File Explorer

1/**2 * xml-naming3 * Validates XML Name productions as defined in the XML 1.0 and 1.1 specifications.4 * Covers: Name, NCName, QName, NMToken, NMTokens5 *6 * XML 1.0 spec: https://www.w3.org/TR/xml/#NT-Name7 * XML 1.1 spec: https://www.w3.org/TR/xml11/#NT-NameStartChar8 * XML NS spec:  https://www.w3.org/TR/xml-names/#NT-NCName9 */10 11// ---------------------------------------------------------------------------12// Character class strings — XML 1.013//14// NameStartChar ::= ":" | [A-Z] | "_" | [a-z]15//   | [#xC0-#xD6]   | [#xD8-#xF6]   | [#xF8-#x2FF]16//   | [#x370-#x37D] | [#x37F-#x1FFF]    <- split to exclude #x048717//   | [#x200C-#x200D]18//   | [#x2070-#x218F] | [#x2C00-#x2FEF]19//   | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD]20//21// NameChar ::= NameStartChar | "-" | "." | [0-9]22//   | #xB7 | [#x0300-#x036F] | [#x203F-#x2040]23//24// Note: \u0487 (Combining Cyrillic Millions Sign) was added in Unicode 4.0,25// after XML 1.0 was defined against Unicode 2.0. It falls inside the range26// \u037F-\u1FFF but must be excluded. We split that range into27// \u037F-\u0486 and \u0488-\u1FFF to exclude it explicitly.28// ---------------------------------------------------------------------------29 30const nameStartChar10 =31  ':A-Za-z_' +32  '\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF' +33  '\u0370-\u037D' +34  '\u037F-\u0486\u0488-\u1FFF' +  // split to exclude \u048735  '\u200C-\u200D' +36  '\u2070-\u218F' +37  '\u2C00-\u2FEF' +38  '\u3001-\uD7FF' +39  '\uF900-\uFDCF' +40  '\uFDF0-\uFFFD';41 42const nameChar10 =43  nameStartChar10 +44  '\\-\\.\\d' +45  '\u00B7' +46  '\u0300-\u036F' +47  '\u203F-\u2040';48 49// ---------------------------------------------------------------------------50// Character class strings — XML 1.151//52// Differences from XML 1.0:53//54// NameStartChar:55//   1.0 has split ranges: \u00C0-\u00D6, \u00D8-\u00F6, \u00F8-\u02FF56//   1.1 merges them into: \u00C0-\u02FF57//   (\u00D7 x and \u00F7 / are division symbols, excluded in both versions)58//59//   1.0 tops out at \uFFFD (BMP only)60//   1.1 adds \u{10000}-\u{EFFFF} (supplementary planes)61//   These require the /u flag on the RegExp — see buildRegexes below.62//63// NameChar:64//   1.1 adds \u0487 (Combining Cyrillic Millions Sign, added in Unicode 4.0)65// ---------------------------------------------------------------------------66 67const nameStartChar11 =68  ':A-Za-z_' +69  '\u00C0-\u02FF' +                    // merged — 1.0 had three split ranges here70  '\u0370-\u037D' +71  '\u037F-\u0486\u0488-\u1FFF' +       // split to exclude \u0487 (combining mark, never a NameStartChar)72  '\u200C-\u200D' +73  '\u2070-\u218F' +74  '\u2C00-\u2FEF' +75  '\u3001-\uD7FF' +76  '\uF900-\uFDCF' +77  '\uFDF0-\uFFFD' +78  '\u{10000}-\u{EFFFF}';     // supplementary planes — REQUIRES /u flag on RegExp79 80const nameChar11 =81  nameStartChar11 +82  '\\-\\.\\d' +83  '\u00B7' +84  '\u0300-\u036F' +85  '\u0487' +                 // Combining Cyrillic Millions Sign — valid in 1.1, not 1.086  '\u203F-\u2040';87 88// ---------------------------------------------------------------------------89// Regex builders90//91// XML 1.0 regexes: no flags — BMP only, standard JS regex behaviour.92// XML 1.1 regexes: /u flag — required for \u{10000}-\u{EFFFF} to match actual93//   supplementary code points rather than lone surrogates (which are illegal XML).94// ---------------------------------------------------------------------------95 96const buildRegexes = (startChar, char, flags = '') => {97  const ncStart = startChar.replace(':', '');98  const ncChar = char.replace(':', '');99  const ncNamePat = `[${ncStart}][${ncChar}]*`;100 101  return {102    name: new RegExp(`^[${startChar}][${char}]*$`, flags),103    ncName: new RegExp(`^${ncNamePat}$`, flags),104    qName: new RegExp(`^${ncNamePat}(?::${ncNamePat})?$`, flags),105    nmToken: new RegExp(`^[${char}]+$`, flags),106    nmTokens: new RegExp(`^[${char}]+(?:\\s+[${char}]+)*$`, flags),107  };108};109 110const regexes10 = buildRegexes(nameStartChar10, nameChar10);       // no /u — BMP only111const regexes11 = buildRegexes(nameStartChar11, nameChar11, 'u');  // /u — enables \u{10000}-\u{EFFFF}112 113const getRegexes = (xmlVersion = '1.0') =>114  xmlVersion === '1.1' ? regexes11 : regexes10;115 116// ---------------------------------------------------------------------------117// Boolean validators118// ---------------------------------------------------------------------------119 120/**121 * Returns true if the string is a valid XML Name.122 * Colons are allowed anywhere (Name production).123 * Used for: DOCTYPE entity names, notation names, DTD element declarations.124 */125export const name = (str, { xmlVersion = '1.0' } = {}) =>126  getRegexes(xmlVersion).name.test(str);127 128/**129 * Returns true if the string is a valid NCName (Non-Colonized Name).130 * Colons are not permitted.131 * Used for: namespace prefixes, local names, SVG id attributes.132 */133export const ncName = (str, { xmlVersion = '1.0' } = {}) =>134  getRegexes(xmlVersion).ncName.test(str);135 136/**137 * Returns true if the string is a valid QName (Qualified Name).138 * Allows exactly one colon as a prefix separator: prefix:localName.139 * Used for: element and attribute names in namespace-aware XML/SVG.140 */141export const qName = (str, { xmlVersion = '1.0' } = {}) =>142  getRegexes(xmlVersion).qName.test(str);143 144/**145 * Returns true if the string is a valid NMToken.146 * Like Name but no restriction on the first character.147 * Used for: DTD NMTOKEN attribute values.148 */149export const nmToken = (str, { xmlVersion = '1.0' } = {}) =>150  getRegexes(xmlVersion).nmToken.test(str);151 152/**153 * Returns true if the string is a valid NMTokens value.154 * A whitespace-separated list of NMToken values.155 * Used for: DTD NMTOKENS attribute values.156 */157export const nmTokens = (str, { xmlVersion = '1.0' } = {}) =>158  getRegexes(xmlVersion).nmTokens.test(str);159 160// ---------------------------------------------------------------------------161// Diagnostic validator162// ---------------------------------------------------------------------------163 164const PRODUCTIONS = ['name', 'ncName', 'qName', 'nmToken', 'nmTokens'];165 166/**167 * Validates a string against a named production and returns a detailed result.168 *169 * @param {string} str170 * @param {'name'|'ncName'|'qName'|'nmToken'|'nmTokens'} production171 * @param {{ xmlVersion?: '1.0'|'1.1' }} [opts]172 * @returns {{ valid: boolean, production: string, input: string, reason?: string, position?: number }}173 */174export const validate = (str, production, { xmlVersion = '1.0' } = {}) => {175  if (!PRODUCTIONS.includes(production)) {176    throw new TypeError(177      `Unknown production "${production}". Must be one of: ${PRODUCTIONS.join(', ')}`178    );179  }180 181  const validators = { name, ncName, qName, nmToken, nmTokens };182  const isValid = validators[production](str, { xmlVersion });183 184  if (isValid) return { valid: true, production, input: str };185 186  let reason = 'Does not match the production rules';187  let position;188 189  if (str.length === 0) {190    reason = 'Input is empty';191  } else if (production === 'ncName' && str.includes(':')) {192    position = str.indexOf(':');193    reason = 'Colon is not allowed in NCName';194  } else if (production === 'qName' && str.startsWith(':')) {195    reason = 'QName cannot start with a colon';196    position = 0;197  } else if (production === 'qName' && str.endsWith(':')) {198    reason = 'QName cannot end with a colon';199    position = str.length - 1;200  } else if (production === 'qName' && (str.match(/:/g) || []).length > 1) {201    reason = 'QName can have at most one colon';202    position = str.lastIndexOf(':');203  } else if (204    ['name', 'ncName', 'qName'].includes(production) &&205    !/^[:A-Za-z_\u00C0-\uFFFD]/.test(str[0])206  ) {207    reason = `First character "${str[0]}" is not a valid NameStartChar`;208    position = 0;209  } else {210    for (let i = 0; i < str.length; i++) {211      if (!/[\w\-\\.:\u00B7\u00C0-\uFFFD]/.test(str[i])) {212        reason = `Character "${str[i]}" at position ${i} is not a valid NameChar`;213        position = i;214        break;215      }216    }217  }218 219  return { valid: false, production, input: str, reason, position };220};221 222// ---------------------------------------------------------------------------223// Batch validator224// ---------------------------------------------------------------------------225 226/**227 * Validates an array of strings against a named production.228 *229 * @param {string[]} strings230 * @param {'name'|'ncName'|'qName'|'nmToken'|'nmTokens'} production231 * @param {{ xmlVersion?: '1.0'|'1.1' }} [opts]232 * @returns {Array<{ valid: boolean, production: string, input: string, reason?: string, position?: number }>}233 */234export const validateAll = (strings, production, opts = {}) =>235  strings.map(str => validate(str, production, opts));236 237// ---------------------------------------------------------------------------238// Sanitizer239// ---------------------------------------------------------------------------240 241/**242 * Transforms an invalid string into the nearest valid XML name for the given production.243 *244 * @param {string} str245 * @param {'name'|'ncName'|'qName'|'nmToken'|'nmTokens'} production246 * @param {{ replacement?: string }} [opts]247 * @returns {string}248 */249export const sanitize = (str, production = 'name', { replacement = '_' } = {}) => {250  if (!str) return replacement;251 252  let result = str;253 254  // Strip colons for NCName255  if (production === 'ncName') {256    result = result.replace(/:/g, '');257  }258 259  // Replace illegal characters260  result = result.replace(/[^\w\-\.:\u00B7\u00C0-\uFFFD]/g, replacement);261 262  // Fix invalid start character for Name / NCName / QName263  if (production !== 'nmToken' && production !== 'nmTokens') {264    if (/^[\-\.\d]/.test(result)) {265      result = replacement + result;266    }267  }268 269  return result || replacement;270};