marc-ts 0.2.0 → 0.4.2

package/dist/marctxt.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"marctxt.cjs","names":[],"sources":["../src/marctxt.ts"],"sourcesContent":["/**\n * MARCBreaker (marctxt) parser and serializer.\n *\n * Also known as MARCMaker format. Each field is one line:\n *\n *   =LDR  00706cam a2200217 a 4500\n *   =001  5490\n *   =245  14$aThe Hobbit /$cJ.R.R. Tolkien.\n *   =650  \\1$aHobbits (Fictitious characters)$vFiction.\n *\n * Blank indicators are represented as `\\`. Records are separated by blank lines.\n * Subfield delimiter is `$` followed by a single character code.\n *\n * Curly-brace escape sequences (per LC MARCMaker spec):\n *   `{`  → `{lcub}`   (left curly brace; reserved by the format)\n *   `}`  → `{rcub}`   (right curly brace; reserved by the format)\n *   `$`  → `{dollar}` (subfield delimiter)\n *   `\\`  → `{bsol}`   (backslash; reserved as blank-indicator stand-in)\n *\n */\n\nimport type { MarcRecord, ControlField, DataField, Subfield } from './types';\nimport { isControlField } from './types';\n\n// ─── Indicator encoding ───────────────────────────────────────────────────────\n\nfunction encodeIndicator(ind: string): string {\n  return ind === ' ' ? '\\\\' : ind;\n}\n\nfunction decodeIndicator(ch: string): string {\n  return ch === '\\\\' ? ' ' : ch;\n}\n\n// ─── Value escape (see file header) ───────────────────────────────────────────\n\nfunction escapeValue(s: string): string {\n  // Single-pass replacement avoids the problem of earlier escapes being\n  // re-escaped by later passes (e.g. '{' → '{lcub}' then '}' → '{lcub{rcub}').\n  return s.replace(/[{}\\\\\\n$]/g, (ch) => {\n    if (ch === '{') return '{lcub}';\n    if (ch === '}') return '{rcub}';\n    if (ch === '$') return '{dollar}';\n    if (ch === '\\\\') return '{bsol}';\n    return ' ';\n  });\n}\n\nfunction unescapeValue(s: string): string {\n  return s.replace(/\\{(lcub|rcub|dollar|bsol)\\}/g, (_, name) => {\n    if (name === 'lcub') return '{';\n    if (name === 'rcub') return '}';\n    if (name === 'dollar') return '$';\n    return '\\\\';\n  });\n}\n\n// ─── Subfield parsing ─────────────────────────────────────────────────────────\n\n/**\n * Parse a subfield string like \"$aValue$bOther\" into Subfield objects.\n * Uses split with a capturing group: \"$aFoo$bBar\" → [\"\", \"a\", \"Foo\", \"b\", \"Bar\"].\n * Any character following `$` is treated as a subfield code.\n */\nfunction parseSubfields(str: string): Subfield[] {\n  const parts = str.split(/\\$(.)/);\n  const subfields: Subfield[] = [];\n  // parts[0] is content before the first $ — should be empty for well-formed data\n  for (let i = 1; i < parts.length; i += 2) {\n    subfields.push({ code: parts[i]!, value: unescapeValue(parts[i + 1] ?? '') });\n  }\n  return subfields;\n}\n\n// ─── Record block parser ──────────────────────────────────────────────────────\n\n/**\n * Parse a block of non-empty marctxt lines into a MarcRecord.\n * Each line has the form `=TAG  content`.\n */\nfunction parseRecordLines(lines: string[]): MarcRecord {\n  let leader = '';\n  const fields: (ControlField | DataField)[] = [];\n\n  for (const line of lines) {\n    if (!line.startsWith('=')) continue;\n    const tag = line.slice(1, 4);\n    // positions 4-5 are the two separator spaces; content starts at 6\n    const content = line.slice(6);\n\n    if (tag === 'LDR' || tag === '000') {\n      leader = content;\n      continue;\n    }\n\n    if (tag < '010') {\n      // Control field: content is the raw field data\n      fields.push({ tag, data: unescapeValue(content) });\n      continue;\n    }\n\n    // Data field: first two chars are indicators, rest are subfields\n    const indicator1 = decodeIndicator(content[0] ?? '\\\\');\n    const indicator2 = decodeIndicator(content[1] ?? '\\\\');\n    const subfields = parseSubfields(content.slice(2));\n    fields.push({ tag, indicator1, indicator2, subfields });\n  }\n\n  return { leader, fields };\n}\n\n// ─── Public parse API ─────────────────────────────────────────────────────────\n\n/**\n * Parse a marctxt string containing one or more records separated by blank lines.\n * Returns all records found.\n */\nexport function parseMarcTxt(text: string): MarcRecord[] {\n  const lines = text.replace(/\\r\\n/g, '\\n').split('\\n');\n  const records: MarcRecord[] = [];\n  let buffer: string[] = [];\n\n  for (const line of lines) {\n    if (line.trim() === '') {\n      if (buffer.length > 0) {\n        records.push(parseRecordLines(buffer));\n        buffer = [];\n      }\n    } else {\n      buffer.push(line);\n    }\n  }\n\n  if (buffer.length > 0) {\n    records.push(parseRecordLines(buffer));\n  }\n\n  return records;\n}\n\n// ─── Serializer ───────────────────────────────────────────────────────────────\n\nfunction serializeMarcTxtRecord(record: MarcRecord): string {\n  const lines: string[] = [];\n\n  lines.push(`=LDR  ${record.leader}`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(`=${field.tag}  ${escapeValue(field.data)}`);\n    } else {\n      const ind1 = encodeIndicator(field.indicator1);\n      const ind2 = encodeIndicator(field.indicator2);\n      const subfields = field.subfields\n        .map((sf) => `$${sf.code}${escapeValue(sf.value)}`)\n        .join('');\n      lines.push(`=${field.tag}  ${ind1}${ind2}${subfields}`);\n    }\n  }\n\n  return lines.join('\\n') + '\\n';\n}\n\n/**\n * Serialize one or more MarcRecords into a marctxt string.\n * Records are separated by blank lines.\n */\nexport function serializeMarcTxt(records: MarcRecord[]): string {\n  // Each record ends with '\\n'; joining with '\\n' produces blank lines between records.\n  return records.map(serializeMarcTxtRecord).join('\\n');\n}\n"],"mappings":"2GA0BA,SAAS,EAAgB,EAAqB,CAC5C,OAAO,IAAQ,IAAM,KAAO,CAC9B,CAEA,SAAS,EAAgB,EAAoB,CAC3C,OAAO,IAAO,KAAO,IAAM,CAC7B,CAIA,SAAS,EAAY,EAAmB,CAGtC,OAAO,EAAE,QAAQ,aAAe,GAC1B,IAAO,IAAY,SACnB,IAAO,IAAY,SACnB,IAAO,IAAY,WACnB,IAAO,KAAa,SACjB,GACR,CACH,CAEA,SAAS,EAAc,EAAmB,CACxC,OAAO,EAAE,QAAQ,+BAAA,CAAiC,EAAG,IAC/C,IAAS,OAAe,IACxB,IAAS,OAAe,IACxB,IAAS,SAAiB,IACvB,IACR,CACH,CASA,SAAS,EAAe,EAAyB,CAC/C,MAAM,EAAQ,EAAI,MAAM,OAAO,EACzB,EAAwB,CAAC,EAE/B,QAAS,EAAI,EAAG,EAAI,EAAM,OAAQ,GAAK,EACrC,EAAU,KAAK,CAAE,KAAM,EAAM,CAAA,EAAK,MAAO,EAAc,EAAM,EAAI,CAAA,GAAM,EAAE,CAAE,CAAC,EAE9E,OAAO,CACT,CAQA,SAAS,EAAiB,EAA6B,CACrD,IAAI,EAAS,GACb,MAAM,EAAuC,CAAC,EAE9C,UAAW,KAAQ,EAAO,CACxB,GAAI,CAAC,EAAK,WAAW,GAAG,EAAG,SAC3B,MAAM,EAAM,EAAK,MAAM,EAAG,CAAC,EAErB,EAAU,EAAK,MAAM,CAAC,EAE5B,GAAI,IAAQ,OAAS,IAAQ,MAAO,CAClC,EAAS,EACT,QACF,CAEA,GAAI,EAAM,MAAO,CAEf,EAAO,KAAK,CAAE,IAAA,EAAK,KAAM,EAAc,CAAO,CAAE,CAAC,EACjD,QACF,CAGA,MAAM,EAAa,EAAgB,EAAQ,CAAA,GAAM,IAAI,EAC/C,EAAa,EAAgB,EAAQ,CAAA,GAAM,IAAI,EAC/C,EAAY,EAAe,EAAQ,MAAM,CAAC,CAAC,EACjD,EAAO,KAAK,CAAE,IAAA,EAAK,WAAA,EAAY,WAAA,EAAY,UAAA,CAAU,CAAC,CACxD,CAEA,MAAO,CAAE,OAAA,EAAQ,OAAA,CAAO,CAC1B,CAQA,SAAgB,EAAa,EAA4B,CACvD,MAAM,EAAQ,EAAK,QAAQ,QAAS;AAAA,CAAI,EAAE,MAAM;AAAA,CAAI,EAC9C,EAAwB,CAAC,EAC/B,IAAI,EAAmB,CAAC,EAExB,UAAW,KAAQ,EACb,EAAK,KAAK,IAAM,GACd,EAAO,OAAS,IAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,EACrC,EAAS,CAAC,GAGZ,EAAO,KAAK,CAAI,EAIpB,OAAI,EAAO,OAAS,GAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,EAGhC,CACT,CAIA,SAAS,EAAuB,EAA4B,CAC1D,MAAM,EAAkB,CAAC,EAEzB,EAAM,KAAK,SAAS,EAAO,MAAA,EAAQ,EAEnC,UAAW,KAAS,EAAO,OACzB,GAAI,EAAA,eAAe,CAAK,EACtB,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,EAAY,EAAM,IAAI,CAAA,EAAG,MACjD,CACL,MAAM,EAAO,EAAgB,EAAM,UAAU,EACvC,EAAO,EAAgB,EAAM,UAAU,EACvC,EAAY,EAAM,UACrB,IAAK,GAAO,IAAI,EAAG,IAAA,GAAO,EAAY,EAAG,KAAK,CAAA,EAAG,EACjD,KAAK,EAAE,EACV,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,CAAA,GAAO,CAAA,GAAO,CAAA,EAAW,CACxD,CAGF,OAAO,EAAM,KAAK;AAAA,CAAI,EAAI;AAAA,CAC5B,CAMA,SAAgB,EAAiB,EAA+B,CAE9D,OAAO,EAAQ,IAAI,CAAsB,EAAE,KAAK;AAAA,CAAI,CACtD"}
+{"version":3,"file":"marctxt.cjs","names":[],"sources":["../src/marctxt.ts"],"sourcesContent":["/**\n * MARCBreaker (marctxt) parser and serializer.\n *\n * Also known as MARCMaker format. Each field is one line:\n *\n *   =LDR  00706cam a2200217 a 4500\n *   =001  5490\n *   =245  14$aThe Hobbit /$cJ.R.R. Tolkien.\n *   =650  \\1$aHobbits (Fictitious characters)$vFiction.\n *\n * Blank indicators are represented as `\\`. Records are separated by blank lines.\n * Subfield delimiter is `$` followed by a single character code.\n *\n * Curly-brace escape sequences (per LC MARCMaker spec):\n *   `{`  → `{lcub}`   (left curly brace; reserved by the format)\n *   `}`  → `{rcub}`   (right curly brace; reserved by the format)\n *   `$`  → `{dollar}` (subfield delimiter)\n *   `\\`  → `{bsol}`   (backslash; reserved as blank-indicator stand-in)\n *\n */\n\nimport type { MarcRecord, ControlField, DataField, Subfield } from './types';\nimport { isControlField } from './types';\n\n// ─── Indicator encoding ───────────────────────────────────────────────────────\n\nfunction encodeIndicator(ind: string): string {\n  return ind === ' ' ? '\\\\' : ind;\n}\n\nfunction decodeIndicator(ch: string): string {\n  return ch === '\\\\' ? ' ' : ch;\n}\n\n// ─── Value escape (see file header) ───────────────────────────────────────────\n\nfunction escapeValue(s: string): string {\n  // Single-pass replacement avoids the problem of earlier escapes being\n  // re-escaped by later passes (e.g. '{' → '{lcub}' then '}' → '{lcub{rcub}').\n  return s.replace(/[{}\\\\\\n$]/g, (ch) => {\n    if (ch === '{') return '{lcub}';\n    if (ch === '}') return '{rcub}';\n    if (ch === '$') return '{dollar}';\n    if (ch === '\\\\') return '{bsol}';\n    return ' ';\n  });\n}\n\nfunction unescapeValue(s: string): string {\n  return s.replace(/\\{(lcub|rcub|dollar|bsol)\\}/g, (_, name) => {\n    if (name === 'lcub') return '{';\n    if (name === 'rcub') return '}';\n    if (name === 'dollar') return '$';\n    return '\\\\';\n  });\n}\n\n// ─── Subfield parsing ─────────────────────────────────────────────────────────\n\n/**\n * Parse a subfield string like \"$aValue$bOther\" into Subfield objects.\n * Uses split with a capturing group: \"$aFoo$bBar\" → [\"\", \"a\", \"Foo\", \"b\", \"Bar\"].\n * Any character following `$` is treated as a subfield code.\n */\nfunction parseSubfields(str: string): Subfield[] {\n  const parts = str.split(/\\$(.)/);\n  const subfields: Subfield[] = [];\n  // parts[0] is content before the first $ — should be empty for well-formed data\n  for (let i = 1; i < parts.length; i += 2) {\n    subfields.push({ code: parts[i]!, value: unescapeValue(parts[i + 1] ?? '') });\n  }\n  return subfields;\n}\n\n// ─── Record block parser ──────────────────────────────────────────────────────\n\n/**\n * Parse a block of non-empty marctxt lines into a MarcRecord.\n * Each line has the form `=TAG  content`.\n */\nfunction parseRecordLines(lines: string[]): MarcRecord {\n  let leader = '';\n  const fields: (ControlField | DataField)[] = [];\n\n  for (const line of lines) {\n    if (!line.startsWith('=')) continue;\n    const tag = line.slice(1, 4);\n    // positions 4-5 are the two separator spaces; content starts at 6\n    const content = line.slice(6);\n\n    if (tag === 'LDR' || tag === '000') {\n      leader = content;\n      continue;\n    }\n\n    if (tag < '010') {\n      // Control field: content is the raw field data\n      fields.push({ tag, data: unescapeValue(content) });\n      continue;\n    }\n\n    // Data field: first two chars are indicators, rest are subfields\n    const indicator1 = decodeIndicator(content[0] ?? '\\\\');\n    const indicator2 = decodeIndicator(content[1] ?? '\\\\');\n    const subfields = parseSubfields(content.slice(2));\n    fields.push({ tag, indicator1, indicator2, subfields });\n  }\n\n  return { leader, fields };\n}\n\n// ─── Public parse API ─────────────────────────────────────────────────────────\n\n/**\n * Parse a marctxt string containing one or more records separated by blank lines.\n * Returns all records found.\n */\nexport function parseMarcTxt(text: string): MarcRecord[] {\n  const lines = text.replace(/\\r\\n/g, '\\n').split('\\n');\n  const records: MarcRecord[] = [];\n  let buffer: string[] = [];\n\n  for (const line of lines) {\n    if (line.trim() === '') {\n      if (buffer.length > 0) {\n        records.push(parseRecordLines(buffer));\n        buffer = [];\n      }\n    } else {\n      buffer.push(line);\n    }\n  }\n\n  if (buffer.length > 0) {\n    records.push(parseRecordLines(buffer));\n  }\n\n  return records;\n}\n\n// ─── Serializer ───────────────────────────────────────────────────────────────\n\nfunction serializeMarcTxtRecord(record: MarcRecord): string {\n  const lines: string[] = [];\n\n  lines.push(`=LDR  ${record.leader}`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(`=${field.tag}  ${escapeValue(field.data)}`);\n    } else {\n      const ind1 = encodeIndicator(field.indicator1);\n      const ind2 = encodeIndicator(field.indicator2);\n      const subfields = field.subfields.map((sf) => `$${sf.code}${escapeValue(sf.value)}`).join('');\n      lines.push(`=${field.tag}  ${ind1}${ind2}${subfields}`);\n    }\n  }\n\n  return lines.join('\\n') + '\\n';\n}\n\n/**\n * Serialize one or more MarcRecords into a marctxt string.\n * Records are separated by blank lines.\n */\nexport function serializeMarcTxt(records: MarcRecord[]): string {\n  // Each record ends with '\\n'; joining with '\\n' produces blank lines between records.\n  return records.map(serializeMarcTxtRecord).join('\\n');\n}\n"],"mappings":"2GA0BA,SAAS,EAAgB,EAAqB,CAC5C,OAAO,IAAQ,IAAM,KAAO,CAC9B,CAEA,SAAS,EAAgB,EAAoB,CAC3C,OAAO,IAAO,KAAO,IAAM,CAC7B,CAIA,SAAS,EAAY,EAAmB,CAGtC,OAAO,EAAE,QAAQ,aAAe,GAC1B,IAAO,IAAY,SACnB,IAAO,IAAY,SACnB,IAAO,IAAY,WACnB,IAAO,KAAa,SACjB,GACR,CACH,CAEA,SAAS,EAAc,EAAmB,CACxC,OAAO,EAAE,QAAQ,+BAAA,CAAiC,EAAG,IAC/C,IAAS,OAAe,IACxB,IAAS,OAAe,IACxB,IAAS,SAAiB,IACvB,IACR,CACH,CASA,SAAS,EAAe,EAAyB,CAC/C,MAAM,EAAQ,EAAI,MAAM,OAAO,EACzB,EAAwB,CAAC,EAE/B,QAAS,EAAI,EAAG,EAAI,EAAM,OAAQ,GAAK,EACrC,EAAU,KAAK,CAAE,KAAM,EAAM,CAAA,EAAK,MAAO,EAAc,EAAM,EAAI,CAAA,GAAM,EAAE,CAAE,CAAC,EAE9E,OAAO,CACT,CAQA,SAAS,EAAiB,EAA6B,CACrD,IAAI,EAAS,GACb,MAAM,EAAuC,CAAC,EAE9C,UAAW,KAAQ,EAAO,CACxB,GAAI,CAAC,EAAK,WAAW,GAAG,EAAG,SAC3B,MAAM,EAAM,EAAK,MAAM,EAAG,CAAC,EAErB,EAAU,EAAK,MAAM,CAAC,EAE5B,GAAI,IAAQ,OAAS,IAAQ,MAAO,CAClC,EAAS,EACT,QACF,CAEA,GAAI,EAAM,MAAO,CAEf,EAAO,KAAK,CAAE,IAAA,EAAK,KAAM,EAAc,CAAO,CAAE,CAAC,EACjD,QACF,CAGA,MAAM,EAAa,EAAgB,EAAQ,CAAA,GAAM,IAAI,EAC/C,EAAa,EAAgB,EAAQ,CAAA,GAAM,IAAI,EAC/C,EAAY,EAAe,EAAQ,MAAM,CAAC,CAAC,EACjD,EAAO,KAAK,CAAE,IAAA,EAAK,WAAA,EAAY,WAAA,EAAY,UAAA,CAAU,CAAC,CACxD,CAEA,MAAO,CAAE,OAAA,EAAQ,OAAA,CAAO,CAC1B,CAQA,SAAgB,EAAa,EAA4B,CACvD,MAAM,EAAQ,EAAK,QAAQ,QAAS;AAAA,CAAI,EAAE,MAAM;AAAA,CAAI,EAC9C,EAAwB,CAAC,EAC/B,IAAI,EAAmB,CAAC,EAExB,UAAW,KAAQ,EACb,EAAK,KAAK,IAAM,GACd,EAAO,OAAS,IAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,EACrC,EAAS,CAAC,GAGZ,EAAO,KAAK,CAAI,EAIpB,OAAI,EAAO,OAAS,GAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,EAGhC,CACT,CAIA,SAAS,EAAuB,EAA4B,CAC1D,MAAM,EAAkB,CAAC,EAEzB,EAAM,KAAK,SAAS,EAAO,MAAA,EAAQ,EAEnC,UAAW,KAAS,EAAO,OACzB,GAAI,EAAA,eAAe,CAAK,EACtB,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,EAAY,EAAM,IAAI,CAAA,EAAG,MACjD,CACL,MAAM,EAAO,EAAgB,EAAM,UAAU,EACvC,EAAO,EAAgB,EAAM,UAAU,EACvC,EAAY,EAAM,UAAU,IAAK,GAAO,IAAI,EAAG,IAAA,GAAO,EAAY,EAAG,KAAK,CAAA,EAAG,EAAE,KAAK,EAAE,EAC5F,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,CAAA,GAAO,CAAA,GAAO,CAAA,EAAW,CACxD,CAGF,OAAO,EAAM,KAAK;AAAA,CAAI,EAAI;AAAA,CAC5B,CAMA,SAAgB,EAAiB,EAA+B,CAE9D,OAAO,EAAQ,IAAI,CAAsB,EAAE,KAAK;AAAA,CAAI,CACtD"}

package/dist/marctxt.js

@@ -1,4 +1,4 @@
-import { t as b } from "./types-c4Mo9m9u.js";
+import { t as b } from "./types-BMKDHD1l.js";
 function u(e) {
   return e === " " ? "\\" : e;
 }

package/dist/marctxt.js.map

@@ -1 +1 @@
-{"version":3,"file":"marctxt.js","names":[],"sources":["../src/marctxt.ts"],"sourcesContent":["/**\n * MARCBreaker (marctxt) parser and serializer.\n *\n * Also known as MARCMaker format. Each field is one line:\n *\n *   =LDR  00706cam a2200217 a 4500\n *   =001  5490\n *   =245  14$aThe Hobbit /$cJ.R.R. Tolkien.\n *   =650  \\1$aHobbits (Fictitious characters)$vFiction.\n *\n * Blank indicators are represented as `\\`. Records are separated by blank lines.\n * Subfield delimiter is `$` followed by a single character code.\n *\n * Curly-brace escape sequences (per LC MARCMaker spec):\n *   `{`  → `{lcub}`   (left curly brace; reserved by the format)\n *   `}`  → `{rcub}`   (right curly brace; reserved by the format)\n *   `$`  → `{dollar}` (subfield delimiter)\n *   `\\`  → `{bsol}`   (backslash; reserved as blank-indicator stand-in)\n *\n */\n\nimport type { MarcRecord, ControlField, DataField, Subfield } from './types';\nimport { isControlField } from './types';\n\n// ─── Indicator encoding ───────────────────────────────────────────────────────\n\nfunction encodeIndicator(ind: string): string {\n  return ind === ' ' ? '\\\\' : ind;\n}\n\nfunction decodeIndicator(ch: string): string {\n  return ch === '\\\\' ? ' ' : ch;\n}\n\n// ─── Value escape (see file header) ───────────────────────────────────────────\n\nfunction escapeValue(s: string): string {\n  // Single-pass replacement avoids the problem of earlier escapes being\n  // re-escaped by later passes (e.g. '{' → '{lcub}' then '}' → '{lcub{rcub}').\n  return s.replace(/[{}\\\\\\n$]/g, (ch) => {\n    if (ch === '{') return '{lcub}';\n    if (ch === '}') return '{rcub}';\n    if (ch === '$') return '{dollar}';\n    if (ch === '\\\\') return '{bsol}';\n    return ' ';\n  });\n}\n\nfunction unescapeValue(s: string): string {\n  return s.replace(/\\{(lcub|rcub|dollar|bsol)\\}/g, (_, name) => {\n    if (name === 'lcub') return '{';\n    if (name === 'rcub') return '}';\n    if (name === 'dollar') return '$';\n    return '\\\\';\n  });\n}\n\n// ─── Subfield parsing ─────────────────────────────────────────────────────────\n\n/**\n * Parse a subfield string like \"$aValue$bOther\" into Subfield objects.\n * Uses split with a capturing group: \"$aFoo$bBar\" → [\"\", \"a\", \"Foo\", \"b\", \"Bar\"].\n * Any character following `$` is treated as a subfield code.\n */\nfunction parseSubfields(str: string): Subfield[] {\n  const parts = str.split(/\\$(.)/);\n  const subfields: Subfield[] = [];\n  // parts[0] is content before the first $ — should be empty for well-formed data\n  for (let i = 1; i < parts.length; i += 2) {\n    subfields.push({ code: parts[i]!, value: unescapeValue(parts[i + 1] ?? '') });\n  }\n  return subfields;\n}\n\n// ─── Record block parser ──────────────────────────────────────────────────────\n\n/**\n * Parse a block of non-empty marctxt lines into a MarcRecord.\n * Each line has the form `=TAG  content`.\n */\nfunction parseRecordLines(lines: string[]): MarcRecord {\n  let leader = '';\n  const fields: (ControlField | DataField)[] = [];\n\n  for (const line of lines) {\n    if (!line.startsWith('=')) continue;\n    const tag = line.slice(1, 4);\n    // positions 4-5 are the two separator spaces; content starts at 6\n    const content = line.slice(6);\n\n    if (tag === 'LDR' || tag === '000') {\n      leader = content;\n      continue;\n    }\n\n    if (tag < '010') {\n      // Control field: content is the raw field data\n      fields.push({ tag, data: unescapeValue(content) });\n      continue;\n    }\n\n    // Data field: first two chars are indicators, rest are subfields\n    const indicator1 = decodeIndicator(content[0] ?? '\\\\');\n    const indicator2 = decodeIndicator(content[1] ?? '\\\\');\n    const subfields = parseSubfields(content.slice(2));\n    fields.push({ tag, indicator1, indicator2, subfields });\n  }\n\n  return { leader, fields };\n}\n\n// ─── Public parse API ─────────────────────────────────────────────────────────\n\n/**\n * Parse a marctxt string containing one or more records separated by blank lines.\n * Returns all records found.\n */\nexport function parseMarcTxt(text: string): MarcRecord[] {\n  const lines = text.replace(/\\r\\n/g, '\\n').split('\\n');\n  const records: MarcRecord[] = [];\n  let buffer: string[] = [];\n\n  for (const line of lines) {\n    if (line.trim() === '') {\n      if (buffer.length > 0) {\n        records.push(parseRecordLines(buffer));\n        buffer = [];\n      }\n    } else {\n      buffer.push(line);\n    }\n  }\n\n  if (buffer.length > 0) {\n    records.push(parseRecordLines(buffer));\n  }\n\n  return records;\n}\n\n// ─── Serializer ───────────────────────────────────────────────────────────────\n\nfunction serializeMarcTxtRecord(record: MarcRecord): string {\n  const lines: string[] = [];\n\n  lines.push(`=LDR  ${record.leader}`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(`=${field.tag}  ${escapeValue(field.data)}`);\n    } else {\n      const ind1 = encodeIndicator(field.indicator1);\n      const ind2 = encodeIndicator(field.indicator2);\n      const subfields = field.subfields\n        .map((sf) => `$${sf.code}${escapeValue(sf.value)}`)\n        .join('');\n      lines.push(`=${field.tag}  ${ind1}${ind2}${subfields}`);\n    }\n  }\n\n  return lines.join('\\n') + '\\n';\n}\n\n/**\n * Serialize one or more MarcRecords into a marctxt string.\n * Records are separated by blank lines.\n */\nexport function serializeMarcTxt(records: MarcRecord[]): string {\n  // Each record ends with '\\n'; joining with '\\n' produces blank lines between records.\n  return records.map(serializeMarcTxtRecord).join('\\n');\n}\n"],"mappings":";AA0BA,SAAS,EAAgB,GAAqB;AAC5C,SAAO,MAAQ,MAAM,OAAO;AAC9B;AAEA,SAAS,EAAgB,GAAoB;AAC3C,SAAO,MAAO,OAAO,MAAM;AAC7B;AAIA,SAAS,EAAY,GAAmB;AAGtC,SAAO,EAAE,QAAQ,cAAA,CAAe,MAC1B,MAAO,MAAY,WACnB,MAAO,MAAY,WACnB,MAAO,MAAY,aACnB,MAAO,OAAa,WACjB,GACR;AACH;AAEA,SAAS,EAAc,GAAmB;AACxC,SAAO,EAAE,QAAQ,gCAAA,CAAiC,GAAG,MAC/C,MAAS,SAAe,MACxB,MAAS,SAAe,MACxB,MAAS,WAAiB,MACvB,IACR;AACH;AASA,SAAS,EAAe,GAAyB;AAC/C,QAAM,IAAQ,EAAI,MAAM,OAAO,GACzB,IAAwB,CAAC;AAE/B,WAAS,IAAI,GAAG,IAAI,EAAM,QAAQ,KAAK,EACrC,CAAA,EAAU,KAAK;AAAA,IAAE,MAAM,EAAM,CAAA;AAAA,IAAK,OAAO,EAAc,EAAM,IAAI,CAAA,KAAM,EAAE;AAAA,EAAE,CAAC;AAE9E,SAAO;AACT;AAQA,SAAS,EAAiB,GAA6B;AACrD,MAAI,IAAS;AACb,QAAM,IAAuC,CAAC;AAE9C,aAAW,KAAQ,GAAO;AACxB,QAAI,CAAC,EAAK,WAAW,GAAG,EAAG;AAC3B,UAAM,IAAM,EAAK,MAAM,GAAG,CAAC,GAErB,IAAU,EAAK,MAAM,CAAC;AAE5B,QAAI,MAAQ,SAAS,MAAQ,OAAO;AAClC,MAAA,IAAS;AACT;AAAA,IACF;AAEA,QAAI,IAAM,OAAO;AAEf,MAAA,EAAO,KAAK;AAAA,QAAE,KAAA;AAAA,QAAK,MAAM,EAAc,CAAO;AAAA,MAAE,CAAC;AACjD;AAAA,IACF;AAGA,UAAM,IAAa,EAAgB,EAAQ,CAAA,KAAM,IAAI,GAC/C,IAAa,EAAgB,EAAQ,CAAA,KAAM,IAAI,GAC/C,IAAY,EAAe,EAAQ,MAAM,CAAC,CAAC;AACjD,IAAA,EAAO,KAAK;AAAA,MAAE,KAAA;AAAA,MAAK,YAAA;AAAA,MAAY,YAAA;AAAA,MAAY,WAAA;AAAA,IAAU,CAAC;AAAA,EACxD;AAEA,SAAO;AAAA,IAAE,QAAA;AAAA,IAAQ,QAAA;AAAA,EAAO;AAC1B;AAQA,SAAgB,EAAa,GAA4B;AACvD,QAAM,IAAQ,EAAK,QAAQ,SAAS;AAAA,CAAI,EAAE,MAAM;AAAA,CAAI,GAC9C,IAAwB,CAAC;AAC/B,MAAI,IAAmB,CAAC;AAExB,aAAW,KAAQ,EACjB,CAAI,EAAK,KAAK,MAAM,KACd,EAAO,SAAS,MAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,GACrC,IAAS,CAAC,KAGZ,EAAO,KAAK,CAAI;AAIpB,SAAI,EAAO,SAAS,KAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,GAGhC;AACT;AAIA,SAAS,EAAuB,GAA4B;AAC1D,QAAM,IAAkB,CAAC;AAEzB,EAAA,EAAM,KAAK,SAAS,EAAO,MAAA,EAAQ;AAEnC,aAAW,KAAS,EAAO,OACzB,KAAI,EAAe,CAAK,EACtB,CAAA,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,EAAY,EAAM,IAAI,CAAA,EAAG;AAAA,OACjD;AACL,UAAM,IAAO,EAAgB,EAAM,UAAU,GACvC,IAAO,EAAgB,EAAM,UAAU,GACvC,IAAY,EAAM,UACrB,IAAA,CAAK,MAAO,IAAI,EAAG,IAAA,GAAO,EAAY,EAAG,KAAK,CAAA,EAAG,EACjD,KAAK,EAAE;AACV,IAAA,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,CAAA,GAAO,CAAA,GAAO,CAAA,EAAW;AAAA,EACxD;AAGF,SAAO,EAAM,KAAK;AAAA,CAAI,IAAI;AAAA;AAC5B;AAMA,SAAgB,EAAiB,GAA+B;AAE9D,SAAO,EAAQ,IAAI,CAAsB,EAAE,KAAK;AAAA,CAAI;AACtD"}
+{"version":3,"file":"marctxt.js","names":[],"sources":["../src/marctxt.ts"],"sourcesContent":["/**\n * MARCBreaker (marctxt) parser and serializer.\n *\n * Also known as MARCMaker format. Each field is one line:\n *\n *   =LDR  00706cam a2200217 a 4500\n *   =001  5490\n *   =245  14$aThe Hobbit /$cJ.R.R. Tolkien.\n *   =650  \\1$aHobbits (Fictitious characters)$vFiction.\n *\n * Blank indicators are represented as `\\`. Records are separated by blank lines.\n * Subfield delimiter is `$` followed by a single character code.\n *\n * Curly-brace escape sequences (per LC MARCMaker spec):\n *   `{`  → `{lcub}`   (left curly brace; reserved by the format)\n *   `}`  → `{rcub}`   (right curly brace; reserved by the format)\n *   `$`  → `{dollar}` (subfield delimiter)\n *   `\\`  → `{bsol}`   (backslash; reserved as blank-indicator stand-in)\n *\n */\n\nimport type { MarcRecord, ControlField, DataField, Subfield } from './types';\nimport { isControlField } from './types';\n\n// ─── Indicator encoding ───────────────────────────────────────────────────────\n\nfunction encodeIndicator(ind: string): string {\n  return ind === ' ' ? '\\\\' : ind;\n}\n\nfunction decodeIndicator(ch: string): string {\n  return ch === '\\\\' ? ' ' : ch;\n}\n\n// ─── Value escape (see file header) ───────────────────────────────────────────\n\nfunction escapeValue(s: string): string {\n  // Single-pass replacement avoids the problem of earlier escapes being\n  // re-escaped by later passes (e.g. '{' → '{lcub}' then '}' → '{lcub{rcub}').\n  return s.replace(/[{}\\\\\\n$]/g, (ch) => {\n    if (ch === '{') return '{lcub}';\n    if (ch === '}') return '{rcub}';\n    if (ch === '$') return '{dollar}';\n    if (ch === '\\\\') return '{bsol}';\n    return ' ';\n  });\n}\n\nfunction unescapeValue(s: string): string {\n  return s.replace(/\\{(lcub|rcub|dollar|bsol)\\}/g, (_, name) => {\n    if (name === 'lcub') return '{';\n    if (name === 'rcub') return '}';\n    if (name === 'dollar') return '$';\n    return '\\\\';\n  });\n}\n\n// ─── Subfield parsing ─────────────────────────────────────────────────────────\n\n/**\n * Parse a subfield string like \"$aValue$bOther\" into Subfield objects.\n * Uses split with a capturing group: \"$aFoo$bBar\" → [\"\", \"a\", \"Foo\", \"b\", \"Bar\"].\n * Any character following `$` is treated as a subfield code.\n */\nfunction parseSubfields(str: string): Subfield[] {\n  const parts = str.split(/\\$(.)/);\n  const subfields: Subfield[] = [];\n  // parts[0] is content before the first $ — should be empty for well-formed data\n  for (let i = 1; i < parts.length; i += 2) {\n    subfields.push({ code: parts[i]!, value: unescapeValue(parts[i + 1] ?? '') });\n  }\n  return subfields;\n}\n\n// ─── Record block parser ──────────────────────────────────────────────────────\n\n/**\n * Parse a block of non-empty marctxt lines into a MarcRecord.\n * Each line has the form `=TAG  content`.\n */\nfunction parseRecordLines(lines: string[]): MarcRecord {\n  let leader = '';\n  const fields: (ControlField | DataField)[] = [];\n\n  for (const line of lines) {\n    if (!line.startsWith('=')) continue;\n    const tag = line.slice(1, 4);\n    // positions 4-5 are the two separator spaces; content starts at 6\n    const content = line.slice(6);\n\n    if (tag === 'LDR' || tag === '000') {\n      leader = content;\n      continue;\n    }\n\n    if (tag < '010') {\n      // Control field: content is the raw field data\n      fields.push({ tag, data: unescapeValue(content) });\n      continue;\n    }\n\n    // Data field: first two chars are indicators, rest are subfields\n    const indicator1 = decodeIndicator(content[0] ?? '\\\\');\n    const indicator2 = decodeIndicator(content[1] ?? '\\\\');\n    const subfields = parseSubfields(content.slice(2));\n    fields.push({ tag, indicator1, indicator2, subfields });\n  }\n\n  return { leader, fields };\n}\n\n// ─── Public parse API ─────────────────────────────────────────────────────────\n\n/**\n * Parse a marctxt string containing one or more records separated by blank lines.\n * Returns all records found.\n */\nexport function parseMarcTxt(text: string): MarcRecord[] {\n  const lines = text.replace(/\\r\\n/g, '\\n').split('\\n');\n  const records: MarcRecord[] = [];\n  let buffer: string[] = [];\n\n  for (const line of lines) {\n    if (line.trim() === '') {\n      if (buffer.length > 0) {\n        records.push(parseRecordLines(buffer));\n        buffer = [];\n      }\n    } else {\n      buffer.push(line);\n    }\n  }\n\n  if (buffer.length > 0) {\n    records.push(parseRecordLines(buffer));\n  }\n\n  return records;\n}\n\n// ─── Serializer ───────────────────────────────────────────────────────────────\n\nfunction serializeMarcTxtRecord(record: MarcRecord): string {\n  const lines: string[] = [];\n\n  lines.push(`=LDR  ${record.leader}`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(`=${field.tag}  ${escapeValue(field.data)}`);\n    } else {\n      const ind1 = encodeIndicator(field.indicator1);\n      const ind2 = encodeIndicator(field.indicator2);\n      const subfields = field.subfields.map((sf) => `$${sf.code}${escapeValue(sf.value)}`).join('');\n      lines.push(`=${field.tag}  ${ind1}${ind2}${subfields}`);\n    }\n  }\n\n  return lines.join('\\n') + '\\n';\n}\n\n/**\n * Serialize one or more MarcRecords into a marctxt string.\n * Records are separated by blank lines.\n */\nexport function serializeMarcTxt(records: MarcRecord[]): string {\n  // Each record ends with '\\n'; joining with '\\n' produces blank lines between records.\n  return records.map(serializeMarcTxtRecord).join('\\n');\n}\n"],"mappings":";AA0BA,SAAS,EAAgB,GAAqB;AAC5C,SAAO,MAAQ,MAAM,OAAO;AAC9B;AAEA,SAAS,EAAgB,GAAoB;AAC3C,SAAO,MAAO,OAAO,MAAM;AAC7B;AAIA,SAAS,EAAY,GAAmB;AAGtC,SAAO,EAAE,QAAQ,cAAA,CAAe,MAC1B,MAAO,MAAY,WACnB,MAAO,MAAY,WACnB,MAAO,MAAY,aACnB,MAAO,OAAa,WACjB,GACR;AACH;AAEA,SAAS,EAAc,GAAmB;AACxC,SAAO,EAAE,QAAQ,gCAAA,CAAiC,GAAG,MAC/C,MAAS,SAAe,MACxB,MAAS,SAAe,MACxB,MAAS,WAAiB,MACvB,IACR;AACH;AASA,SAAS,EAAe,GAAyB;AAC/C,QAAM,IAAQ,EAAI,MAAM,OAAO,GACzB,IAAwB,CAAC;AAE/B,WAAS,IAAI,GAAG,IAAI,EAAM,QAAQ,KAAK,EACrC,CAAA,EAAU,KAAK;AAAA,IAAE,MAAM,EAAM,CAAA;AAAA,IAAK,OAAO,EAAc,EAAM,IAAI,CAAA,KAAM,EAAE;AAAA,EAAE,CAAC;AAE9E,SAAO;AACT;AAQA,SAAS,EAAiB,GAA6B;AACrD,MAAI,IAAS;AACb,QAAM,IAAuC,CAAC;AAE9C,aAAW,KAAQ,GAAO;AACxB,QAAI,CAAC,EAAK,WAAW,GAAG,EAAG;AAC3B,UAAM,IAAM,EAAK,MAAM,GAAG,CAAC,GAErB,IAAU,EAAK,MAAM,CAAC;AAE5B,QAAI,MAAQ,SAAS,MAAQ,OAAO;AAClC,MAAA,IAAS;AACT;AAAA,IACF;AAEA,QAAI,IAAM,OAAO;AAEf,MAAA,EAAO,KAAK;AAAA,QAAE,KAAA;AAAA,QAAK,MAAM,EAAc,CAAO;AAAA,MAAE,CAAC;AACjD;AAAA,IACF;AAGA,UAAM,IAAa,EAAgB,EAAQ,CAAA,KAAM,IAAI,GAC/C,IAAa,EAAgB,EAAQ,CAAA,KAAM,IAAI,GAC/C,IAAY,EAAe,EAAQ,MAAM,CAAC,CAAC;AACjD,IAAA,EAAO,KAAK;AAAA,MAAE,KAAA;AAAA,MAAK,YAAA;AAAA,MAAY,YAAA;AAAA,MAAY,WAAA;AAAA,IAAU,CAAC;AAAA,EACxD;AAEA,SAAO;AAAA,IAAE,QAAA;AAAA,IAAQ,QAAA;AAAA,EAAO;AAC1B;AAQA,SAAgB,EAAa,GAA4B;AACvD,QAAM,IAAQ,EAAK,QAAQ,SAAS;AAAA,CAAI,EAAE,MAAM;AAAA,CAAI,GAC9C,IAAwB,CAAC;AAC/B,MAAI,IAAmB,CAAC;AAExB,aAAW,KAAQ,EACjB,CAAI,EAAK,KAAK,MAAM,KACd,EAAO,SAAS,MAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,GACrC,IAAS,CAAC,KAGZ,EAAO,KAAK,CAAI;AAIpB,SAAI,EAAO,SAAS,KAClB,EAAQ,KAAK,EAAiB,CAAM,CAAC,GAGhC;AACT;AAIA,SAAS,EAAuB,GAA4B;AAC1D,QAAM,IAAkB,CAAC;AAEzB,EAAA,EAAM,KAAK,SAAS,EAAO,MAAA,EAAQ;AAEnC,aAAW,KAAS,EAAO,OACzB,KAAI,EAAe,CAAK,EACtB,CAAA,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,EAAY,EAAM,IAAI,CAAA,EAAG;AAAA,OACjD;AACL,UAAM,IAAO,EAAgB,EAAM,UAAU,GACvC,IAAO,EAAgB,EAAM,UAAU,GACvC,IAAY,EAAM,UAAU,IAAA,CAAK,MAAO,IAAI,EAAG,IAAA,GAAO,EAAY,EAAG,KAAK,CAAA,EAAG,EAAE,KAAK,EAAE;AAC5F,IAAA,EAAM,KAAK,IAAI,EAAM,GAAA,KAAQ,CAAA,GAAO,CAAA,GAAO,CAAA,EAAW;AAAA,EACxD;AAGF,SAAO,EAAM,KAAK;AAAA,CAAI,IAAI;AAAA;AAC5B;AAMA,SAAgB,EAAiB,GAA+B;AAE9D,SAAO,EAAQ,IAAI,CAAsB,EAAE,KAAK;AAAA,CAAI;AACtD"}

package/dist/marcxml.cjs

@@ -1,8 +1,8 @@
-Object.defineProperty(exports,Symbol.toStringTag,{value:"Module"});const $=require("./types-CJcxHJff.cjs");var b=new Map([["amp","&"],["lt","<"],["gt",">"],["quot",'"'],["apos","'"]]);function g(t){return t.replace(/&(?:#x([0-9a-fA-F]+)|#([0-9]+)|([a-zA-Z]+));/g,(i,r,s,e)=>{if(r!==void 0){const n=parseInt(r,16);return n>=0&&n<=1114111?String.fromCodePoint(n):"�"}if(s!==void 0){const n=parseInt(s,10);return n>=0&&n<=1114111?String.fromCodePoint(n):"�"}return b.get(e)??i})}function u(t){return t.replace(/[\x00-\x08\x0B\x0C\x0E-\x1F]/g,"�").replace(/&/g,"&amp;").replace(/</g,"&lt;").replace(/>/g,"&gt;").replace(/"/g,"&quot;").replace(/\r/g,"&#13;")}function f(t){const i=t.indexOf(":");return i===-1?t:t.slice(i+1)}function m(t){const i={},r=/([a-zA-Z_:][^\s=]*)\s*=\s*(?:"([^"]*)"|'([^']*)')/g;let s;for(;(s=r.exec(t))!==null;){const e=f(s[1]);i[e]=g(s[2]??s[3]??"")}return i}function v(t){const i=[];let r=0;for(;r<t.length;){const s=t.indexOf("<",r);if(s===-1){t.slice(r).trim()&&i.push({type:"text",text:g(t.slice(r))});break}if(s>r){const c=t.slice(r,s);c.trim()&&i.push({type:"text",text:g(c)})}const e=t.indexOf(">",s);if(e===-1)break;const n=t.slice(s+1,e);if(n.startsWith("!")||n.startsWith("?")){r=e+1;continue}if(n.startsWith("/"))i.push({type:"close",name:f(n.slice(1).trim())});else if(n.endsWith("/")){const c=n.slice(0,-1).trim(),o=c.search(/\s/),l=o===-1?c:c.slice(0,o),p=o===-1?"":c.slice(o);i.push({type:"self-close",name:f(l),attrs:m(p)})}else{const c=n.search(/\s/),o=c===-1?n:n.slice(0,c),l=c===-1?"":n.slice(c);i.push({type:"open",name:f(o),attrs:m(l)})}r=e+1}return i}function M(t,i){let r="";const s=[];let e=i;for(;e<t.length;){const n=t[e];if(n.type==="close"&&n.name==="record")return{record:{leader:r,fields:s},end:e+1};if(n.type==="open"&&n.name==="leader"){e++,e<t.length&&t[e].type==="text"&&(r=t[e].text.trim(),e++),e<t.length&&t[e].type==="close"&&e++;continue}if(n.type==="self-close"&&n.name==="controlfield"){s.push({tag:n.attrs?.tag??"",data:""}),e++;continue}if(n.type==="open"&&n.name==="controlfield"){const c=n.attrs?.tag??"";e++;let o="";e<t.length&&t[e].type==="text"&&(o=t[e].text??"",e++),e<t.length&&t[e].type==="close"&&e++,s.push({tag:c,data:o});continue}if(n.type==="self-close"&&n.name==="datafield"){s.push({tag:n.attrs?.tag??"",indicator1:n.attrs?.ind1??" ",indicator2:n.attrs?.ind2??" ",subfields:[]}),e++;continue}if(n.type==="open"&&n.name==="datafield"){const c=n.attrs?.tag??"",o=n.attrs?.ind1??" ",l=n.attrs?.ind2??" ",p=[];for(e++;e<t.length;){const d=t[e];if(d.type==="close"&&d.name==="datafield"){e++;break}if(d.type==="open"&&d.name==="subfield"){const x=d.attrs?.code??"";e++;let h="";e<t.length&&t[e].type==="text"&&(h=t[e].text??"",e++),e<t.length&&t[e].type==="close"&&e++,p.push({code:x,value:h});continue}e++}s.push({tag:c,indicator1:o,indicator2:l,subfields:p});continue}e++}return{record:{leader:r,fields:s},end:e}}function X(t){const i=v(t),r=[];let s=0;for(;s<i.length;){const e=i[s];if(e.type==="open"&&e.name==="record"){const{record:n,end:c}=M(i,s+1);r.push(n),s=c;continue}s++}return r}var w=`<?xml version="1.0" encoding="UTF-8"?>
-`,y='xmlns="http://www.loc.gov/MARC21/slim"',a="  ";function A(t){const i=[`<record ${y}>`];i.push(`${a}<leader>${u(t.leader)}</leader>`);for(const r of t.fields)if($.isControlField(r))i.push(`${a}<controlfield tag="${r.tag}">${u(r.data)}</controlfield>`);else{const s=r.indicator1===" "?" ":r.indicator1,e=r.indicator2===" "?" ":r.indicator2;i.push(`${a}<datafield tag="${r.tag}" ind1="${s}" ind2="${e}">`);for(const n of r.subfields)i.push(`${a}${a}<subfield code="${n.code}">${u(n.value)}</subfield>`);i.push(`${a}</datafield>`)}return i.push("</record>"),i.join(`
-`)}function C(t){const i=[w,`<collection ${y}>`];for(const r of t){const s=A(r).split(`
-`).map(e=>a+e).join(`
-`);i.push(s)}return i.push("</collection>"),i.join(`
-`)}exports.parseMarcXml=X;exports.serializeMarcXml=C;
+Object.defineProperty(exports,Symbol.toStringTag,{value:"Module"});const M=require("./types-CsOhH4OF.cjs"),d=require("./warnings-6yoB06xI.cjs");var A=new Map([["amp","&"],["lt","<"],["gt",">"],["quot",'"'],["apos","'"]]);function y(e){return e.replace(/&(?:#x([0-9a-fA-F]+)|#([0-9]+)|([a-zA-Z]+));/g,(s,n,i,r)=>{if(n!==void 0){const c=parseInt(n,16);return c>=0&&c<=1114111?String.fromCodePoint(c):"�"}if(i!==void 0){const c=parseInt(i,10);return c>=0&&c<=1114111?String.fromCodePoint(c):"�"}return A.get(r)??s})}function u(e){return e.replace(/[\x00-\x08\x0B\x0C\x0E-\x1F]/g,"�").replace(/&/g,"&amp;").replace(/</g,"&lt;").replace(/>/g,"&gt;").replace(/"/g,"&quot;").replace(/\r/g,"&#13;")}function m(e){const s=e.indexOf(":");return s===-1?e:e.slice(s+1)}function _(e){const s={},n=/([a-zA-Z_:][^\s=]*)\s*=\s*(?:"([^"]*)"|'([^']*)')/g;let i;for(;(i=n.exec(e))!==null;){const r=m(i[1]);s[r]=y(i[2]??i[3]??"")}return s}function T(e,s){let n=!1,i=!1;for(let r=s;r<e.length;r++){const c=e[r];if(c==='"'&&!n)i=!i;else if(c==="'"&&!i)n=!n;else if(c===">"&&!n&&!i)return r}return-1}function X(e,s){const n=[];let i=0;for(;i<e.length;){const r=e.indexOf("<",i);if(r===-1){e.slice(i).trim()&&n.push({type:"text",text:y(e.slice(i))});break}if(r>i){const t=e.slice(i,r);t.trim()&&n.push({type:"text",text:y(t)})}if(e.startsWith("<!--",r)){const t=e.indexOf("-->",r+4);t===-1?(s?.push(d.createWarning("malformed_xml","Unterminated comment")),i=e.length):i=t+3;continue}if(e.startsWith("<![CDATA[",r)){const t=e.indexOf("]]>",r+9);if(t===-1)s?.push(d.createWarning("malformed_xml","Unterminated CDATA section")),i=e.length;else{const a=e.slice(r+9,t);a&&n.push({type:"text",text:a}),i=t+3}continue}if(e.startsWith("<?",r)){const t=e.indexOf("?>",r+2);i=t===-1?e.length:t+2;continue}if(e.startsWith("<!",r)){const t=e.indexOf(">",r+2);i=t===-1?e.length:t+1;continue}const c=T(e,r+1);if(c===-1){s?.push(d.createWarning("malformed_xml","Unclosed tag at end of input"));break}const o=e.slice(r+1,c);if(o.startsWith("/"))n.push({type:"close",name:m(o.slice(1).trim())});else if(o.endsWith("/")){const t=o.slice(0,-1).trim(),a=t.search(/\s/),l=a===-1?t:t.slice(0,a),f=a===-1?"":t.slice(a);n.push({type:"self-close",name:m(l),attrs:_(f)})}else{const t=o.search(/\s/),a=t===-1?o:o.slice(0,t),l=t===-1?"":o.slice(t);n.push({type:"open",name:m(a),attrs:_(l)})}i=c+1}return n}function g(e,s,n){if(n.strict)throw new Error(s.message);const i=n.maxWarnings??100;e.length<i&&e.push(s)}function C(e,s,n,i){let r="",c=!1;const o=[];let t=s;for(;t<e.length;){const a=e[t];if(a.type==="close"&&a.name==="record")return c||g(n,d.createWarning("missing_element","Record has no <leader> element"),i),{record:{leader:r,fields:o},end:t+1};if(a.type==="open"&&a.name==="leader"){c=!0,t++,t<e.length&&e[t].type==="text"&&(r=e[t].text.trim(),t++),r.length!==24&&g(n,d.createWarning("invalid_leader",`Leader is ${r.length} characters, expected 24`),i),t<e.length&&e[t].type==="close"&&t++;continue}if(a.type==="self-close"&&a.name==="controlfield"){const l=a.attrs?.tag;l===void 0&&g(n,d.createWarning("missing_element","controlfield missing tag attribute"),i),o.push({tag:l??"",data:""}),t++;continue}if(a.type==="open"&&a.name==="controlfield"){const l=a.attrs?.tag;l===void 0&&g(n,d.createWarning("missing_element","controlfield missing tag attribute"),i),t++;let f="";t<e.length&&e[t].type==="text"&&(f=e[t].text??"",t++),t<e.length&&e[t].type==="close"&&t++,o.push({tag:l??"",data:f});continue}if(a.type==="self-close"&&a.name==="datafield"){const l=a.attrs?.tag;l===void 0&&g(n,d.createWarning("missing_element","datafield missing tag attribute"),i),o.push({tag:l??"",indicator1:a.attrs?.ind1??" ",indicator2:a.attrs?.ind2??" ",subfields:[]}),t++;continue}if(a.type==="open"&&a.name==="datafield"){const l=a.attrs?.tag;l===void 0&&g(n,d.createWarning("missing_element","datafield missing tag attribute"),i);const f=a.attrs?.ind1??" ",E=a.attrs?.ind2??" ",W=[];for(t++;t<e.length;){const h=e[t];if(h.type==="close"&&h.name==="datafield"){t++;break}if(h.type==="open"&&h.name==="subfield"){const b=h.attrs?.code;b===void 0&&g(n,d.createWarning("missing_element","subfield missing code attribute",void 0,l),i),t++;let v="";t<e.length&&e[t].type==="text"&&(v=e[t].text??"",t++),t<e.length&&e[t].type==="close"&&t++,W.push({code:b??"",value:v});continue}t++}o.push({tag:l??"",indicator1:f,indicator2:E,subfields:W});continue}t++}return c||g(n,d.createWarning("missing_element","Record has no <leader> element"),i),{record:{leader:r,fields:o},end:t}}function $(e,s){const n=s??{},i=[],r=X(e,i),c=[];let o=0;for(;o<r.length;){const t=r[o];if(t.type==="open"&&t.name==="record"){const a=[],{record:l,end:f}=C(r,o+1,a,n);c.push({record:l,warnings:a}),o=f;continue}o++}return c.length===0&&i.length>0?c.push({record:null,warnings:i}):i.length>0&&c.length>0&&(c[0]={record:c[0].record,warnings:[...i,...c[0].warnings]}),{results:c}}function O(e,s){return $(e,s).results.map(n=>n.record).filter(n=>n!==null)}var S=`<?xml version="1.0" encoding="UTF-8"?>
+`,x='xmlns="http://www.loc.gov/MARC21/slim"',p="  ";function I(e){const s=[`<record ${x}>`];s.push(`${p}<leader>${u(e.leader)}</leader>`);for(const n of e.fields)if(M.isControlField(n))s.push(`${p}<controlfield tag="${u(n.tag)}">${u(n.data)}</controlfield>`);else{const i=n.indicator1===" "?" ":n.indicator1,r=n.indicator2===" "?" ":n.indicator2;s.push(`${p}<datafield tag="${u(n.tag)}" ind1="${u(i)}" ind2="${u(r)}">`);for(const c of n.subfields)s.push(`${p}${p}<subfield code="${u(c.code)}">${u(c.value)}</subfield>`);s.push(`${p}</datafield>`)}return s.push("</record>"),s.join(`
+`)}function q(e){const s=[S,`<collection ${x}>`];for(const n of e){const i=I(n).split(`
+`).map(r=>p+r).join(`
+`);s.push(i)}return s.push("</collection>"),s.join(`
+`)}exports.parseMarcXml=O;exports.parseMarcXmlWithWarnings=$;exports.serializeMarcXml=q;
 
 //# sourceMappingURL=marcxml.cjs.map

package/dist/marcxml.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"marcxml.cjs","names":[],"sources":["../src/marcxml.ts"],"sourcesContent":["/**\n * MARCXML parser and serializer.\n *\n * Supports the Library of Congress MARCXML schema:\n * http://www.loc.gov/MARC21/slim\n *\n * Parsing is done with a hand-rolled state machine — no XML library needed.\n * The MARCXML format is sufficiently regular (fixed element names, no arbitrary\n * nesting) that a full DOM parser is unnecessary.\n */\n\nimport type { MarcRecord, ControlField, DataField, Subfield } from './types';\nimport { isControlField } from './types';\n\n// ─── XML entity handling ─────────────────────────────────────────────────────\n\nconst ENTITY_MAP: ReadonlyMap<string, string> = new Map([\n  ['amp', '&'],\n  ['lt', '<'],\n  ['gt', '>'],\n  ['quot', '\"'],\n  ['apos', \"'\"],\n]);\n\nfunction unescapeXml(text: string): string {\n  return text.replace(/&(?:#x([0-9a-fA-F]+)|#([0-9]+)|([a-zA-Z]+));/g, (_, hex, dec, name) => {\n    if (hex !== undefined) {\n      const cp = parseInt(hex, 16);\n      return cp >= 0 && cp <= 0x10ffff ? String.fromCodePoint(cp) : '�';\n    }\n    if (dec !== undefined) {\n      const cp = parseInt(dec, 10);\n      return cp >= 0 && cp <= 0x10ffff ? String.fromCodePoint(cp) : '�';\n    }\n    return ENTITY_MAP.get(name) ?? _;\n  });\n}\n\nfunction escapeXml(text: string): string {\n  return text\n    // XML 1.0 forbids most C0 control characters in document text. There is no\n    // valid XML 1.0 representation for them, so substitute the Unicode\n    // replacement character to keep the output well-formed.\n    .replace(/[\\x00-\\x08\\x0B\\x0C\\x0E-\\x1F]/g, '�')\n    .replace(/&/g, '&amp;')\n    .replace(/</g, '&lt;')\n    .replace(/>/g, '&gt;')\n    .replace(/\"/g, '&quot;')\n    // Preserve literal CR through the XML round-trip: XML parsers normalize\n    // bare \\r and \\r\\n to \\n, so we must encode CR as a numeric reference.\n    .replace(/\\r/g, '&#13;');\n}\n\n// ─── Minimal tokeniser ────────────────────────────────────────────────────────\n\ninterface Token {\n  type: 'open' | 'close' | 'self-close' | 'text';\n  /** Local name (no namespace prefix) */\n  name?: string;\n  attrs?: Record<string, string>;\n  text?: string;\n}\n\n/**\n * Strip namespace prefix from a tag name, e.g. \"marc:record\" → \"record\".\n */\nfunction localName(raw: string): string {\n  const colon = raw.indexOf(':');\n  return colon === -1 ? raw : raw.slice(colon + 1);\n}\n\n/**\n * Parse `key=\"value\"` pairs out of an attribute string.\n */\nfunction parseAttrs(attrStr: string): Record<string, string> {\n  const attrs: Record<string, string> = {};\n  const re = /([a-zA-Z_:][^\\s=]*)\\s*=\\s*(?:\"([^\"]*)\"|'([^']*)')/g;\n  let m: RegExpExecArray | null;\n  while ((m = re.exec(attrStr)) !== null) {\n    const key = localName(m[1]!);\n    attrs[key] = unescapeXml(m[2] ?? m[3] ?? '');\n  }\n  return attrs;\n}\n\n/**\n * Tokenise an XML string into a flat stream of open/close/text tokens.\n * Skips processing instructions, comments, and DOCTYPE declarations.\n * Sufficient for the well-constrained MARCXML format.\n */\nfunction tokenise(xml: string): Token[] {\n  const tokens: Token[] = [];\n  let i = 0;\n\n  while (i < xml.length) {\n    const ltPos = xml.indexOf('<', i);\n\n    // Text node before next tag\n    if (ltPos === -1) {\n      const text = xml.slice(i).trim();\n      if (text) tokens.push({ type: 'text', text: unescapeXml(xml.slice(i)) });\n      break;\n    }\n\n    if (ltPos > i) {\n      const raw = xml.slice(i, ltPos);\n      const text = raw.trim();\n      if (text) tokens.push({ type: 'text', text: unescapeXml(raw) });\n    }\n\n    const gtPos = xml.indexOf('>', ltPos);\n    if (gtPos === -1) break;\n\n    const tag = xml.slice(ltPos + 1, gtPos);\n\n    // Skip comments, PIs, DOCTYPE\n    if (tag.startsWith('!') || tag.startsWith('?')) {\n      i = gtPos + 1;\n      continue;\n    }\n\n    if (tag.startsWith('/')) {\n      tokens.push({ type: 'close', name: localName(tag.slice(1).trim()) });\n    } else if (tag.endsWith('/')) {\n      const inner = tag.slice(0, -1).trim();\n      const spaceIdx = inner.search(/\\s/);\n      const name = spaceIdx === -1 ? inner : inner.slice(0, spaceIdx);\n      const attrStr = spaceIdx === -1 ? '' : inner.slice(spaceIdx);\n      tokens.push({ type: 'self-close', name: localName(name), attrs: parseAttrs(attrStr) });\n    } else {\n      const spaceIdx = tag.search(/\\s/);\n      const name = spaceIdx === -1 ? tag : tag.slice(0, spaceIdx);\n      const attrStr = spaceIdx === -1 ? '' : tag.slice(spaceIdx);\n      tokens.push({ type: 'open', name: localName(name), attrs: parseAttrs(attrStr) });\n    }\n\n    i = gtPos + 1;\n  }\n\n  return tokens;\n}\n\n// ─── MARCXML parser ───────────────────────────────────────────────────────────\n\n/**\n * Parse one `<record>` element's worth of tokens into a MarcRecord.\n * Mutates `pos` via the returned index.\n */\nfunction parseRecordTokens(tokens: Token[], start: number): { record: MarcRecord; end: number } {\n  let leader = '';\n  const fields: (ControlField | DataField)[] = [];\n  let i = start;\n\n  while (i < tokens.length) {\n    const tok = tokens[i]!;\n\n    if (tok.type === 'close' && tok.name === 'record') {\n      return { record: { leader, fields }, end: i + 1 };\n    }\n\n    if (tok.type === 'open' && tok.name === 'leader') {\n      i++;\n      if (i < tokens.length && tokens[i]!.type === 'text') {\n        leader = tokens[i]!.text!.trim();\n        i++;\n      }\n      // consume </leader>\n      if (i < tokens.length && tokens[i]!.type === 'close') i++;\n      continue;\n    }\n\n    if (tok.type === 'self-close' && tok.name === 'controlfield') {\n      fields.push({ tag: tok.attrs?.['tag'] ?? '', data: '' });\n      i++;\n      continue;\n    }\n\n    if (tok.type === 'open' && tok.name === 'controlfield') {\n      const tag = tok.attrs?.['tag'] ?? '';\n      i++;\n      let data = '';\n      if (i < tokens.length && tokens[i]!.type === 'text') {\n        data = tokens[i]!.text ?? '';\n        i++;\n      }\n      // consume </controlfield>\n      if (i < tokens.length && tokens[i]!.type === 'close') i++;\n      fields.push({ tag, data });\n      continue;\n    }\n\n    if (tok.type === 'self-close' && tok.name === 'datafield') {\n      fields.push({\n        tag: tok.attrs?.['tag'] ?? '',\n        indicator1: tok.attrs?.['ind1'] ?? ' ',\n        indicator2: tok.attrs?.['ind2'] ?? ' ',\n        subfields: [],\n      });\n      i++;\n      continue;\n    }\n\n    if (tok.type === 'open' && tok.name === 'datafield') {\n      const tag = tok.attrs?.['tag'] ?? '';\n      const indicator1 = tok.attrs?.['ind1'] ?? ' ';\n      const indicator2 = tok.attrs?.['ind2'] ?? ' ';\n      const subfields: Subfield[] = [];\n      i++;\n\n      while (i < tokens.length) {\n        const stok = tokens[i]!;\n        if (stok.type === 'close' && stok.name === 'datafield') {\n          i++;\n          break;\n        }\n        if (stok.type === 'open' && stok.name === 'subfield') {\n          const code = stok.attrs?.['code'] ?? '';\n          i++;\n          let value = '';\n          if (i < tokens.length && tokens[i]!.type === 'text') {\n            value = tokens[i]!.text ?? '';\n            i++;\n          }\n          // consume </subfield>\n          if (i < tokens.length && tokens[i]!.type === 'close') i++;\n          subfields.push({ code, value });\n          continue;\n        }\n        i++;\n      }\n\n      fields.push({ tag, indicator1, indicator2, subfields });\n      continue;\n    }\n\n    i++;\n  }\n\n  return { record: { leader, fields }, end: i };\n}\n\n/**\n * Parse a MARCXML string containing one `<collection>` or one bare `<record>`.\n * Returns all records found.\n */\nexport function parseMarcXml(xml: string): MarcRecord[] {\n  const tokens = tokenise(xml);\n  const records: MarcRecord[] = [];\n  let i = 0;\n\n  while (i < tokens.length) {\n    const tok = tokens[i]!;\n    if (tok.type === 'open' && tok.name === 'record') {\n      const { record, end } = parseRecordTokens(tokens, i + 1);\n      records.push(record);\n      i = end;\n      continue;\n    }\n    i++;\n  }\n\n  return records;\n}\n\n// ─── MARCXML serializer ───────────────────────────────────────────────────────\n\nconst XML_HEADER = '<?xml version=\"1.0\" encoding=\"UTF-8\"?>\\n';\nconst COLLECTION_NS = 'xmlns=\"http://www.loc.gov/MARC21/slim\"';\nconst INDENT = '  ';\n\nfunction serializeMarcXmlRecord(record: MarcRecord): string {\n  const lines: string[] = [`<record ${COLLECTION_NS}>`];\n  lines.push(`${INDENT}<leader>${escapeXml(record.leader)}</leader>`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(`${INDENT}<controlfield tag=\"${field.tag}\">${escapeXml(field.data)}</controlfield>`);\n    } else {\n      const ind1 = field.indicator1 === ' ' ? ' ' : field.indicator1;\n      const ind2 = field.indicator2 === ' ' ? ' ' : field.indicator2;\n      lines.push(`${INDENT}<datafield tag=\"${field.tag}\" ind1=\"${ind1}\" ind2=\"${ind2}\">`);\n      for (const sf of field.subfields) {\n        lines.push(\n          `${INDENT}${INDENT}<subfield code=\"${sf.code}\">${escapeXml(sf.value)}</subfield>`\n        );\n      }\n      lines.push(`${INDENT}</datafield>`);\n    }\n  }\n\n  lines.push('</record>');\n  return lines.join('\\n');\n}\n\n/**\n * Serialize one or more MarcRecords into a MARCXML `<collection>` document.\n */\nexport function serializeMarcXml(records: MarcRecord[]): string {\n  const parts: string[] = [\n    XML_HEADER,\n    `<collection ${COLLECTION_NS}>`,\n  ];\n\n  for (const record of records) {\n    // Indent each record element by one level inside <collection>\n    const recordXml = serializeMarcXmlRecord(record)\n      .split('\\n')\n      .map((line) => INDENT + line)\n      .join('\\n');\n    parts.push(recordXml);\n  }\n\n  parts.push('</collection>');\n  return parts.join('\\n');\n}\n"],"mappings":"2GAgBA,IAAM,EAA0C,IAAI,IAAI,CACtD,CAAC,MAAO,GAAG,EACX,CAAC,KAAM,GAAG,EACV,CAAC,KAAM,GAAG,EACV,CAAC,OAAQ,GAAG,EACZ,CAAC,OAAQ,GAAG,CACd,CAAC,EAED,SAAS,EAAY,EAAsB,CACzC,OAAO,EAAK,QAAQ,gDAAA,CAAkD,EAAG,EAAK,EAAK,IAAS,CAC1F,GAAI,IAAQ,OAAW,CACrB,MAAM,EAAK,SAAS,EAAK,EAAE,EAC3B,OAAO,GAAM,GAAK,GAAM,QAAW,OAAO,cAAc,CAAE,EAAI,GAChE,CACA,GAAI,IAAQ,OAAW,CACrB,MAAM,EAAK,SAAS,EAAK,EAAE,EAC3B,OAAO,GAAM,GAAK,GAAM,QAAW,OAAO,cAAc,CAAE,EAAI,GAChE,CACA,OAAO,EAAW,IAAI,CAAI,GAAK,CACjC,CAAC,CACH,CAEA,SAAS,EAAU,EAAsB,CACvC,OAAO,EAIJ,QAAQ,gCAAiC,GAAG,EAC5C,QAAQ,KAAM,OAAO,EACrB,QAAQ,KAAM,MAAM,EACpB,QAAQ,KAAM,MAAM,EACpB,QAAQ,KAAM,QAAQ,EAGtB,QAAQ,MAAO,OAAO,CAC3B,CAeA,SAAS,EAAU,EAAqB,CACtC,MAAM,EAAQ,EAAI,QAAQ,GAAG,EAC7B,OAAO,IAAU,GAAK,EAAM,EAAI,MAAM,EAAQ,CAAC,CACjD,CAKA,SAAS,EAAW,EAAyC,CAC3D,MAAM,EAAgC,CAAC,EACjC,EAAK,qDACX,IAAI,EACJ,MAAQ,EAAI,EAAG,KAAK,CAAO,KAAO,MAAM,CACtC,MAAM,EAAM,EAAU,EAAE,CAAA,CAAG,EAC3B,EAAM,CAAA,EAAO,EAAY,EAAE,CAAA,GAAM,EAAE,CAAA,GAAM,EAAE,CAC7C,CACA,OAAO,CACT,CAOA,SAAS,EAAS,EAAsB,CACtC,MAAM,EAAkB,CAAC,EACzB,IAAI,EAAI,EAER,KAAO,EAAI,EAAI,QAAQ,CACrB,MAAM,EAAQ,EAAI,QAAQ,IAAK,CAAC,EAGhC,GAAI,IAAU,GAAI,CACH,EAAI,MAAM,CAAC,EAAE,KACtB,GAAM,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,EAAY,EAAI,MAAM,CAAC,CAAC,CAAE,CAAC,EACvE,KACF,CAEA,GAAI,EAAQ,EAAG,CACb,MAAM,EAAM,EAAI,MAAM,EAAG,CAAK,EACjB,EAAI,KACb,GAAM,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,EAAY,CAAG,CAAE,CAAC,CAChE,CAEA,MAAM,EAAQ,EAAI,QAAQ,IAAK,CAAK,EACpC,GAAI,IAAU,GAAI,MAElB,MAAM,EAAM,EAAI,MAAM,EAAQ,EAAG,CAAK,EAGtC,GAAI,EAAI,WAAW,GAAG,GAAK,EAAI,WAAW,GAAG,EAAG,CAC9C,EAAI,EAAQ,EACZ,QACF,CAEA,GAAI,EAAI,WAAW,GAAG,EACpB,EAAO,KAAK,CAAE,KAAM,QAAS,KAAM,EAAU,EAAI,MAAM,CAAC,EAAE,KAAK,CAAC,CAAE,CAAC,UAC1D,EAAI,SAAS,GAAG,EAAG,CAC5B,MAAM,EAAQ,EAAI,MAAM,EAAG,EAAE,EAAE,KAAK,EAC9B,EAAW,EAAM,OAAO,IAAI,EAC5B,EAAO,IAAa,GAAK,EAAQ,EAAM,MAAM,EAAG,CAAQ,EACxD,EAAU,IAAa,GAAK,GAAK,EAAM,MAAM,CAAQ,EAC3D,EAAO,KAAK,CAAE,KAAM,aAAc,KAAM,EAAU,CAAI,EAAG,MAAO,EAAW,CAAO,CAAE,CAAC,CACvF,KAAO,CACL,MAAM,EAAW,EAAI,OAAO,IAAI,EAC1B,EAAO,IAAa,GAAK,EAAM,EAAI,MAAM,EAAG,CAAQ,EACpD,EAAU,IAAa,GAAK,GAAK,EAAI,MAAM,CAAQ,EACzD,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,EAAU,CAAI,EAAG,MAAO,EAAW,CAAO,CAAE,CAAC,CACjF,CAEA,EAAI,EAAQ,CACd,CAEA,OAAO,CACT,CAQA,SAAS,EAAkB,EAAiB,EAAoD,CAC9F,IAAI,EAAS,GACb,MAAM,EAAuC,CAAC,EAC9C,IAAI,EAAI,EAER,KAAO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAM,EAAO,CAAA,EAEnB,GAAI,EAAI,OAAS,SAAW,EAAI,OAAS,SACvC,MAAO,CAAE,OAAQ,CAAE,OAAA,EAAQ,OAAA,CAAO,EAAG,IAAK,EAAI,CAAE,EAGlD,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,SAAU,CAChD,IACI,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAS,EAAO,CAAA,EAAI,KAAM,KAAK,EAC/B,KAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,QACF,CAEA,GAAI,EAAI,OAAS,cAAgB,EAAI,OAAS,eAAgB,CAC5D,EAAO,KAAK,CAAE,IAAK,EAAI,OAAQ,KAAU,GAAI,KAAM,EAAG,CAAC,EACvD,IACA,QACF,CAEA,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,eAAgB,CACtD,MAAM,EAAM,EAAI,OAAQ,KAAU,GAClC,IACA,IAAI,EAAO,GACP,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAO,EAAO,CAAA,EAAI,MAAQ,GAC1B,KAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,EAAO,KAAK,CAAE,IAAA,EAAK,KAAA,CAAK,CAAC,EACzB,QACF,CAEA,GAAI,EAAI,OAAS,cAAgB,EAAI,OAAS,YAAa,CACzD,EAAO,KAAK,CACV,IAAK,EAAI,OAAQ,KAAU,GAC3B,WAAY,EAAI,OAAQ,MAAW,IACnC,WAAY,EAAI,OAAQ,MAAW,IACnC,UAAW,CAAC,CACd,CAAC,EACD,IACA,QACF,CAEA,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,YAAa,CACnD,MAAM,EAAM,EAAI,OAAQ,KAAU,GAC5B,EAAa,EAAI,OAAQ,MAAW,IACpC,EAAa,EAAI,OAAQ,MAAW,IACpC,EAAwB,CAAC,EAG/B,IAFA,IAEO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAO,EAAO,CAAA,EACpB,GAAI,EAAK,OAAS,SAAW,EAAK,OAAS,YAAa,CACtD,IACA,KACF,CACA,GAAI,EAAK,OAAS,QAAU,EAAK,OAAS,WAAY,CACpD,MAAM,EAAO,EAAK,OAAQ,MAAW,GACrC,IACA,IAAI,EAAQ,GACR,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAQ,EAAO,CAAA,EAAI,MAAQ,GAC3B,KAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,EAAU,KAAK,CAAE,KAAA,EAAM,MAAA,CAAM,CAAC,EAC9B,QACF,CACA,GACF,CAEA,EAAO,KAAK,CAAE,IAAA,EAAK,WAAA,EAAY,WAAA,EAAY,UAAA,CAAU,CAAC,EACtD,QACF,CAEA,GACF,CAEA,MAAO,CAAE,OAAQ,CAAE,OAAA,EAAQ,OAAA,CAAO,EAAG,IAAK,CAAE,CAC9C,CAMA,SAAgB,EAAa,EAA2B,CACtD,MAAM,EAAS,EAAS,CAAG,EACrB,EAAwB,CAAC,EAC/B,IAAI,EAAI,EAER,KAAO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAM,EAAO,CAAA,EACnB,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,SAAU,CAChD,KAAM,CAAE,OAAA,EAAQ,IAAA,CAAA,EAAQ,EAAkB,EAAQ,EAAI,CAAC,EACvD,EAAQ,KAAK,CAAM,EACnB,EAAI,EACJ,QACF,CACA,GACF,CAEA,OAAO,CACT,CAIA,IAAM,EAAa;AAAA,EACb,EAAgB,yCAChB,EAAS,KAEf,SAAS,EAAuB,EAA4B,CAC1D,MAAM,EAAkB,CAAC,WAAW,CAAA,GAAgB,EACpD,EAAM,KAAK,GAAG,CAAA,WAAiB,EAAU,EAAO,MAAM,CAAA,WAAY,EAElE,UAAW,KAAS,EAAO,OACzB,GAAI,EAAA,eAAe,CAAK,EACtB,EAAM,KAAK,GAAG,CAAA,sBAA4B,EAAM,GAAA,KAAQ,EAAU,EAAM,IAAI,CAAA,iBAAkB,MACzF,CACL,MAAM,EAAO,EAAM,aAAe,IAAM,IAAM,EAAM,WAC9C,EAAO,EAAM,aAAe,IAAM,IAAM,EAAM,WACpD,EAAM,KAAK,GAAG,CAAA,mBAAyB,EAAM,GAAA,WAAc,CAAA,WAAe,CAAA,IAAQ,EAClF,UAAW,KAAM,EAAM,UACrB,EAAM,KACJ,GAAG,CAAA,GAAS,CAAA,mBAAyB,EAAG,IAAA,KAAS,EAAU,EAAG,KAAK,CAAA,aACrE,EAEF,EAAM,KAAK,GAAG,CAAA,cAAoB,CACpC,CAGF,OAAA,EAAM,KAAK,WAAW,EACf,EAAM,KAAK;AAAA,CAAI,CACxB,CAKA,SAAgB,EAAiB,EAA+B,CAC9D,MAAM,EAAkB,CACtB,EACA,eAAe,CAAA,GACjB,EAEA,UAAW,KAAU,EAAS,CAE5B,MAAM,EAAY,EAAuB,CAAM,EAC5C,MAAM;AAAA,CAAI,EACV,IAAK,GAAS,EAAS,CAAI,EAC3B,KAAK;AAAA,CAAI,EACZ,EAAM,KAAK,CAAS,CACtB,CAEA,OAAA,EAAM,KAAK,eAAe,EACnB,EAAM,KAAK;AAAA,CAAI,CACxB"}
+{"version":3,"file":"marcxml.cjs","names":[],"sources":["../src/marcxml.ts"],"sourcesContent":["/**\n * MARCXML parser and serializer.\n *\n * Supports the Library of Congress MARCXML schema:\n * http://www.loc.gov/MARC21/slim\n *\n * Parsing is done with a hand-rolled state machine — no XML library needed.\n * The MARCXML format is sufficiently regular (fixed element names, no arbitrary\n * nesting) that a full DOM parser is unnecessary.\n */\n\nimport type {\n  MarcRecord,\n  ControlField,\n  DataField,\n  Subfield,\n  ParseOptions,\n  ParseResult,\n  ParseBatchResult,\n  MarcWarning,\n} from './types';\nimport { isControlField } from './types';\nimport { createWarning } from './warnings';\n\n// ─── XML entity handling ─────────────────────────────────────────────────────\n\nconst ENTITY_MAP: ReadonlyMap<string, string> = new Map([\n  ['amp', '&'],\n  ['lt', '<'],\n  ['gt', '>'],\n  ['quot', '\"'],\n  ['apos', \"'\"],\n]);\n\nfunction unescapeXml(text: string): string {\n  return text.replace(/&(?:#x([0-9a-fA-F]+)|#([0-9]+)|([a-zA-Z]+));/g, (_, hex, dec, name) => {\n    if (hex !== undefined) {\n      const cp = parseInt(hex, 16);\n      return cp >= 0 && cp <= 0x10ffff ? String.fromCodePoint(cp) : '�';\n    }\n    if (dec !== undefined) {\n      const cp = parseInt(dec, 10);\n      return cp >= 0 && cp <= 0x10ffff ? String.fromCodePoint(cp) : '�';\n    }\n    return ENTITY_MAP.get(name) ?? _;\n  });\n}\n\nfunction escapeXml(text: string): string {\n  return (\n    text\n      // XML 1.0 forbids most C0 control characters in document text. There is no\n      // valid XML 1.0 representation for them, so substitute the Unicode\n      // replacement character to keep the output well-formed.\n      .replace(/[\\x00-\\x08\\x0B\\x0C\\x0E-\\x1F]/g, '�')\n      .replace(/&/g, '&amp;')\n      .replace(/</g, '&lt;')\n      .replace(/>/g, '&gt;')\n      .replace(/\"/g, '&quot;')\n      // Preserve literal CR through the XML round-trip: XML parsers normalize\n      // bare \\r and \\r\\n to \\n, so we must encode CR as a numeric reference.\n      .replace(/\\r/g, '&#13;')\n  );\n}\n\n// ─── Minimal tokeniser ────────────────────────────────────────────────────────\n\ninterface Token {\n  type: 'open' | 'close' | 'self-close' | 'text';\n  /** Local name (no namespace prefix) */\n  name?: string;\n  attrs?: Record<string, string>;\n  text?: string;\n}\n\n/**\n * Strip namespace prefix from a tag name, e.g. \"marc:record\" → \"record\".\n */\nfunction localName(raw: string): string {\n  const colon = raw.indexOf(':');\n  return colon === -1 ? raw : raw.slice(colon + 1);\n}\n\n/**\n * Parse `key=\"value\"` pairs out of an attribute string.\n */\nfunction parseAttrs(attrStr: string): Record<string, string> {\n  const attrs: Record<string, string> = {};\n  const re = /([a-zA-Z_:][^\\s=]*)\\s*=\\s*(?:\"([^\"]*)\"|'([^']*)')/g;\n  let m: RegExpExecArray | null;\n  while ((m = re.exec(attrStr)) !== null) {\n    const key = localName(m[1]!);\n    attrs[key] = unescapeXml(m[2] ?? m[3] ?? '');\n  }\n  return attrs;\n}\n\n/**\n * Find the closing `>` of an XML element tag, respecting quoted attribute values.\n */\nfunction findTagEnd(xml: string, start: number): number {\n  let inSingle = false;\n  let inDouble = false;\n  for (let j = start; j < xml.length; j++) {\n    const ch = xml[j];\n    if (ch === '\"' && !inSingle) inDouble = !inDouble;\n    else if (ch === \"'\" && !inDouble) inSingle = !inSingle;\n    else if (ch === '>' && !inSingle && !inDouble) return j;\n  }\n  return -1;\n}\n\n/**\n * Tokenise an XML string into a flat stream of open/close/text tokens.\n * Skips processing instructions, comments, and DOCTYPE declarations.\n * Handles CDATA sections as literal text.\n */\nfunction tokenise(xml: string, warnings?: MarcWarning[]): Token[] {\n  const tokens: Token[] = [];\n  let i = 0;\n\n  while (i < xml.length) {\n    const ltPos = xml.indexOf('<', i);\n\n    // Text node before next tag\n    if (ltPos === -1) {\n      const text = xml.slice(i).trim();\n      if (text) tokens.push({ type: 'text', text: unescapeXml(xml.slice(i)) });\n      break;\n    }\n\n    if (ltPos > i) {\n      const raw = xml.slice(i, ltPos);\n      const text = raw.trim();\n      if (text) tokens.push({ type: 'text', text: unescapeXml(raw) });\n    }\n\n    // Handle special constructs before finding the regular tag end\n    if (xml.startsWith('<!--', ltPos)) {\n      const commentEnd = xml.indexOf('-->', ltPos + 4);\n      if (commentEnd === -1) {\n        warnings?.push(createWarning('malformed_xml', 'Unterminated comment'));\n        i = xml.length;\n      } else {\n        i = commentEnd + 3;\n      }\n      continue;\n    }\n\n    if (xml.startsWith('<![CDATA[', ltPos)) {\n      const cdataEnd = xml.indexOf(']]>', ltPos + 9);\n      if (cdataEnd === -1) {\n        warnings?.push(createWarning('malformed_xml', 'Unterminated CDATA section'));\n        i = xml.length;\n      } else {\n        // CDATA content is literal — no entity unescaping\n        const cdataText = xml.slice(ltPos + 9, cdataEnd);\n        if (cdataText) tokens.push({ type: 'text', text: cdataText });\n        i = cdataEnd + 3;\n      }\n      continue;\n    }\n\n    if (xml.startsWith('<?', ltPos)) {\n      const piEnd = xml.indexOf('?>', ltPos + 2);\n      i = piEnd === -1 ? xml.length : piEnd + 2;\n      continue;\n    }\n\n    if (xml.startsWith('<!', ltPos)) {\n      const bangEnd = xml.indexOf('>', ltPos + 2);\n      i = bangEnd === -1 ? xml.length : bangEnd + 1;\n      continue;\n    }\n\n    // Regular element tag — use quote-aware scanning\n    const gtPos = findTagEnd(xml, ltPos + 1);\n    if (gtPos === -1) {\n      warnings?.push(createWarning('malformed_xml', 'Unclosed tag at end of input'));\n      break;\n    }\n\n    const tag = xml.slice(ltPos + 1, gtPos);\n\n    if (tag.startsWith('/')) {\n      tokens.push({ type: 'close', name: localName(tag.slice(1).trim()) });\n    } else if (tag.endsWith('/')) {\n      const inner = tag.slice(0, -1).trim();\n      const spaceIdx = inner.search(/\\s/);\n      const name = spaceIdx === -1 ? inner : inner.slice(0, spaceIdx);\n      const attrStr = spaceIdx === -1 ? '' : inner.slice(spaceIdx);\n      tokens.push({\n        type: 'self-close',\n        name: localName(name),\n        attrs: parseAttrs(attrStr),\n      });\n    } else {\n      const spaceIdx = tag.search(/\\s/);\n      const name = spaceIdx === -1 ? tag : tag.slice(0, spaceIdx);\n      const attrStr = spaceIdx === -1 ? '' : tag.slice(spaceIdx);\n      tokens.push({\n        type: 'open',\n        name: localName(name),\n        attrs: parseAttrs(attrStr),\n      });\n    }\n\n    i = gtPos + 1;\n  }\n\n  return tokens;\n}\n\n// ─── MARCXML parser ───────────────────────────────────────────────────────────\n\nfunction emitWarning(warnings: MarcWarning[], warning: MarcWarning, options: ParseOptions): void {\n  if (options.strict) {\n    throw new Error(warning.message);\n  }\n  const max = options.maxWarnings ?? 100;\n  if (warnings.length < max) {\n    warnings.push(warning);\n  }\n}\n\n/**\n * Parse one `<record>` element's worth of tokens into a MarcRecord.\n */\nfunction parseRecordTokens(\n  tokens: Token[],\n  start: number,\n  warnings: MarcWarning[],\n  options: ParseOptions\n): { record: MarcRecord; end: number } {\n  let leader = '';\n  let hasLeader = false;\n  const fields: (ControlField | DataField)[] = [];\n  let i = start;\n\n  while (i < tokens.length) {\n    const tok = tokens[i]!;\n\n    if (tok.type === 'close' && tok.name === 'record') {\n      if (!hasLeader) {\n        emitWarning(\n          warnings,\n          createWarning('missing_element', 'Record has no <leader> element'),\n          options\n        );\n      }\n      return { record: { leader, fields }, end: i + 1 };\n    }\n\n    if (tok.type === 'open' && tok.name === 'leader') {\n      hasLeader = true;\n      i++;\n      if (i < tokens.length && tokens[i]!.type === 'text') {\n        leader = tokens[i]!.text!.trim();\n        i++;\n      }\n      if (leader.length !== 24) {\n        emitWarning(\n          warnings,\n          createWarning('invalid_leader', `Leader is ${leader.length} characters, expected 24`),\n          options\n        );\n      }\n      // consume </leader>\n      if (i < tokens.length && tokens[i]!.type === 'close') i++;\n      continue;\n    }\n\n    if (tok.type === 'self-close' && tok.name === 'controlfield') {\n      const tag = tok.attrs?.['tag'];\n      if (tag === undefined) {\n        emitWarning(\n          warnings,\n          createWarning('missing_element', 'controlfield missing tag attribute'),\n          options\n        );\n      }\n      fields.push({ tag: tag ?? '', data: '' });\n      i++;\n      continue;\n    }\n\n    if (tok.type === 'open' && tok.name === 'controlfield') {\n      const tag = tok.attrs?.['tag'];\n      if (tag === undefined) {\n        emitWarning(\n          warnings,\n          createWarning('missing_element', 'controlfield missing tag attribute'),\n          options\n        );\n      }\n      i++;\n      let data = '';\n      if (i < tokens.length && tokens[i]!.type === 'text') {\n        data = tokens[i]!.text ?? '';\n        i++;\n      }\n      // consume </controlfield>\n      if (i < tokens.length && tokens[i]!.type === 'close') i++;\n      fields.push({ tag: tag ?? '', data });\n      continue;\n    }\n\n    if (tok.type === 'self-close' && tok.name === 'datafield') {\n      const tag = tok.attrs?.['tag'];\n      if (tag === undefined) {\n        emitWarning(\n          warnings,\n          createWarning('missing_element', 'datafield missing tag attribute'),\n          options\n        );\n      }\n      fields.push({\n        tag: tag ?? '',\n        indicator1: tok.attrs?.['ind1'] ?? ' ',\n        indicator2: tok.attrs?.['ind2'] ?? ' ',\n        subfields: [],\n      });\n      i++;\n      continue;\n    }\n\n    if (tok.type === 'open' && tok.name === 'datafield') {\n      const tag = tok.attrs?.['tag'];\n      if (tag === undefined) {\n        emitWarning(\n          warnings,\n          createWarning('missing_element', 'datafield missing tag attribute'),\n          options\n        );\n      }\n      const indicator1 = tok.attrs?.['ind1'] ?? ' ';\n      const indicator2 = tok.attrs?.['ind2'] ?? ' ';\n      const subfields: Subfield[] = [];\n      i++;\n\n      while (i < tokens.length) {\n        const stok = tokens[i]!;\n        if (stok.type === 'close' && stok.name === 'datafield') {\n          i++;\n          break;\n        }\n        if (stok.type === 'open' && stok.name === 'subfield') {\n          const code = stok.attrs?.['code'];\n          if (code === undefined) {\n            emitWarning(\n              warnings,\n              createWarning('missing_element', 'subfield missing code attribute', undefined, tag),\n              options\n            );\n          }\n          i++;\n          let value = '';\n          if (i < tokens.length && tokens[i]!.type === 'text') {\n            value = tokens[i]!.text ?? '';\n            i++;\n          }\n          // consume </subfield>\n          if (i < tokens.length && tokens[i]!.type === 'close') i++;\n          subfields.push({ code: code ?? '', value });\n          continue;\n        }\n        i++;\n      }\n\n      fields.push({ tag: tag ?? '', indicator1, indicator2, subfields });\n      continue;\n    }\n\n    i++;\n  }\n\n  if (!hasLeader) {\n    emitWarning(\n      warnings,\n      createWarning('missing_element', 'Record has no <leader> element'),\n      options\n    );\n  }\n  return { record: { leader, fields }, end: i };\n}\n\n/**\n * Parse a MARCXML string, returning per-record parse results including warnings.\n * Unlike parseMarcXml, every record attempt is included even if it produced warnings.\n */\nexport function parseMarcXmlWithWarnings(xml: string, options?: ParseOptions): ParseBatchResult {\n  const opts = options ?? {};\n  const tokWarnings: MarcWarning[] = [];\n  const tokens = tokenise(xml, tokWarnings);\n  const results: ParseResult[] = [];\n  let i = 0;\n\n  while (i < tokens.length) {\n    const tok = tokens[i]!;\n    if (tok.type === 'open' && tok.name === 'record') {\n      const recordWarnings: MarcWarning[] = [];\n      const { record, end } = parseRecordTokens(tokens, i + 1, recordWarnings, opts);\n      results.push({ record, warnings: recordWarnings });\n      i = end;\n      continue;\n    }\n    i++;\n  }\n\n  if (results.length === 0 && tokWarnings.length > 0) {\n    results.push({ record: null, warnings: tokWarnings });\n  } else if (tokWarnings.length > 0 && results.length > 0) {\n    results[0] = {\n      record: results[0]!.record,\n      warnings: [...tokWarnings, ...results[0]!.warnings],\n    };\n  }\n\n  return { results };\n}\n\n/**\n * Parse a MARCXML string containing one `<collection>` or one bare `<record>`.\n * Returns all successfully parsed records.\n */\nexport function parseMarcXml(xml: string, options?: ParseOptions): MarcRecord[] {\n  const batch = parseMarcXmlWithWarnings(xml, options);\n  return batch.results.map((r) => r.record).filter((r): r is MarcRecord => r !== null);\n}\n\n// ─── MARCXML serializer ───────────────────────────────────────────────────────\n\nconst XML_HEADER = '<?xml version=\"1.0\" encoding=\"UTF-8\"?>\\n';\nconst COLLECTION_NS = 'xmlns=\"http://www.loc.gov/MARC21/slim\"';\nconst INDENT = '  ';\n\nfunction serializeMarcXmlRecord(record: MarcRecord): string {\n  const lines: string[] = [`<record ${COLLECTION_NS}>`];\n  lines.push(`${INDENT}<leader>${escapeXml(record.leader)}</leader>`);\n\n  for (const field of record.fields) {\n    if (isControlField(field)) {\n      lines.push(\n        `${INDENT}<controlfield tag=\"${escapeXml(field.tag)}\">${escapeXml(field.data)}</controlfield>`\n      );\n    } else {\n      const ind1 = field.indicator1 === ' ' ? ' ' : field.indicator1;\n      const ind2 = field.indicator2 === ' ' ? ' ' : field.indicator2;\n      lines.push(\n        `${INDENT}<datafield tag=\"${escapeXml(field.tag)}\" ind1=\"${escapeXml(ind1)}\" ind2=\"${escapeXml(ind2)}\">`\n      );\n      for (const sf of field.subfields) {\n        lines.push(\n          `${INDENT}${INDENT}<subfield code=\"${escapeXml(sf.code)}\">${escapeXml(sf.value)}</subfield>`\n        );\n      }\n      lines.push(`${INDENT}</datafield>`);\n    }\n  }\n\n  lines.push('</record>');\n  return lines.join('\\n');\n}\n\n/**\n * Serialize one or more MarcRecords into a MARCXML `<collection>` document.\n */\nexport function serializeMarcXml(records: MarcRecord[]): string {\n  const parts: string[] = [XML_HEADER, `<collection ${COLLECTION_NS}>`];\n\n  for (const record of records) {\n    // Indent each record element by one level inside <collection>\n    const recordXml = serializeMarcXmlRecord(record)\n      .split('\\n')\n      .map((line) => INDENT + line)\n      .join('\\n');\n    parts.push(recordXml);\n  }\n\n  parts.push('</collection>');\n  return parts.join('\\n');\n}\n"],"mappings":"gJA0BA,IAAM,EAA0C,IAAI,IAAI,CACtD,CAAC,MAAO,GAAG,EACX,CAAC,KAAM,GAAG,EACV,CAAC,KAAM,GAAG,EACV,CAAC,OAAQ,GAAG,EACZ,CAAC,OAAQ,GAAG,CACd,CAAC,EAED,SAAS,EAAY,EAAsB,CACzC,OAAO,EAAK,QAAQ,gDAAA,CAAkD,EAAG,EAAK,EAAK,IAAS,CAC1F,GAAI,IAAQ,OAAW,CACrB,MAAM,EAAK,SAAS,EAAK,EAAE,EAC3B,OAAO,GAAM,GAAK,GAAM,QAAW,OAAO,cAAc,CAAE,EAAI,GAChE,CACA,GAAI,IAAQ,OAAW,CACrB,MAAM,EAAK,SAAS,EAAK,EAAE,EAC3B,OAAO,GAAM,GAAK,GAAM,QAAW,OAAO,cAAc,CAAE,EAAI,GAChE,CACA,OAAO,EAAW,IAAI,CAAI,GAAK,CACjC,CAAC,CACH,CAEA,SAAS,EAAU,EAAsB,CACvC,OACE,EAIG,QAAQ,gCAAiC,GAAG,EAC5C,QAAQ,KAAM,OAAO,EACrB,QAAQ,KAAM,MAAM,EACpB,QAAQ,KAAM,MAAM,EACpB,QAAQ,KAAM,QAAQ,EAGtB,QAAQ,MAAO,OAAO,CAE7B,CAeA,SAAS,EAAU,EAAqB,CACtC,MAAM,EAAQ,EAAI,QAAQ,GAAG,EAC7B,OAAO,IAAU,GAAK,EAAM,EAAI,MAAM,EAAQ,CAAC,CACjD,CAKA,SAAS,EAAW,EAAyC,CAC3D,MAAM,EAAgC,CAAC,EACjC,EAAK,qDACX,IAAI,EACJ,MAAQ,EAAI,EAAG,KAAK,CAAO,KAAO,MAAM,CACtC,MAAM,EAAM,EAAU,EAAE,CAAA,CAAG,EAC3B,EAAM,CAAA,EAAO,EAAY,EAAE,CAAA,GAAM,EAAE,CAAA,GAAM,EAAE,CAC7C,CACA,OAAO,CACT,CAKA,SAAS,EAAW,EAAa,EAAuB,CACtD,IAAI,EAAW,GACX,EAAW,GACf,QAAS,EAAI,EAAO,EAAI,EAAI,OAAQ,IAAK,CACvC,MAAM,EAAK,EAAI,CAAA,EACf,GAAI,IAAO,KAAO,CAAC,EAAU,EAAW,CAAC,UAChC,IAAO,KAAO,CAAC,EAAU,EAAW,CAAC,UACrC,IAAO,KAAO,CAAC,GAAY,CAAC,EAAU,OAAO,CACxD,CACA,MAAO,EACT,CAOA,SAAS,EAAS,EAAa,EAAmC,CAChE,MAAM,EAAkB,CAAC,EACzB,IAAI,EAAI,EAER,KAAO,EAAI,EAAI,QAAQ,CACrB,MAAM,EAAQ,EAAI,QAAQ,IAAK,CAAC,EAGhC,GAAI,IAAU,GAAI,CACH,EAAI,MAAM,CAAC,EAAE,KACtB,GAAM,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,EAAY,EAAI,MAAM,CAAC,CAAC,CAAE,CAAC,EACvE,KACF,CAEA,GAAI,EAAQ,EAAG,CACb,MAAM,EAAM,EAAI,MAAM,EAAG,CAAK,EACjB,EAAI,KACb,GAAM,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,EAAY,CAAG,CAAE,CAAC,CAChE,CAGA,GAAI,EAAI,WAAW,OAAQ,CAAK,EAAG,CACjC,MAAM,EAAa,EAAI,QAAQ,MAAO,EAAQ,CAAC,EAC3C,IAAe,IACjB,GAAU,KAAK,EAAA,cAAc,gBAAiB,sBAAsB,CAAC,EACrE,EAAI,EAAI,QAER,EAAI,EAAa,EAEnB,QACF,CAEA,GAAI,EAAI,WAAW,YAAa,CAAK,EAAG,CACtC,MAAM,EAAW,EAAI,QAAQ,MAAO,EAAQ,CAAC,EAC7C,GAAI,IAAa,GACf,GAAU,KAAK,EAAA,cAAc,gBAAiB,4BAA4B,CAAC,EAC3E,EAAI,EAAI,WACH,CAEL,MAAM,EAAY,EAAI,MAAM,EAAQ,EAAG,CAAQ,EAC3C,GAAW,EAAO,KAAK,CAAE,KAAM,OAAQ,KAAM,CAAU,CAAC,EAC5D,EAAI,EAAW,CACjB,CACA,QACF,CAEA,GAAI,EAAI,WAAW,KAAM,CAAK,EAAG,CAC/B,MAAM,EAAQ,EAAI,QAAQ,KAAM,EAAQ,CAAC,EACzC,EAAI,IAAU,GAAK,EAAI,OAAS,EAAQ,EACxC,QACF,CAEA,GAAI,EAAI,WAAW,KAAM,CAAK,EAAG,CAC/B,MAAM,EAAU,EAAI,QAAQ,IAAK,EAAQ,CAAC,EAC1C,EAAI,IAAY,GAAK,EAAI,OAAS,EAAU,EAC5C,QACF,CAGA,MAAM,EAAQ,EAAW,EAAK,EAAQ,CAAC,EACvC,GAAI,IAAU,GAAI,CAChB,GAAU,KAAK,EAAA,cAAc,gBAAiB,8BAA8B,CAAC,EAC7E,KACF,CAEA,MAAM,EAAM,EAAI,MAAM,EAAQ,EAAG,CAAK,EAEtC,GAAI,EAAI,WAAW,GAAG,EACpB,EAAO,KAAK,CAAE,KAAM,QAAS,KAAM,EAAU,EAAI,MAAM,CAAC,EAAE,KAAK,CAAC,CAAE,CAAC,UAC1D,EAAI,SAAS,GAAG,EAAG,CAC5B,MAAM,EAAQ,EAAI,MAAM,EAAG,EAAE,EAAE,KAAK,EAC9B,EAAW,EAAM,OAAO,IAAI,EAC5B,EAAO,IAAa,GAAK,EAAQ,EAAM,MAAM,EAAG,CAAQ,EACxD,EAAU,IAAa,GAAK,GAAK,EAAM,MAAM,CAAQ,EAC3D,EAAO,KAAK,CACV,KAAM,aACN,KAAM,EAAU,CAAI,EACpB,MAAO,EAAW,CAAO,CAC3B,CAAC,CACH,KAAO,CACL,MAAM,EAAW,EAAI,OAAO,IAAI,EAC1B,EAAO,IAAa,GAAK,EAAM,EAAI,MAAM,EAAG,CAAQ,EACpD,EAAU,IAAa,GAAK,GAAK,EAAI,MAAM,CAAQ,EACzD,EAAO,KAAK,CACV,KAAM,OACN,KAAM,EAAU,CAAI,EACpB,MAAO,EAAW,CAAO,CAC3B,CAAC,CACH,CAEA,EAAI,EAAQ,CACd,CAEA,OAAO,CACT,CAIA,SAAS,EAAY,EAAyB,EAAsB,EAA6B,CAC/F,GAAI,EAAQ,OACV,MAAM,IAAI,MAAM,EAAQ,OAAO,EAEjC,MAAM,EAAM,EAAQ,aAAe,IAC/B,EAAS,OAAS,GACpB,EAAS,KAAK,CAAO,CAEzB,CAKA,SAAS,EACP,EACA,EACA,EACA,EACqC,CACrC,IAAI,EAAS,GACT,EAAY,GAChB,MAAM,EAAuC,CAAC,EAC9C,IAAI,EAAI,EAER,KAAO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAM,EAAO,CAAA,EAEnB,GAAI,EAAI,OAAS,SAAW,EAAI,OAAS,SACvC,OAAK,GACH,EACE,EACA,EAAA,cAAc,kBAAmB,gCAAgC,EACjE,CACF,EAEK,CAAE,OAAQ,CAAE,OAAA,EAAQ,OAAA,CAAO,EAAG,IAAK,EAAI,CAAE,EAGlD,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,SAAU,CAChD,EAAY,GACZ,IACI,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAS,EAAO,CAAA,EAAI,KAAM,KAAK,EAC/B,KAEE,EAAO,SAAW,IACpB,EACE,EACA,EAAA,cAAc,iBAAkB,aAAa,EAAO,MAAA,0BAAgC,EACpF,CACF,EAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,QACF,CAEA,GAAI,EAAI,OAAS,cAAgB,EAAI,OAAS,eAAgB,CAC5D,MAAM,EAAM,EAAI,OAAQ,IACpB,IAAQ,QACV,EACE,EACA,EAAA,cAAc,kBAAmB,oCAAoC,EACrE,CACF,EAEF,EAAO,KAAK,CAAE,IAAK,GAAO,GAAI,KAAM,EAAG,CAAC,EACxC,IACA,QACF,CAEA,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,eAAgB,CACtD,MAAM,EAAM,EAAI,OAAQ,IACpB,IAAQ,QACV,EACE,EACA,EAAA,cAAc,kBAAmB,oCAAoC,EACrE,CACF,EAEF,IACA,IAAI,EAAO,GACP,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAO,EAAO,CAAA,EAAI,MAAQ,GAC1B,KAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,EAAO,KAAK,CAAE,IAAK,GAAO,GAAI,KAAA,CAAK,CAAC,EACpC,QACF,CAEA,GAAI,EAAI,OAAS,cAAgB,EAAI,OAAS,YAAa,CACzD,MAAM,EAAM,EAAI,OAAQ,IACpB,IAAQ,QACV,EACE,EACA,EAAA,cAAc,kBAAmB,iCAAiC,EAClE,CACF,EAEF,EAAO,KAAK,CACV,IAAK,GAAO,GACZ,WAAY,EAAI,OAAQ,MAAW,IACnC,WAAY,EAAI,OAAQ,MAAW,IACnC,UAAW,CAAC,CACd,CAAC,EACD,IACA,QACF,CAEA,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,YAAa,CACnD,MAAM,EAAM,EAAI,OAAQ,IACpB,IAAQ,QACV,EACE,EACA,EAAA,cAAc,kBAAmB,iCAAiC,EAClE,CACF,EAEF,MAAM,EAAa,EAAI,OAAQ,MAAW,IACpC,EAAa,EAAI,OAAQ,MAAW,IACpC,EAAwB,CAAC,EAG/B,IAFA,IAEO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAO,EAAO,CAAA,EACpB,GAAI,EAAK,OAAS,SAAW,EAAK,OAAS,YAAa,CACtD,IACA,KACF,CACA,GAAI,EAAK,OAAS,QAAU,EAAK,OAAS,WAAY,CACpD,MAAM,EAAO,EAAK,OAAQ,KACtB,IAAS,QACX,EACE,EACA,EAAA,cAAc,kBAAmB,kCAAmC,OAAW,CAAG,EAClF,CACF,EAEF,IACA,IAAI,EAAQ,GACR,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAC3C,EAAQ,EAAO,CAAA,EAAI,MAAQ,GAC3B,KAGE,EAAI,EAAO,QAAU,EAAO,CAAA,EAAI,OAAS,SAAS,IACtD,EAAU,KAAK,CAAE,KAAM,GAAQ,GAAI,MAAA,CAAM,CAAC,EAC1C,QACF,CACA,GACF,CAEA,EAAO,KAAK,CAAE,IAAK,GAAO,GAAI,WAAA,EAAY,WAAA,EAAY,UAAA,CAAU,CAAC,EACjE,QACF,CAEA,GACF,CAEA,OAAK,GACH,EACE,EACA,EAAA,cAAc,kBAAmB,gCAAgC,EACjE,CACF,EAEK,CAAE,OAAQ,CAAE,OAAA,EAAQ,OAAA,CAAO,EAAG,IAAK,CAAE,CAC9C,CAMA,SAAgB,EAAyB,EAAa,EAA0C,CAC9F,MAAM,EAAO,GAAW,CAAC,EACnB,EAA6B,CAAC,EAC9B,EAAS,EAAS,EAAK,CAAW,EAClC,EAAyB,CAAC,EAChC,IAAI,EAAI,EAER,KAAO,EAAI,EAAO,QAAQ,CACxB,MAAM,EAAM,EAAO,CAAA,EACnB,GAAI,EAAI,OAAS,QAAU,EAAI,OAAS,SAAU,CAChD,MAAM,EAAgC,CAAC,EACjC,CAAE,OAAA,EAAQ,IAAA,CAAA,EAAQ,EAAkB,EAAQ,EAAI,EAAG,EAAgB,CAAI,EAC7E,EAAQ,KAAK,CAAE,OAAA,EAAQ,SAAU,CAAe,CAAC,EACjD,EAAI,EACJ,QACF,CACA,GACF,CAEA,OAAI,EAAQ,SAAW,GAAK,EAAY,OAAS,EAC/C,EAAQ,KAAK,CAAE,OAAQ,KAAM,SAAU,CAAY,CAAC,EAC3C,EAAY,OAAS,GAAK,EAAQ,OAAS,IACpD,EAAQ,CAAA,EAAK,CACX,OAAQ,EAAQ,CAAA,EAAI,OACpB,SAAU,CAAC,GAAG,EAAa,GAAG,EAAQ,CAAA,EAAI,QAAQ,CACpD,GAGK,CAAE,QAAA,CAAQ,CACnB,CAMA,SAAgB,EAAa,EAAa,EAAsC,CAE9E,OADc,EAAyB,EAAK,CACrC,EAAM,QAAQ,IAAK,GAAM,EAAE,MAAM,EAAE,OAAQ,GAAuB,IAAM,IAAI,CACrF,CAIA,IAAM,EAAa;AAAA,EACb,EAAgB,yCAChB,EAAS,KAEf,SAAS,EAAuB,EAA4B,CAC1D,MAAM,EAAkB,CAAC,WAAW,CAAA,GAAgB,EACpD,EAAM,KAAK,GAAG,CAAA,WAAiB,EAAU,EAAO,MAAM,CAAA,WAAY,EAElE,UAAW,KAAS,EAAO,OACzB,GAAI,EAAA,eAAe,CAAK,EACtB,EAAM,KACJ,GAAG,CAAA,sBAA4B,EAAU,EAAM,GAAG,CAAA,KAAM,EAAU,EAAM,IAAI,CAAA,iBAC9E,MACK,CACL,MAAM,EAAO,EAAM,aAAe,IAAM,IAAM,EAAM,WAC9C,EAAO,EAAM,aAAe,IAAM,IAAM,EAAM,WACpD,EAAM,KACJ,GAAG,CAAA,mBAAyB,EAAU,EAAM,GAAG,CAAA,WAAY,EAAU,CAAI,CAAA,WAAY,EAAU,CAAI,CAAA,IACrG,EACA,UAAW,KAAM,EAAM,UACrB,EAAM,KACJ,GAAG,CAAA,GAAS,CAAA,mBAAyB,EAAU,EAAG,IAAI,CAAA,KAAM,EAAU,EAAG,KAAK,CAAA,aAChF,EAEF,EAAM,KAAK,GAAG,CAAA,cAAoB,CACpC,CAGF,OAAA,EAAM,KAAK,WAAW,EACf,EAAM,KAAK;AAAA,CAAI,CACxB,CAKA,SAAgB,EAAiB,EAA+B,CAC9D,MAAM,EAAkB,CAAC,EAAY,eAAe,CAAA,GAAgB,EAEpE,UAAW,KAAU,EAAS,CAE5B,MAAM,EAAY,EAAuB,CAAM,EAC5C,MAAM;AAAA,CAAI,EACV,IAAK,GAAS,EAAS,CAAI,EAC3B,KAAK;AAAA,CAAI,EACZ,EAAM,KAAK,CAAS,CACtB,CAEA,OAAA,EAAM,KAAK,eAAe,EACnB,EAAM,KAAK;AAAA,CAAI,CACxB"}

package/dist/marcxml.d.ts

@@ -1,9 +1,14 @@
-import { MarcRecord } from './types';
+import { MarcRecord, ParseOptions, ParseBatchResult } from './types';
+/**
+ * Parse a MARCXML string, returning per-record parse results including warnings.
+ * Unlike parseMarcXml, every record attempt is included even if it produced warnings.
+ */
+export declare function parseMarcXmlWithWarnings(xml: string, options?: ParseOptions): ParseBatchResult;
 /**
  * Parse a MARCXML string containing one `<collection>` or one bare `<record>`.
- * Returns all records found.
+ * Returns all successfully parsed records.
  */
-export declare function parseMarcXml(xml: string): MarcRecord[];
+export declare function parseMarcXml(xml: string, options?: ParseOptions): MarcRecord[];
 /**
  * Serialize one or more MarcRecords into a MARCXML `<collection>` document.
  */