bind(null,r,{flags:e});if({}.toString.call(e)==="[object Object]"&&!t.length)return P.bind(null,r,e);throw new Error(`Unexpected arguments: ${JSON.stringify([e,...t])}`)};function P(e,t,r,...n){let{flags:o="",postprocessors:a=[],__flagN:s=!0,__flagX:l=!0,__rake:u=!0}=t;if(/[vu]/.test(o))throw new Error("Flags v/u cannot be explicitly added since v is always enabled");l&&({template:r,values:n}=D(r,n,q)),s&&({template:r,values:n}=D(r,n,j));let E=0,p="",i={};r.raw.forEach((d,_)=>{let h=r.raw[_]||r.raw[_+1];E+=F(d),p+=w(d,A.CHAR_CLASS),i=L(p,i);let{regexContext:v,charClassContext:z}=i;if(_<r.raw.length-1){let x=ie(n[_],o,v,z,h,E);E+=x.capturesAdded||0,p+=x.value}});let N=[k,...a];u&&N.push(W);for(let d of N)p=d(p);return new e(p,`v${o}`)}function ie(e,t,r,n,o,a){if(e instanceof RegExp&&r!==f.DEFAULT)throw new Error("Cannot interpolate a RegExp at this position because the syntax context does not match");if(r===f.INVALID_INCOMPLETE_TOKEN||n===c.INVALID_INCOMPLETE_TOKEN)throw new Error("Interpolation preceded by invalid incomplete token");let s=e instanceof T,l;if(!(e instanceof RegExp)){e=String(e),s||(l=C(e,r));let u=M(l||e,r,n);if(u)throw new Error(`Unescaped stray "${u}" in the interpolated value would have side effects outside it`)}if(r===f.ENCLOSED_TOKEN||r===f.INTERVAL_QUANTIFIER||r===f.GROUP_NAME||n===c.ENCLOSED_TOKEN||n===c.Q_TOKEN)return{value:s?e:l};if(r===f.CHAR_CLASS){if(s){if(m(e,"^-|^&&|-$|&&$","")!==e)throw new Error("In character classes, a partial cannot use a range/set operator at its boundary; move the operation into the partial or the operator outside of it");let E=V(I(e));return{value:$(e)?`[${E}]`:w(E)}}return{value:$(l)?`[${l}]`:l}}if(e instanceof RegExp){let u=ae(e,t),E=y(u.value,a);return{value:u.usedModifier?E:`(?:${E})`,capturesAdded:F(e.source)}}return s?{value:`(?:${e})`}:{value:o?`(?:${l})`:l}}function ae(e,t){let r={i:null,m:null,s:null},n="\\n\\r\\u2028\\u2029",o=e.source;if(e.ignoreCase!==t.includes("i"))if(S)r.i=e.ignoreCase;else throw new Error("Pattern modifiers not supported, so the value of flag i on the interpolated RegExp must match the outer regex");if(e.dotAll!==t.includes("s")&&(S?r.s=e.dotAll:o=m(o,"\\.",e.dotAll?"[^]":`[^${n}]`,A.DEFAULT)),e.multiline!==t.includes("m")&&(S?r.m=e.multiline:(o=m(o,"\\^",e.multiline?`(?<=^|[${n}])`:"(?<![^])",A.DEFAULT),o=m(o,"\\$",e.multiline?`(?=$|[${n}])`:"(?![^])",A.DEFAULT))),S){let a=Object.keys(r),s=a.filter(u=>r[u]===!0).join(""),l=a.filter(u=>r[u]===!1).join("");if(l&&(s+=`-${l}`),s)return{value:`(?${s}:${o})`,usedModifier:!0}}return{value:o}}var Z=b;var fe={make:Z,partial:O,regex:b};return te(le);})();
+//! regex 1.1.1; Steven Levithan; MIT License
 //# sourceMappingURL=regex.min.js.map

package.json

 {
   "name": "regex",
-  "version": "1.1.0",
-  "description": "Context-aware regex template strings with batteries included",
+  "version": "1.1.1",
+  "description": "Context-aware regex template tag with best practices built-in and advanced features",
   "exports": "./src/index.js",
@@ -17,9 +17,11 @@ "type": "module",
     "type": "git",
-    "url": "git+https://github.com/slevithan/regex-make.git"
+    "url": "git+https://github.com/slevithan/regex.git"
   },
   "keywords": [
     "regex",
-    "regexp",
-    "template"
+    "regexp"
   ],
+  "dependencies": {
+    "regex-utilities": "^1.0.0"
+  },
   "devDependencies": {
@@ -26,0 +28,0 @@ "esbuild": "^0.21.4",

README.md

@@ -6,3 +6,3 @@ # `regex`
 
-`regex` is a template tag for dynamically creating readable, high performance, native JavaScript regular expressions with advanced features. It's lightweight (5.5KB), has no dependencies, and supports all ES2024+ regex features.
+`regex` is a template tag for dynamically creating readable, high performance, native JavaScript regular expressions with advanced features. It's lightweight (5.6KB) and supports all ES2024+ regex features.
 
@@ -98,3 +98,3 @@ Highlights include using whitespace and comments in regexes, atomic groups via `(?>…)` which can help you avoid [ReDoS](https://en.wikipedia.org/wiki/ReDoS), and context-aware interpolation of `RegExp` instances, escaped strings, and partial patterns.
 
-Additionally, JavaScript regex syntax is hard to write and even harder to read and refactor. But it doesn't have to be that way! With a few key features — raw template strings, insignificant whitespace, comments, *named capture only* mode, and interpolation (coming soon: definition blocks and subexpressions as subroutines) — even long and complex regexes can be beautiful, grammatical, and easy to understand.
+Additionally, JavaScript regex syntax is hard to write and even harder to read and refactor. But it doesn't have to be that way! With a few key features — raw multiline template strings, insignificant whitespace, comments, *named capture only* mode, and interpolation (coming soon: definition blocks and subexpressions as subroutines) — even long and complex regexes can be beautiful, grammatical, and easy to understand.
 
@@ -205,6 +205,14 @@ `regex` adds all of these features and returns native `RegExp` instances. It always uses flag <kbd>v</kbd> (already a best practice for new regexes) so you never forget to turn it on and don't have to worry about the differences in other parsing modes. It supports atomic groups via `(?>…)` to help you improve the performance of your regexes and avoid catastrophic backtracking. And it gives you best-in-class, context-aware interpolation of `RegExp` instances, escaped strings, and partial patterns.
 
-Requiring the syntactically clumsy `(?:…)` where you could just use `(…)` hurts readability and encourages adding unneeded captures (which hurt efficiency and refactoring). Flag <kbd>n</kbd> fixes this, making your regexes more readable.
+Motivation: Requiring the syntactically clumsy `(?:…)` where you could just use `(…)` hurts readability and encourages adding unneeded captures (which hurt efficiency and refactoring). Flag <kbd>n</kbd> fixes this, making your regexes more readable.
 
+Example:
+
+```js
+// Doesn't capture
+regex`\b(ab|cd)\b`
+// Use standard (?<name>…) to capture as `name`
+```
+
 > [!NOTE]
-> Flag <kbd>n</kbd> is based on .NET, C++, PCRE, Perl, and XRegExp, which share the `n` flag letter but call it *explicit capture*, *no auto capture*, or *nosubs*. In `regex`, the implicit flag <kbd>n</kbd> also prevents using numbered backreferences to named groups in the outer regex, which follows the behavior of C++. Referring to named groups by number is a footgun, and the way that named groups are numbered is inconsistent across regex flavors.
+> Flag <kbd>n</kbd> is based on .NET, C++, PCRE, Perl, and XRegExp, which share the <kbd>n</kbd> flag letter but call it *explicit capture*, *no auto capture*, or *nosubs*. In `regex`, the implicit flag <kbd>n</kbd> also prevents using numbered backreferences to named groups in the outer regex, which follows the behavior of C++. Referring to named groups by number is a footgun, and the way that named groups are numbered is inconsistent across regex flavors.
 
@@ -234,3 +242,3 @@ > Aside: Flag <kbd>n</kbd>'s behavior also enables `regex` to emulate atomic groups and recursion.
 
-- Regexes can't be interpolated in the middle of a character class (so `` regex`[${/./}]` `` is an error) because the syntax context doesn't match. See [*Interpolating partial patterns*](#interpolating-partial-patterns) for a way to safely embed regex syntax (rather than `RegExp` instances) in character classes and other edge-case locations with different context.
+- Regexes can't be interpolated inside character classes (so `` regex`[${/./}]` `` is an error) because the syntax context doesn't match. See [*Interpolating partial patterns*](#interpolating-partial-patterns) for a way to safely embed regex syntax (rather than `RegExp` instances) in character classes and other edge-case locations with different context.
 - To change the flags used by an interpolated regex, use the built-in capability of `RegExp` to copy a regex while providing new flags. Ex: `new RegExp(/./, 's')`.
@@ -243,3 +251,3 @@ </details>
 
-> As with all interpolation in `regex`, escaped strings are sandboxed and treated as complete units. For example, a following quantifier repeats the whole unit rather than just the last character. And if interpolating into a character class, the escaped string is treated as a flag-<kbd>v</kbd>-mode nested union if it contains more than one character node.
+> As with all interpolation in `regex`, escaped strings are sandboxed and treated as complete units. For example, a following quantifier repeats the entire escaped string rather than just its last character. And if interpolating into a character class, the escaped string is treated as a flag-<kbd>v</kbd>-mode nested union if it contains more than one character node.
 
@@ -260,3 +268,3 @@ As a result, `regex` is a safe and context-aware alternative to JavaScript proposal [`RegExp.escape`](https://github.com/tc39/proposal-regex-escaping).
 // Instead of
-new RegExp(`[a-${RegExp.escape(str)}]`, 'u') // Flag u/v required to avoid bug
+new RegExp(`[a-${RegExp.escape(str)}]`, 'u') // Flag u/v required to avoid bugs
 // You can say
@@ -288,3 +296,3 @@ regex`[a-${str}]`
 - Composing a dynamic number of strings.
-- Adding a pattern in the middle of a character class (not allowed for `RegExp` instances since their top-level syntax context doesn't match).
+- Adding a pattern inside a character class (not allowed for `RegExp` instances since their top-level syntax context doesn't match).
 - Dynamically adding backreferences without their corresponding captures (which wouldn't be valid as a standalone `RegExp`).
@@ -317,3 +325,3 @@ - When you don't want the pattern to specify its own, local flags.
 
-Moving on, the following lines all throw because otherwise the partial patterns would break out of their interpolation sandboxes and change the meaning of the surrounding patterns:
+Moving on, the following lines all throw because otherwise the partial patterns would break out of their interpolation sandboxes and change the meaning of their surrounding patterns:
 
@@ -358,3 +366,3 @@ ```js
 
-1. This is an uncompleted `\u` token (which is an error) followed by the tokens `0`, `0`, `0`, `A`. That's because the interpolation does not happen within an enclosed `\u{…}` context.
+1. This is an uncompleted `\u` token (which is an error) followed by the tokens `0`, `0`, `0`, `A`. That's because the interpolation doesn't happen within an enclosed `\u{…}` context.
 2. The unescaped `}` within the partial is not allowed to break out of its interpolation sandbox.
@@ -369,3 +377,3 @@ 3. The group opening `(` can't be quantified with `?`.
 // This works fine
-regex`[\0-${partial('\\cZ')}]`
+regex`[\0-${partial`\cZ`}]`
 
@@ -430,6 +438,6 @@ // But this is an error since you can't create a range from 'a' to the set 'de'
     <td>•&nbsp;Sandboxed <br> •&nbsp;Atomized <br><br><br></td>
-    <td>•&nbsp;Sandboxed <br> •&nbsp;Atomized <br> •&nbsp;Backrefs adjusted <br> •&nbsp;Own flags apply locally</td>
+    <td>•&nbsp;Sandboxed <br> •&nbsp;Atomized <br> •&nbsp;Backrefs adjusted <br> •&nbsp;Flags localized</td>
   </tr>
   <tr>
-    <td>Character class: <code>[…]</code>, <code>[^…]</code>, <code>[…[…]]</code>, etc.</td>
+    <td>Character class: <code>[…]</code>, <code>[^…]</code>, <code>[[…]]</code>, etc.</td>
     <td><code>regex`[${'a-z'}]`</code><br><br></td>
@@ -466,3 +474,3 @@ <td>•&nbsp;Sandboxed <br> •&nbsp;Atomized <br> •&nbsp;Escaped</td>
 
-- `regex` relies on flag <kbd>v</kbd> (`unicodeSets`), which has had universal browser support since ~mid-2023 and is available in Node.js 20+.
+- `regex` relies on flag <kbd>v</kbd> (`unicodeSets`), which has had universal browser support since ~mid-2023 (see [compat table](https://caniuse.com/mdn-javascript_builtins_regexp_unicodesets)) and is available in Node.js 20+. But it's possible to extend support to older browsers (see [#2](https://github.com/slevithan/regex/issues/2)).
 - Using an interpolated `RegExp` instance with a different value for flag <kbd>i</kbd> than its outer regex relies on [regex modifiers](https://github.com/tc39/proposal-regexp-modifiers), a bleeding-edge feature available in Chrome and Edge 125+. A descriptive error is thrown in environments without support, which you can avoid by aligning the use of flag <kbd>i</kbd> on inner and outer regexes. Local-only application of other flags does not rely on this feature.
@@ -475,5 +483,5 @@ - If you want `regex` to use a `RegExp` subclass or other constructor, you can do so by modifying `this`: `` regex.bind(RegExpSubclass)`…` ``.
 
-Version 1.0.0 was named Regex.make and used the tag name `make` instead of `regex`; hence the repo name.
+Version 1.0.0 was named Regex.make and used tag name `make` instead of `regex`. `make` is still available as an alias.
 
-Crafted with ❤︎ (for developers and regular expressions) by Steven Levithan.<br>
+Crafted by Steven Levithan with ❤︎ for regular expressions and their enthusiasts.<br>
 MIT License.

spec/flag-x-spec.js

@@ -5,3 +5,3 @@ describe('flag x', () => {
       const ws = '\t\n\v\f\r \xA0\u1680\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200A\u2028\u2029\u202F\u205F\u3000\uFEFF';
-      expect('ab').toMatch(regex({raw: [`^a${ws}b$`]}, []));
+      expect('ab').toMatch(regex({raw: [`^a${ws}b$`]}));
     });
@@ -16,3 +16,3 @@
       newlines.forEach(n => {
-        expect('ab').not.toMatch(regex({raw: [`^a#comment${n}b\n$`]}, []));
+        expect('ab').not.toMatch(regex({raw: [`^a#comment${n}b\n$`]}));
       });
@@ -26,3 +26,3 @@ });
     it('should allow mixing whitespace and line comments', function() {
-      expect('ab').toMatch(regex({raw: ['\f^ a \t\n ##comment\n #\nb $ # ignored']}, []));
+      expect('ab').toMatch(regex({raw: ['\f^ a \t\n ##comment\n #\nb $ # ignored']}));
     });
@@ -32,12 +32,36 @@
       expect('aaa').toMatch(regex`^a +$`);
-      expect('aaa').toMatch(regex({raw: ['^a#comment\n+$']}, []));
-      expect('aaa').toMatch(regex({raw: ['^a  #comment\n +$']}, []));
+      expect('aaa').toMatch(regex`^(a) +$`);
+      expect('aaa').toMatch(regex({raw: ['^a#comment\n+$']}));
+      expect('aaa').toMatch(regex({raw: ['^a  #comment\n +$']}));
+      expect(() => regex`a | ?`).toThrow();
     });
 
-    it('should not let the token following whitespace or line comments modify the preceding token', () => {
-        expect('\u{0}0').toMatch(regex`\0 0`);
-        expect('\u{0}1').toMatch(regex`\0 1`);
-        expect('\u{0}1').toMatch(regex({raw: ['\0#\n1']}, []));
+    it('should not allow quantifiers to follow other quantifiers', function() {
+      expect(() => regex`a?? ?`).toThrow();
+      expect(() => regex`a*? ?`).toThrow();
+      expect(() => regex`a+? ?`).toThrow();
+      expect(() => regex`a{2}? ?`).toThrow();
+      expect(() => regex`a* *`).toThrow();
+      expect(() => regex`a+ +`).toThrow();
+      expect(() => regex`a{2} {2}`).toThrow();
     });
 
+    // Follows Perl, PCRE, .NET, Java
+    // Not allowed in Python
+    it('should allow whitespace between a quantifier and the ? that makes it lazy', function() {
+      expect(regex`^aa? ?`.exec('aaa')[0]).toBe('a');
+      expect(regex`^aa* ?`.exec('aaa')[0]).toBe('a');
+      expect(regex`^aa+ ?`.exec('aaa')[0]).toBe('aa');
+      expect(regex`^aa{1,2} ?`.exec('aaa')[0]).toBe('aa');
+    });
+
+    it('should not let the token following whitespace modify the preceding token', () => {
+      expect('\u{0}0').toMatch(regex`^\0 0$`);
+      expect('\u{0}1').toMatch(regex`^\0 1$`);
+    });
+
+    it('should not let the token following a line comment modify the preceding token', () => {
+      expect('\u{0}1').toMatch(regex({raw: ['^\0#\n1$']}));
+    });
+
     it('should preserve the error status of incomplete tokens separated from their completing chars by whitespace', () => {
@@ -52,11 +76,20 @@ const values = [
         '\\x0 0',
+        '(? :)',
       ];
       values.forEach(v => {
-        expect(() => regex({raw: [v]}, [])).withContext(v).toThrow();
+        expect(() => regex({raw: [v]})).withContext(v).toThrow();
       });
     });
 
+    it('should not allow quantifying ( with ? when separated by whitespace and followed by group type char', () => {
+      expect(() => regex`( ?:)`).toThrow();
+      expect(() => regex`( ?=)`).toThrow();
+      expect(() => regex`( ? : )`).toThrow();
+      expect(() => regex`( ? = )`).toThrow();
+    });
+
     it('should allow escaping whitespace to make it significant', () => {
-      expect(' ').toMatch(regex`\ `);
-      expect('  ').toMatch(regex` \ \  `);
+      expect(' ').toMatch(regex`^\ $`);
+      expect('  ').toMatch(regex`^ \ \  $`);
+      expect(' t').toMatch(regex`^\ t$`);
     });
@@ -68,2 +101,11 @@
     });
+
+    it('should treat whitespace in enclosed tokens as significant', () => {
+      expect(() => regex`a{ 6 }`).toThrow();
+      expect(() => regex`\p{ L }`).toThrow();
+      expect(() => regex`\P{ L }`).toThrow();
+      expect(() => regex`\u{ 0 }`).toThrow();
+      expect(() => regex`(?< n >)`).toThrow();
+      expect(() => regex`(?<n>)\k< n >`).toThrow();
+    });
   });
@@ -74,9 +116,9 @@
       expect(' ').not.toMatch(regex`[ a]`);
-      expect('\t').not.toMatch(regex({raw: ['[\ta]']}, []));
+      expect('\t').not.toMatch(regex({raw: ['[\ta]']}));
     });
 
     it('should not treat whitespace characters apart from space and tab as insignificant', () => {
-      expect('\n').toMatch(regex({raw: ['[\na]']}, []));
-      expect('\xA0').toMatch(regex({raw: ['[\xA0a]']}, []));
-      expect('\u2028').toMatch(regex({raw: ['[\u2028a]']}, []));
+      expect('\n').toMatch(regex({raw: ['[\na]']}));
+      expect('\xA0').toMatch(regex({raw: ['[\xA0a]']}));
+      expect('\u2028').toMatch(regex({raw: ['[\u2028a]']}));
     });
@@ -96,4 +138,4 @@
     it('should not let the token following whitespace modify the preceding token', () => {
-        expect('0').toMatch(regex`[\0 0]`);
-        expect('1').toMatch(regex`[\0 1]`);
+        expect('0').toMatch(regex`^[\0 0]$`);
+        expect('1').toMatch(regex`^[\0 1]$`);
     });
@@ -112,3 +154,3 @@
       values.forEach(v => {
-        expect(() => regex({raw: [v]}, [])).withContext(v).toThrow();
+        expect(() => regex({raw: [v]})).withContext(v).toThrow();
       });
@@ -133,3 +175,3 @@ });
       values.forEach(v => {
-        expect(() => regex({raw: [v]}, [])).withContext(v).toThrow();
+        expect(() => regex({raw: [v]})).withContext(v).toThrow();
       });
@@ -170,6 +212,6 @@ });
       doublePunctuatorChars.forEach(c => {
-        expect(c).withContext(`[a${c} ${c}b]`).toMatch(regex({raw: [`[a${c} ${c}b]`]}, []));
-        expect(c).withContext(`[a${c} ${c} b]`).toMatch(regex({raw: [`[a${c} ${c} b]`]}, []));
-        expect(c).withContext(`[a ${c} ${c}b]`).toMatch(regex({raw: [`[a ${c} ${c}b]`]}, []));
-        expect(c).withContext(`[a ${c} ${c} b]`).toMatch(regex({raw: [`[a ${c} ${c} b]`]}, []));
+        expect(c).withContext(`[a${c} ${c}b]`).toMatch(regex({raw: [`[a${c} ${c}b]`]}));
+        expect(c).withContext(`[a${c} ${c} b]`).toMatch(regex({raw: [`[a${c} ${c} b]`]}));
+        expect(c).withContext(`[a ${c} ${c}b]`).toMatch(regex({raw: [`[a ${c} ${c}b]`]}));
+        expect(c).withContext(`[a ${c} ${c} b]`).toMatch(regex({raw: [`[a ${c} ${c} b]`]}));
       });
@@ -179,5 +221,22 @@ });
     it('should allow escaping whitespace to make it significant', () => {
-      expect(' ').toMatch(regex`[ \ ]`);
-      expect(' ').toMatch(regex`[\q{ \ }]`);
+      expect(' ').toMatch(regex`^[ \ ]$`);
+      expect(' ').toMatch(regex`^[\q{ \ }]$`);
+      expect('t ').toMatch(regex`^[\ t]{2}$`);
     });
+
+    it('should treat whitespace in enclosed tokens as significant', () => {
+      expect(() => regex`[\p{ L }]`).toThrow();
+      expect(() => regex`[\P{ L }]`).toThrow();
+      expect(() => regex`[\u{ 0 }]`).toThrow();
+    });
+
+    it('should treat whitespace in [\\q{}] as insignificant', () => {
+      expect('ab').toMatch(regex`^[\q{ a b | c }]$`);
+    });
+
+    it('should handle empty character classes with insignificant whitespace', () => {
+      expect(/[]/v.test('a')).toBe(regex`[ ]`.test('a'));
+      expect(/[^]/v.test('a')).toBe(regex`[^ ]`.test('a'));
+      expect(/[\q{}]/v.test('a')).toBe(regex`[ \q{ } ]`.test('a'));
+    });
   });
@@ -184,0 +243,0 @@

spec/interpolate-partial-spec.js

@@ -50,2 +50,7 @@ describe('interpolation: partial patterns', () => {
 
+    it('should not let > end an enclosed token', () => {
+      expect(() => regex`(?<n>)\k<${partial`n>`}`).toThrow();
+      expect(() => regex`(?<${partial`n>`}>)`).toThrow();
+    });
+
     it('should not let a preceding unescaped \\ change the first character inside the interpolation', () => {
@@ -52,0 +57,0 @@ // Raw string syntax prevents `\${'w'}` since the raw \ escapes the $

spec/interpolate-string-spec.js

@@ -18,2 +18,7 @@ describe('interpolation: escaped strings', () => {
 
+    it('should not let > end an enclosed token', () => {
+      expect(() => regex`(?<n>)\k<${'n>'}`).toThrow();
+      expect(() => regex`(?<${'n>'}>)`).toThrow();
+    });
+
     it('should not let a preceding unescaped \\ change the first character inside the interpolation', () => {
@@ -20,0 +25,0 @@ // Raw string syntax prevents `\${'w'}` since the raw \ escapes the $

spec/partial-spec.js

 describe('partial', () => {
+  it('should coerce non-string values', () => {
+    expect(String(partial('1'))).toBe('1');
+    expect(String(partial(1))).toBe('1');
+    expect(String(partial(NaN))).toBe('NaN');
+  });
+
   describe('strings', () => {
@@ -7,8 +13,2 @@ it('should accept empty arguments', () => {
     });
-
-    it('should coerce to string', () => {
-      expect(String(partial('1'))).toBe('1');
-      expect(String(partial(1))).toBe('1');
-      expect(String(partial(NaN))).toBe('NaN');
-    });
   });
@@ -15,0 +15,0 @@

spec/regex-syntax-spec.js

@@ -48,8 +48,8 @@ // Ensure transformations have not changed native syntax and errors
     reservedDoublePunctuators.forEach(dp => {
-      expect(() => regex({raw: [`[a${dp}b]`]}, [])).withContext(dp).toThrow();
+      expect(() => regex({raw: [`[a${dp}b]`]})).withContext(dp).toThrow();
     });
     allDoublePunctuatorsExcludingCaret.forEach(dp => {
-      expect(() => regex({raw: [`[${dp}]`]}, [])).withContext(dp).toThrow();
-      expect(() => regex({raw: [`[${dp}b]`]}, [])).withContext(dp).toThrow();
-      expect(() => regex({raw: [`[a${dp}]`]}, [])).withContext(dp).toThrow();
+      expect(() => regex({raw: [`[${dp}]`]})).withContext(dp).toThrow();
+      expect(() => regex({raw: [`[${dp}b]`]})).withContext(dp).toThrow();
+      expect(() => regex({raw: [`[a${dp}]`]})).withContext(dp).toThrow();
     });
@@ -56,0 +56,0 @@ expect(() => regex`[^^^]`).withContext('^^').toThrow();

src/atomic-groups.js

@@ -1,8 +0,8 @@
-import {RegexContext, hasUnescapedInDefaultRegexContext, replaceUnescaped} from './utils.js';
+import {Context, hasUnescaped, replaceUnescaped} from 'regex-utilities';
 
 export function transformAtomicGroups(pattern) {
-  if (!hasUnescapedInDefaultRegexContext(pattern, String.raw`\(\?>`)) {
+  if (!hasUnescaped(pattern, String.raw`\(\?>`, Context.DEFAULT)) {
     return pattern;
   }
-  const token = new RegExp(String.raw`(?<noncapturingStart>\(\?(?:[:=!>]|<[=!]|[ims\-]+:))|(?<capturingStart>\((?:\?<[^>]+>)?)|(?<backrefNum>\\[1-9]\d*)|\\?.`, 'gsu');
+  const token = new RegExp(String.raw`(?<noncapturingStart>\(\?(?:[:=!>A-Za-z\-]|<[=!]))|(?<capturingStart>\((?:\?<[^>]+>)?)|(?<backrefNum>\\[1-9]\d*)|\\?.`, 'gsu');
   const aGDelimLen = '(?>'.length;
@@ -64,5 +64,5 @@ let hasProcessedAG;
     ({groups: {backrefNum}}) => `\\${backrefNum}`,
-    RegexContext.DEFAULT
+    Context.DEFAULT
   );
   return pattern;
 }

src/flag-n.js

 import {RegexContext, contextToken, getEndContextForIncompletePattern} from './utils.js';
 
+// Applied to the outer regex and interpolated partials, but not interpolated regexes or strings
 export function flagNProcessor(value, runningContext) {
@@ -4,0 +5,0 @@ value = String(value);

src/flag-x.js

@@ -1,2 +0,3 @@
-import {CharClassContext, RegexContext, contextToken, getEndContextForIncompletePattern, replaceUnescaped, sandboxLoneDoublePunctuatorChar, sandboxUnsafeNulls} from './utils.js';
+import {Context, replaceUnescaped} from 'regex-utilities';
+import {CharClassContext, RegexContext, contextToken, getEndContextForIncompletePattern, sandboxLoneDoublePunctuatorChar, sandboxUnsafeNulls} from './utils.js';
 
@@ -8,2 +9,3 @@ const ws = /^\s$/;
 
+// Applied to the outer regex and interpolated partials, but not interpolated regexes or strings
 export function flagXProcessor(value, runningContext) {
@@ -16,2 +18,3 @@ value = String(value);
   let transformed = '';
+  let lastSignificantToken = '';
   let lastSignificantCharClassContext = '';
@@ -60,3 +63,3 @@ let divNeeded = false;
       // turned off
-      transformed += update(m, {prefix: false, postfix: m === '?'});
+      transformed += update(m, {prefix: false, postfix: lastSignificantToken === '('});
     } else if (regexContext === RegexContext.DEFAULT) {
@@ -99,2 +102,3 @@ if (ws.test(m)) {
     if (!(ignoringWs || ignoringCharClassWs || ignoringComment)) {
+      lastSignificantToken = m;
       lastSignificantCharClassContext = charClassContext;
@@ -113,3 +117,3 @@ }
   // No need for repeated separators
-  pattern = replaceUnescaped(pattern, `${sep}(?:${sep})+`, '(?:)', RegexContext.DEFAULT);
+  pattern = replaceUnescaped(pattern, `${sep}(?:${sep})+`, '(?:)', Context.DEFAULT);
   // No need for separators at:
@@ -124,5 +128,5 @@ // - The beginning, if not followed by a quantifier.
     '',
-    RegexContext.DEFAULT
+    Context.DEFAULT
   );
   return pattern;
 }

src/index.js

@@ -1,3 +0,4 @@
-//! regex 1.1.0; Steven Levithan; MIT License
+//! regex 1.1.1; Steven Levithan; MIT License
 
+import {Context, replaceUnescaped} from 'regex-utilities';
 import {transformAtomicGroups} from './atomic-groups.js';
@@ -7,5 +8,14 @@ import {flagNProcessor} from './flag-n.js';
 import {PartialPattern, partial} from './partial.js';
-import {CharClassContext, RegexContext, adjustNumberedBackrefs, containsCharClassUnion, countCaptures, escapeV, getBreakoutChar, getEndContextForIncompletePattern, patternModsOn, replaceUnescaped, sandboxLoneCharClassCaret, sandboxLoneDoublePunctuatorChar, sandboxUnsafeNulls, transformTemplateAndValues} from './utils.js';
+import {CharClassContext, RegexContext, adjustNumberedBackrefs, containsCharClassUnion, countCaptures, escapeV, getBreakoutChar, getEndContextForIncompletePattern, patternModsOn, sandboxLoneCharClassCaret, sandboxLoneDoublePunctuatorChar, sandboxUnsafeNulls, transformTemplateAndValues} from './utils.js';
 
 /**
+@typedef {Object} RegexTagOptions
+@prop {string} [flags]
+@prop {Array<Function>} [postprocessors]
+@prop {boolean} [__flagN]
+@prop {boolean} [__flagX]
+@prop {boolean} [__rake]
+*/
+
+/**
 Template tag for constructing a UnicodeSets-mode RegExp with advanced features and context-aware
@@ -19,7 +29,10 @@ interpolation of regexes, escaped strings, and partial patterns.
 4. `` regex.bind(RegExpSubclass)`…` `` - With a `this` that specifies a different constructor.
-@param {string | TemplateStringsArray} first Flags or a template.
-@param {...any} [values] Values to fill the template holes.
-@returns {RegExp | (TemplateStringsArray, ...any) => RegExp}
+
+@type {{
+  (flags?: string) => (TemplateStringsArray, ...values) => RegExp;
+  (options: RegexTagOptions) => (TemplateStringsArray, ...values) => RegExp;
+  (template: TemplateStringsArray, ...values) => RegExp;
+}}
 */
-function regex(first, ...values) {
+const regex = function(first, ...values) {
   // Allow binding to other constructors
@@ -38,8 +51,8 @@ const constructor = this instanceof Function ? this : RegExp;
   throw new Error(`Unexpected arguments: ${JSON.stringify([first, ...values])}`);
-}
+};
 
 /**
 Makes a UnicodeSets-mode RegExp from a template and values to fill the template holes.
-@param {RegExpConstructor} constructor
-@param {Object} options
+@param {RegExpConstructor | (pattern: string, flags: string) => RegExp} constructor
+@param {RegexTagOptions} options
 @param {TemplateStringsArray} template
@@ -52,5 +65,6 @@ @param {...any} values
     flags = '',
+    postprocessors = [],
     __flagN = true,
     __flagX = true,
-    __rake = options.__flagX ?? true,
+    __rake = true,
   } = options;
@@ -80,3 +94,3 @@ if (/[vu]/.test(flags)) {
     // cases a following interpolated value would always be atomized
-    pattern += sandboxUnsafeNulls(raw, RegexContext.CHAR_CLASS);
+    pattern += sandboxUnsafeNulls(raw, Context.CHAR_CLASS);
     runningContext = getEndContextForIncompletePattern(pattern, runningContext);
@@ -91,4 +105,10 @@ const {regexContext, charClassContext} = runningContext;
 
-  pattern = transformAtomicGroups(pattern);
-  return new constructor(__rake ? rakeSeparators(pattern) : pattern, `v${flags}`);
+  const postp = [transformAtomicGroups, ...postprocessors];
+  if (__rake) {
+    postp.push(rakeSeparators);
+  }
+  for (const p of postp) {
+    pattern = p(pattern);
+  }
+  return new constructor(pattern, `v${flags}`);
 }
@@ -186,3 +206,3 @@
     } else {
-      value = replaceUnescaped(value, '\\.', (re.dotAll ? '[^]' : `[^${newlines}]`), RegexContext.DEFAULT);
+      value = replaceUnescaped(value, '\\.', (re.dotAll ? '[^]' : `[^${newlines}]`), Context.DEFAULT);
     }
@@ -194,4 +214,4 @@ }
     } else {
-      value = replaceUnescaped(value, '\\^', (re.multiline ? `(?<=^|[${newlines}])` : '(?<![^])'), RegexContext.DEFAULT);
-      value = replaceUnescaped(value, '\\$', (re.multiline ? `(?=$|[${newlines}])` : '(?![^])'), RegexContext.DEFAULT);
+      value = replaceUnescaped(value, '\\^', (re.multiline ? `(?<=^|[${newlines}])` : '(?<![^])'), Context.DEFAULT);
+      value = replaceUnescaped(value, '\\$', (re.multiline ? `(?=$|[${newlines}])` : '(?![^])'), Context.DEFAULT);
     }
@@ -198,0 +218,0 @@ }

src/utils.js

@@ -0,6 +1,6 @@
+import {Context, replaceUnescaped} from 'regex-utilities';
 import {PartialPattern, partial} from './partial.js';
 
 export const RegexContext = {
-  DEFAULT: 'DEFAULT',
-  CHAR_CLASS: 'CHAR_CLASS',
+  ...Context,
   GROUP_NAME: 'GROUP_NAME',
@@ -70,4 +70,4 @@ ENCLOSED_TOKEN: 'ENCLOSED_TOKEN',
 // be placed on a range boundary). So escape \0 in character classes as \u{0}
-export function sandboxUnsafeNulls(str, inRegexContext) {
-  return replaceUnescaped(str, String.raw`\\0(?!\d)`, '\\u{0}', inRegexContext);
+export function sandboxUnsafeNulls(str, inContext) {
+  return replaceUnescaped(str, String.raw`\\0(?!\d)`, '\\u{0}', inContext);
 }
@@ -152,3 +152,3 @@
 | \[\^
-| \(\?[:=!<>ims\-]
+| \(\? [:=!<>A-Za-z\-]
 | (?<dp> [${doublePunctuatorChars}] ) \k<dp>
@@ -230,77 +230,2 @@ | \\[1-9]\d*
 
-/**
-Replaces patterns only when they're unescaped and in the given context.
-Doesn't skip over complete multicharacter tokens (only `\` and folowing char) so must be used with
-knowledge of what's safe to do given regex syntax.
-Assumes flag v and doesn't worry about syntax errors that are caught by it.
-@param {string} pattern
-@param {string} needle Search as a regex pattern, with flags `su`
-@param {string | (match: RegExpExecArray) => string} replacement
-@param {'DEFAULT' | 'CHAR_CLASS'} [inRegexContext]
-@returns {string}
-@example
-replaceUnescaped(String.raw`.\.\\.\\\.[[\.].].`, '\\.', '~');
-// -> String.raw`~\.\\~\\\.[[\.]~]~`
-replaceUnescaped(String.raw`.\.\\.\\\.[[\.].].`, '\\.', '~', RegexContext.DEFAULT);
-// -> String.raw`~\.\\~\\\.[[\.].]~`
-*/
-export function replaceUnescaped(pattern, needle, replacement, inRegexContext) {
-  const re = new RegExp(String.raw`(?<found>${needle})|\\?.`, 'gsu');
-  let numCharClassesOpen = 0;
-  let result = '';
-  for (const match of pattern.matchAll(re)) {
-    const {0: m, groups: {found}} = match;
-    if (found && (!inRegexContext || (inRegexContext === RegexContext.DEFAULT) === !numCharClassesOpen)) {
-      if (replacement instanceof Function) {
-        result += replacement(match);
-      } else {
-        result += replacement;
-      }
-      continue;
-    }
-    if (m === '[') {
-      numCharClassesOpen++;
-    } else if (m === ']' && numCharClassesOpen) {
-      numCharClassesOpen--;
-    }
-    result += m;
-  }
-  return result;
-}
-
-/**
-Check if an unescaped version of a pattern appears outside of a character class.
-Doesn't skip over complete multicharacter tokens (only `\` and folowing char) so must be used with
-knowledge of what's safe to do given regex syntax.
-Assumes flag v and doesn't worry about syntax errors that are caught by it.
-@param {string} pattern
-@param {string} needle Search as a regex pattern, with flags `su`
-@param {(match: RegExpExecArray) => void} [callback]
-@returns {boolean}
-*/
-export function hasUnescapedInDefaultRegexContext(pattern, needle, callback) {
-  // Quick partial test; avoid the loop if not needed
-  if (!(new RegExp(needle, 'su')).test(pattern)) {
-    return false;
-  }
-  const re = new RegExp(String.raw`(?<found>${needle})|\\?.`, 'gsu');
-  let numCharClassesOpen = 0;
-  for (const match of pattern.matchAll(re)) {
-    const {0: m, groups: {found}} = match;
-    if (m === '[') {
-      numCharClassesOpen++;
-    } else if (!numCharClassesOpen) {
-      if (found) {
-        if (callback) {
-          callback(match);
-        }
-        return true;
-      }
-    } else if (m === ']') {
-      numCharClassesOpen--;
-    }
-  }
-  return false;
-}
-
 // Assumes flag v and doesn't worry about syntax errors that are caught by it
@@ -307,0 +232,0 @@ export function countCaptures(pattern) {

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

159613

increased by2.7%

Lines of code

1756

increased by1.74%

Number of lines in readme file

475

increased by1.71%

Worsened metrics

Dependency count

1

increased byInfinity%

Dependency changes

• + Addedregex-utilities@^1.0.0

• + Addedregex-utilities@1.1.1(transitive)