read-excel-file 9.2.0 → 9.3.1

package/modules/xlsx/parseSheetData.js

@@ -0,0 +1,55 @@
+import parseCell, { createInitialStateInCell, onOpenTagInCell, onCloseTagInCell, onTextInCell } from './parseCell.js';
+export default function parseSheetData(state, sharedStrings, styles, epoch1904, options) {
+  return state.cells.map(function (_ref) {
+    var attributes = _ref.attributes,
+      value = _ref.value,
+      inlineString = _ref.inlineString;
+    return parseCell(attributes, value, inlineString, sharedStrings, styles, epoch1904, options);
+  });
+}
+export function createInitialStateInSheetData() {
+  return {
+    c: undefined,
+    cells: []
+  };
+}
+export function onOpenTagInSheetData(tagName, attributes, state) {
+  if (tagName === 'c') {
+    state.c = createInitialStateInCell();
+    state.c.attributes = attributes;
+  } else if (state.c) {
+    onOpenTagInCell(tagName, attributes, state.c);
+  }
+}
+export function onCloseTagInSheetData(tagName, state) {
+  if (tagName === 'c') {
+    state.cells.push({
+      attributes: state.c.attributes,
+      value: state.c.value,
+      inlineString: state.c.inlineString
+    });
+    state.c = undefined;
+  } else if (state.c) {
+    onCloseTagInCell(tagName, state.c);
+  }
+}
+export function onTextInSheetData(text, state) {
+  if (state.c) {
+    onTextInCell(text, state.c);
+  }
+}
+
+// Here, it could also parse "merged cells" and then return them in some special way.
+// But then it's not clear what should be the way to return such merged cells.
+// I.e. should it just return it as a duplicate value in each one of the merged cells?
+// Or should it keep the current behavior of only returning the value of the top-most left-most cell
+// and then just return `null` for the rest of the cells in a "merged cells" group?
+// Perhaps the latter (current) approach is the most sensible one, so there's no need
+// to change anything.
+//
+// const mergedCells = getMergedCellCoordinates(sheetDocument)
+// for (const mergedCell of mergedCells) {
+//   const [from, to] = mergedCell.split(':').map(parseCellAddress)
+//   console.log('Merged Cell.', 'From:', from, 'To:', to)
+// }
+//# sourceMappingURL=parseSheetData.js.map

package/modules/xlsx/parseSheetData.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"parseSheetData.js","names":["parseCell","createInitialStateInCell","onOpenTagInCell","onCloseTagInCell","onTextInCell","parseSheetData","state","sharedStrings","styles","epoch1904","options","cells","map","_ref","attributes","value","inlineString","createInitialStateInSheetData","c","undefined","onOpenTagInSheetData","tagName","onCloseTagInSheetData","push","onTextInSheetData","text"],"sources":["../../source/xlsx/parseSheetData.js"],"sourcesContent":["import parseCell, {\r\n  createInitialStateInCell,\r\n  onOpenTagInCell,\r\n  onCloseTagInCell,\r\n  onTextInCell\r\n} from './parseCell.js'\r\n\r\nexport default function parseSheetData(state, sharedStrings, styles, epoch1904, options) {\r\n  return state.cells.map(({ attributes, value, inlineString }) => {\r\n    return parseCell(\r\n      attributes,\r\n      value,\r\n      inlineString,\r\n      sharedStrings,\r\n      styles,\r\n      epoch1904,\r\n      options\r\n    )\r\n  })\r\n}\r\n\r\nexport function createInitialStateInSheetData() {\r\n  return {\r\n    c: undefined,\r\n    cells: []\r\n  }\r\n}\r\n\r\nexport function onOpenTagInSheetData(tagName, attributes, state) {\r\n  if (tagName === 'c') {\r\n    state.c = createInitialStateInCell()\r\n    state.c.attributes = attributes\r\n  } else if (state.c) {\r\n    onOpenTagInCell(tagName, attributes, state.c)\r\n  }\r\n}\r\n\r\nexport function onCloseTagInSheetData(tagName, state) {\r\n  if (tagName === 'c') {\r\n    state.cells.push({\r\n      attributes: state.c.attributes,\r\n      value: state.c.value,\r\n      inlineString: state.c.inlineString\r\n    })\r\n    state.c = undefined\r\n  } else if (state.c) {\r\n    onCloseTagInCell(tagName, state.c)\r\n  }\r\n}\r\n\r\nexport function onTextInSheetData(text, state) {\r\n  if (state.c) {\r\n    onTextInCell(text, state.c)\r\n  }\r\n}\r\n\r\n// Here, it could also parse \"merged cells\" and then return them in some special way.\r\n// But then it's not clear what should be the way to return such merged cells.\r\n// I.e. should it just return it as a duplicate value in each one of the merged cells?\r\n// Or should it keep the current behavior of only returning the value of the top-most left-most cell\r\n// and then just return `null` for the rest of the cells in a \"merged cells\" group?\r\n// Perhaps the latter (current) approach is the most sensible one, so there's no need\r\n// to change anything.\r\n//\r\n// const mergedCells = getMergedCellCoordinates(sheetDocument)\r\n// for (const mergedCell of mergedCells) {\r\n//   const [from, to] = mergedCell.split(':').map(parseCellAddress)\r\n//   console.log('Merged Cell.', 'From:', from, 'To:', to)\r\n// }"],"mappings":"AAAA,OAAOA,SAAS,IACdC,wBAAwB,EACxBC,eAAe,EACfC,gBAAgB,EAChBC,YAAY,QACP,gBAAgB;AAEvB,eAAe,SAASC,cAAcA,CAACC,KAAK,EAAEC,aAAa,EAAEC,MAAM,EAAEC,SAAS,EAAEC,OAAO,EAAE;EACvF,OAAOJ,KAAK,CAACK,KAAK,CAACC,GAAG,CAAC,UAAAC,IAAA,EAAyC;IAAA,IAAtCC,UAAU,GAAAD,IAAA,CAAVC,UAAU;MAAEC,KAAK,GAAAF,IAAA,CAALE,KAAK;MAAEC,YAAY,GAAAH,IAAA,CAAZG,YAAY;IACvD,OAAOhB,SAAS,CACdc,UAAU,EACVC,KAAK,EACLC,YAAY,EACZT,aAAa,EACbC,MAAM,EACNC,SAAS,EACTC,OACF,CAAC;EACH,CAAC,CAAC;AACJ;AAEA,OAAO,SAASO,6BAA6BA,CAAA,EAAG;EAC9C,OAAO;IACLC,CAAC,EAAEC,SAAS;IACZR,KAAK,EAAE;EACT,CAAC;AACH;AAEA,OAAO,SAASS,oBAAoBA,CAACC,OAAO,EAAEP,UAAU,EAAER,KAAK,EAAE;EAC/D,IAAIe,OAAO,KAAK,GAAG,EAAE;IACnBf,KAAK,CAACY,CAAC,GAAGjB,wBAAwB,CAAC,CAAC;IACpCK,KAAK,CAACY,CAAC,CAACJ,UAAU,GAAGA,UAAU;EACjC,CAAC,MAAM,IAAIR,KAAK,CAACY,CAAC,EAAE;IAClBhB,eAAe,CAACmB,OAAO,EAAEP,UAAU,EAAER,KAAK,CAACY,CAAC,CAAC;EAC/C;AACF;AAEA,OAAO,SAASI,qBAAqBA,CAACD,OAAO,EAAEf,KAAK,EAAE;EACpD,IAAIe,OAAO,KAAK,GAAG,EAAE;IACnBf,KAAK,CAACK,KAAK,CAACY,IAAI,CAAC;MACfT,UAAU,EAAER,KAAK,CAACY,CAAC,CAACJ,UAAU;MAC9BC,KAAK,EAAET,KAAK,CAACY,CAAC,CAACH,KAAK;MACpBC,YAAY,EAAEV,KAAK,CAACY,CAAC,CAACF;IACxB,CAAC,CAAC;IACFV,KAAK,CAACY,CAAC,GAAGC,SAAS;EACrB,CAAC,MAAM,IAAIb,KAAK,CAACY,CAAC,EAAE;IAClBf,gBAAgB,CAACkB,OAAO,EAAEf,KAAK,CAACY,CAAC,CAAC;EACpC;AACF;AAEA,OAAO,SAASM,iBAAiBA,CAACC,IAAI,EAAEnB,KAAK,EAAE;EAC7C,IAAIA,KAAK,CAACY,CAAC,EAAE;IACXd,YAAY,CAACqB,IAAI,EAAEnB,KAAK,CAACY,CAAC,CAAC;EAC7B;AACF;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA"}

package/modules/xlsx/parseSpreadsheetContents.js

@@ -1,66 +1,162 @@
+var _excluded = ["parseNumber"];
+function _typeof(o) { "@babel/helpers - typeof"; return _typeof = "function" == typeof Symbol && "symbol" == typeof Symbol.iterator ? function (o) { return typeof o; } : function (o) { return o && "function" == typeof Symbol && o.constructor === Symbol && o !== Symbol.prototype ? "symbol" : typeof o; }, _typeof(o); }
 function _createForOfIteratorHelperLoose(o, allowArrayLike) { var it = typeof Symbol !== "undefined" && o[Symbol.iterator] || o["@@iterator"]; if (it) return (it = it.call(o)).next.bind(it); if (Array.isArray(o) || (it = _unsupportedIterableToArray(o)) || allowArrayLike && o && typeof o.length === "number") { if (it) o = it; var i = 0; return function () { if (i >= o.length) return { done: true }; return { done: false, value: o[i++] }; }; } throw new TypeError("Invalid attempt to iterate non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method."); }
 function _unsupportedIterableToArray(o, minLen) { if (!o) return; if (typeof o === "string") return _arrayLikeToArray(o, minLen); var n = Object.prototype.toString.call(o).slice(8, -1); if (n === "Object" && o.constructor) n = o.constructor.name; if (n === "Map" || n === "Set") return Array.from(o); if (n === "Arguments" || /^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(n)) return _arrayLikeToArray(o, minLen); }
 function _arrayLikeToArray(arr, len) { if (len == null || len > arr.length) len = arr.length; for (var i = 0, arr2 = new Array(len); i < len; i++) arr2[i] = arr[i]; return arr2; }
+function _objectWithoutProperties(source, excluded) { if (source == null) return {}; var target = _objectWithoutPropertiesLoose(source, excluded); var key, i; if (Object.getOwnPropertySymbols) { var sourceSymbolKeys = Object.getOwnPropertySymbols(source); for (i = 0; i < sourceSymbolKeys.length; i++) { key = sourceSymbolKeys[i]; if (excluded.indexOf(key) >= 0) continue; if (!Object.prototype.propertyIsEnumerable.call(source, key)) continue; target[key] = source[key]; } } return target; }
+function _objectWithoutPropertiesLoose(source, excluded) { if (source == null) return {}; var target = {}; var sourceKeys = Object.keys(source); var key, i; for (i = 0; i < sourceKeys.length; i++) { key = sourceKeys[i]; if (excluded.indexOf(key) >= 0) continue; target[key] = source[key]; } return target; }
+function ownKeys(e, r) { var t = Object.keys(e); if (Object.getOwnPropertySymbols) { var o = Object.getOwnPropertySymbols(e); r && (o = o.filter(function (r) { return Object.getOwnPropertyDescriptor(e, r).enumerable; })), t.push.apply(t, o); } return t; }
+function _objectSpread(e) { for (var r = 1; r < arguments.length; r++) { var t = null != arguments[r] ? arguments[r] : {}; r % 2 ? ownKeys(Object(t), !0).forEach(function (r) { _defineProperty(e, r, t[r]); }) : Object.getOwnPropertyDescriptors ? Object.defineProperties(e, Object.getOwnPropertyDescriptors(t)) : ownKeys(Object(t)).forEach(function (r) { Object.defineProperty(e, r, Object.getOwnPropertyDescriptor(t, r)); }); } return e; }
+function _defineProperty(obj, key, value) { key = _toPropertyKey(key); if (key in obj) { Object.defineProperty(obj, key, { value: value, enumerable: true, configurable: true, writable: true }); } else { obj[key] = value; } return obj; }
+function _toPropertyKey(arg) { var key = _toPrimitive(arg, "string"); return _typeof(key) === "symbol" ? key : String(key); }
+function _toPrimitive(input, hint) { if (_typeof(input) !== "object" || input === null) return input; var prim = input[Symbol.toPrimitive]; if (prim !== undefined) { var res = prim.call(input, hint || "default"); if (_typeof(res) !== "object") return res; throw new TypeError("@@toPrimitive must return a primitive value."); } return (hint === "string" ? String : Number)(input); }
 import parseSpreadsheetInfo from './parseSpreadsheetInfo.js';
 import parseFilePaths from './parseFilePaths.js';
 import parseStyles from './parseStyles.js';
 import parseSharedStrings from './parseSharedStrings.js';
 import parseSheet from './parseSheet.js';
+import parseNumberDefault from './parseNumber.js';
+import checkpoint, { latestCheckpointTimestamp } from '../utility/checkpoint.js';
+import isPromise from '../utility/isPromise.js';
 
 // For an introduction in reading `.xlsx` files see "The minimum viable XLSX reader":
 // https://www.brendanlong.com/the-minimum-viable-xlsx-reader.html
 
 /**
  * Reads data from an `.xlsx` file.
+ * @param  {function} parseXmlStream — SAX XML parser.
  * @param  {Record<string,string>} contents - A map of `.xml` files inside the `.xlsx` file (which itself is just a zipped directory).
- * @param  {object} xml — An object having a single property — `createDocument(string)` function.
  * @param  {object} [options]
- * @return {Sheet[]}
+ * @return {Promise<Sheet[]>}
  */
-export default function parseSpreadsheetContents(contents, xml) {
+function parseSpreadsheetContents(parseXmlStream, contents) {
   var options = arguments.length > 2 && arguments[2] !== undefined ? arguments[2] : {};
-  var getFileContent = function getFileContent(filePath) {
-    if (!contents[filePath]) {
-      throw new Error("\"".concat(filePath, "\" file not found inside the *.xlsx file zip archive"));
-    }
-    return contents[filePath];
-  };
+  // Because of how `.xlsx` file contents are defined in the specification,
+  // it will have to be read in 3 passes:
+  // * First pass — read the actual file paths
+  // * Second pass — read "shared strings" and "styles"
+  // * Thirs pass — read the sheets data
 
-  // Read the paths to certain files inside the `.xlsx` file, which is itself just a `.zip` archive.
-  // These paths aren't standardized between different spreadsheet editors.
-  // https://github.com/tidyverse/readxl/issues/104
-  var filePaths = parseFilePaths(getFileContent('xl/_rels/workbook.xml.rels'), xml);
-
-  // The usual file path for "shared strings" is "xl/sharedStrings.xml".
-  var sharedStrings = filePaths.sharedStrings ? parseSharedStrings(getFileContent(filePaths.sharedStrings), xml) : [];
-
-  // The usual file path for "styles" is "xl/styles.xml".
-  var styles = filePaths.styles ? parseStyles(getFileContent(filePaths.styles), xml) : {};
-  var _parseSpreadsheetInfo = parseSpreadsheetInfo(getFileContent('xl/workbook.xml'), xml),
-    sheets = _parseSpreadsheetInfo.sheets,
-    epoch1904 = _parseSpreadsheetInfo.epoch1904;
-  var sheetIdsToRead = options.sheets && options.sheets.map(function (sheet) {
-    return getSheetId(sheet, sheets);
-  });
+  checkpoint('parse spreadsheet info and file paths');
 
-  // Parse sheets data.
+  // Get spreadsheet info and the paths to files.
+  return readFiles(getXmlFilesAtFixedPaths(), contents, parseXmlStream).then(function (_ref) {
+    var spreadsheetInfo = _ref.spreadsheetInfo,
+      filePaths = _ref.filePaths;
+    checkpoint('parse "shared strings" and "styles"');
 
-  var sheetsData = [];
-  for (var _i = 0, _Object$keys = Object.keys(filePaths.sheets); _i < _Object$keys.length; _i++) {
-    var sheetId = _Object$keys[_i];
-    if (sheetIdsToRead && !sheetIdsToRead.includes(sheetId)) {
-      continue;
-    }
-    sheetsData.push({
-      sheet: getSheetNameById(sheetId, sheets),
-      data: parseSheet(getFileContent(filePaths.sheets[sheetId]), xml, sharedStrings, styles, epoch1904, options)
+    // Parse "shared strings" and "styles".
+    return readFiles(getXmlFilesAtNonFixedPaths(filePaths), contents, parseXmlStream).then(function (_ref2) {
+      var sharedStrings = _ref2.sharedStrings,
+        styles = _ref2.styles;
+      var sheetRelationIdsToRead = options.sheets ? options.sheets.map(function (sheet) {
+        return getSheetRelationId(sheet, spreadsheetInfo.sheets);
+      }) : spreadsheetInfo.sheets.map(function (_) {
+        return _.relationId;
+      });
+      checkpoint("parse sheet".concat(sheetRelationIdsToRead.length === 1 ? '' : 's', " data"));
+
+      // Parse sheets data.
+      return readFiles(getSheetDataXmlFiles(filePaths, sheetRelationIdsToRead, {
+        sharedStrings: sharedStrings,
+        styles: styles,
+        epoch1904: spreadsheetInfo.epoch1904,
+        options: options
+      }), contents, parseXmlStream).then(function (sheetsData) {
+        checkpoint('end');
+        // Return sheets data.
+        return sheetRelationIdsToRead.map(function (sheetRelationId) {
+          return {
+            sheet: getSheetNameByRelationId(sheetRelationId, spreadsheetInfo.sheets),
+            data: sheetsData[sheetRelationId]
+          };
+        });
+      });
+    });
+  });
+}
+
+/**
+ * Reads data from an `.xlsx` file in a worker.
+ * @param  {function} [createWorkerFunction] — Creates a worker function.
+ * @param  {function} parseXmlStream — SAX XML parser.
+ * @param  {Record<string,string>} contents - A map of `.xml` files inside the `.xlsx` file (which itself is just a zipped directory).
+ * @param  {object} [options]
+ * @return {Promise<Sheet[]>}
+ */
+export default function parseSpreadsheetContentsInWorker(createWorkerFunction, parseXmlStream, contents, options) {
+  // Assign the default `parseNumber()` function in the `options`.
+  // The reason is that the worker code requires it to be non-`undefined`.
+  // Otherwise, it would throw "parseNumber is not defined".
+  if (!(options && options.parseNumber)) {
+    options = _objectSpread(_objectSpread({}, options), {}, {
+      parseNumber: parseNumberDefault
     });
   }
 
-  // Return spreadsheet data.
-  return sheetsData;
+  // Using a worker requires specifying all the top-level variables or functions that it uses.
+  // Currently, that list looks a little bit too long so for now workers aren't used for parsing sheet data.
+  // See the comment in `createWorkerFunction()` call for more details.
+  // createWorkerFunction = undefined
+
+  // If the environment doesn't support "workers", parse spreadsheet contents "synchronously".
+  // This will "block" the main thread while parsing.
+  if (!createWorkerFunction) {
+    return parseSpreadsheetContents(parseXmlStream, contents, options);
+  }
+
+  // Any functions have to be removed from the `options` in order for them to be "serializable"
+  // before sending them to the worker thread.
+  var _options = options,
+    parseNumber = _options.parseNumber,
+    optionsJson = _objectWithoutProperties(_options, _excluded);
+
+  // Create a worker.
+  var workerFn = createWorkerFunction(function (data) {
+    // Reconstruct the `options`.
+    var options = _objectSpread(_objectSpread({}, data.optionsJson), {}, {
+      parseNumber: parseNumber
+    });
+    // Parse sheet data from the `.xml` files.
+    return parseSpreadsheetContents(parseXmlStream, data.contents, options);
+  });
+  workerFn.addDependencies(
+  // Any "outside" dependencies that're referenced in the function (below).
+  function () {
+    return [parseXmlStream, parseNumber, parseSpreadsheetContents, getSheetRelationId, getSheetNameByRelationId, getXmlFilesAtFixedPaths, getXmlFilesAtNonFixedPaths, getSheetDataXmlFiles, readFiles, checkpoint, latestCheckpointTimestamp, isPromise
+    // parseSheet,
+    //
+    // This is not the full list by any means. There's more. Quite a lot more of them.
+    // It started looking a bit too much so I stopped adding the dependencies here.
+    // Now I understand why `fflate`'s source code is all written in a single
+    // multi-thousand-line `index.ts` file. The thing is, once one starts splitting
+    // the code into modules, they'd have to manually export any top-level variables
+    // or functions from those modules and import them here in order to specify them
+    // in the list of dependencies. And even if doing so for every top-level variable
+    // or function in every imported module doesn't seem like an impossible task,
+    // imagine someone refactoring the code later and extracting new top-level
+    // variables or functions. Without 100% code coverage requirement, it won't be caught
+    // at build time and can only be caught in production, which isn't ideal to say the least.
+    // So it seems like users of this package will have to just deal with the "blocking"
+    // nature of the sheet data parser, because I won't follow into `fflate`'s steps
+    // and rewrite this package as a single `index.js` file.
+    // Users of this package will have to manually put their code in a worker
+    // in case they'd prefer it to run in a separate thread to prevent "blocking"
+    // the main thread when parsing sheet data.
+    ];
+  });
+
+  // Post a message with some data to the worker
+  // so that it starts processing the data
+  // and later posts a message back to the main thread
+  // with the result of the calculation.
+  return workerFn.callOnce({
+    optionsJson: optionsJson,
+    contents: contents
+  }); // (optional) add `transferList` argument.
 }
-function getSheetId(sheet, sheets) {
+
+function getSheetRelationId(sheet, sheets) {
   if (typeof sheet === 'string') {
     for (var _iterator = _createForOfIteratorHelperLoose(sheets), _step; !(_step = _iterator()).done;) {
       var _sheet = _step.value;
@@ -68,8 +164,8 @@ function getSheetId(sheet, sheets) {
         return _sheet.relationId;
       }
     }
-    throw new Error("Sheet \"".concat(sheet, "\" not found. Available sheets: ").concat(sheets.map(function (_ref) {
-      var name = _ref.name;
+    throw new Error("Sheet \"".concat(sheet, "\" not found. Available sheets: ").concat(sheets.map(function (_ref3) {
+      var name = _ref3.name;
       return "\"".concat(name, "\"");
     }).join(', ')));
   } else {
@@ -79,13 +175,116 @@ function getSheetId(sheet, sheets) {
     throw new Error("Sheet number out of bounds: ".concat(sheet, ". Available sheets count: ").concat(sheets.length));
   }
 }
-function getSheetNameById(sheetId, sheets) {
+function getSheetNameByRelationId(sheetRelationId, sheets) {
   for (var _iterator2 = _createForOfIteratorHelperLoose(sheets), _step2; !(_step2 = _iterator2()).done;) {
     var sheet = _step2.value;
-    if (sheet.relationId === sheetId) {
+    if (sheet.relationId === sheetRelationId) {
       return sheet.name;
     }
   }
-  throw new Error("Sheet ID not found: ".concat(sheetId));
+  throw new Error("Sheet relation ID not found: ".concat(sheetRelationId));
+}
+function getXmlFilesAtFixedPaths() {
+  return {
+    // Read the paths to certain files inside the `.xlsx` file, which is itself just a `.zip` archive.
+    // These paths aren't standardized between different spreadsheet editors.
+    // https://github.com/tidyverse/readxl/issues/104
+    'xl/_rels/workbook.xml.rels': {
+      name: 'filePaths',
+      parse: parseFilePaths
+    },
+    // General info on the spreadsheet.
+    'xl/workbook.xml': {
+      name: 'spreadsheetInfo',
+      parse: parseSpreadsheetInfo
+    }
+  };
+}
+function getXmlFilesAtNonFixedPaths(filePaths) {
+  var _ref4;
+  return _ref4 = {}, _defineProperty(_ref4, filePaths.sharedStrings || 'xl/sharedStrings.xml', {
+    name: 'sharedStrings',
+    // `parseSharedStrings()` returns a `Promise`.
+    parse: parseSharedStrings,
+    // It seems that "sharedStrings.xml" is not required to exist.
+    // For example, that could be the case when a spreadsheet doesn't contain any strings.
+    // https://github.com/catamphetamine/read-excel-file/issues/85
+    fallback: []
+  }), _defineProperty(_ref4, filePaths.styles || 'xl/styles.xml', {
+    name: 'styles',
+    parse: parseStyles,
+    fallback: {}
+  }), _ref4;
+}
+
+// Returns the list of sheet data `.xml` files.
+function getSheetDataXmlFiles(filePaths, sheetRelationIdsToRead, sheetDataParserParameters) {
+  return Object.keys(filePaths.sheets).filter(function (sheetRelationId) {
+    return sheetRelationIdsToRead.includes(sheetRelationId);
+  }).reduce(function (filesInfo, sheetRelationId) {
+    return _objectSpread(_objectSpread({}, filesInfo), {}, _defineProperty({}, filePaths.sheets[sheetRelationId], {
+      name: sheetRelationId,
+      // `parseSheet()` returns a `Promise`.
+      parse: function parse(content, parseXmlStream) {
+        return parseSheet(content, parseXmlStream, sheetDataParserParameters);
+      }
+    }));
+  }, {});
+}
+
+// In case of converting `.zip` file reader from a "read-and-return" one to a "streaming" one,
+// this function could be modified to process the files as they come rather than all-at-once.
+
+// Reads files from inside an `.xlsx` archive by file paths.
+//
+// In case of converting `.zip` file reader from a "read-and-return" one to a "streaming" one,
+// this function could be modified to process the files as they come rather than all-at-once.
+//
+// But there's a catch: inside an `.xlsx` file, some file paths are not fixed
+// and are instead defined in "xl/_rels/workbook.xml.rels" file,
+// which presents a "chicken and an egg" dilemma: how could one possibly
+// read an `.xlsx` file in one go when the order of the files inside it isn't fixed
+// and could be random. Most likely, in the majority of cases, "xl/_rels/workbook.xml.rels"
+// file is gonna be one of the first in a given `.xlsx` archive, but still it's not guaranteed.
+// A solution would be reading an `.xlsx` file in two passes: one pass would be just to read
+// the "xl/_rels/workbook.xml.rels" and ignore decompressing anything else,
+// and then the second pass would be to read all other files whose paths are now known.
+//
+// Returns:
+// * If none of the `parse()` functions returned a `Promise`, it returns a map of files' contents.
+// * If any of the `parse()` functions returned a `Promise`, it returns a `Promise` that resolves to a map of files' contents.
+//
+function readFiles(filesInfo, contents, parseXmlStream) {
+  // Get files' contents.
+  var results = {};
+  var _loop = function _loop() {
+    var filePath = _Object$keys[_i];
+    var fileInfo = filesInfo[filePath];
+    results[fileInfo.name] = contents[filePath] === undefined ? fileInfo.fallback === undefined ? function () {
+      throw new Error("\"".concat(filePath, "\" file not found inside the `.xlsx` file"));
+    }() : fileInfo.fallback : fileInfo.parse(contents[filePath], parseXmlStream);
+  };
+  for (var _i = 0, _Object$keys = Object.keys(filesInfo); _i < _Object$keys.length; _i++) {
+    _loop();
+  }
+  // Resolve any `Promise`s.
+  var promises = [];
+  var _loop2 = function _loop2() {
+    var name = _Object$keys2[_i2];
+    if (isPromise(results[name])) {
+      promises.push(results[name].then(function (result) {
+        results[name] = result;
+      }));
+    }
+  };
+  for (var _i2 = 0, _Object$keys2 = Object.keys(results); _i2 < _Object$keys2.length; _i2++) {
+    _loop2();
+  }
+  if (promises.length > 0) {
+    return Promise.all(promises).then(function () {
+      return results;
+    });
+  }
+  return results;
 }
 //# sourceMappingURL=parseSpreadsheetContents.js.map

package/modules/xlsx/parseSpreadsheetContents.js.map

@@ -1 +1 @@
-{"version":3,"file":"parseSpreadsheetContents.js","names":["parseSpreadsheetInfo","parseFilePaths","parseStyles","parseSharedStrings","parseSheet","parseSpreadsheetContents","contents","xml","options","arguments","length","undefined","getFileContent","filePath","Error","concat","filePaths","sharedStrings","styles","_parseSpreadsheetInfo","sheets","epoch1904","sheetIdsToRead","map","sheet","getSheetId","sheetsData","_i","_Object$keys","Object","keys","sheetId","includes","push","getSheetNameById","data","_iterator","_createForOfIteratorHelperLoose","_step","done","_sheet","value","name","relationId","_ref","join","_iterator2","_step2"],"sources":["../../source/xlsx/parseSpreadsheetContents.js"],"sourcesContent":["import parseSpreadsheetInfo from './parseSpreadsheetInfo.js'\r\nimport parseFilePaths from './parseFilePaths.js'\r\nimport parseStyles from './parseStyles.js'\r\nimport parseSharedStrings from './parseSharedStrings.js'\r\nimport parseSheet from './parseSheet.js'\r\n\r\n// For an introduction in reading `.xlsx` files see \"The minimum viable XLSX reader\":\r\n// https://www.brendanlong.com/the-minimum-viable-xlsx-reader.html\r\n\r\n/**\r\n * Reads data from an `.xlsx` file.\r\n * @param  {Record<string,string>} contents - A map of `.xml` files inside the `.xlsx` file (which itself is just a zipped directory).\r\n * @param  {object} xml — An object having a single property — `createDocument(string)` function.\r\n * @param  {object} [options]\r\n * @return {Sheet[]}\r\n */\r\nexport default function parseSpreadsheetContents(contents, xml, options = {}) {\r\n  const getFileContent = (filePath) => {\r\n    if (!contents[filePath]) {\r\n      throw new Error(`\"${filePath}\" file not found inside the *.xlsx file zip archive`)\r\n    }\r\n    return contents[filePath]\r\n  }\r\n\r\n  // Read the paths to certain files inside the `.xlsx` file, which is itself just a `.zip` archive.\r\n  // These paths aren't standardized between different spreadsheet editors.\r\n  // https://github.com/tidyverse/readxl/issues/104\r\n  const filePaths = parseFilePaths(getFileContent('xl/_rels/workbook.xml.rels'), xml)\r\n\r\n  // The usual file path for \"shared strings\" is \"xl/sharedStrings.xml\".\r\n  const sharedStrings = filePaths.sharedStrings\r\n    ? parseSharedStrings(getFileContent(filePaths.sharedStrings), xml)\r\n    : []\r\n\r\n  // The usual file path for \"styles\" is \"xl/styles.xml\".\r\n  const styles = filePaths.styles\r\n    ? parseStyles(getFileContent(filePaths.styles), xml)\r\n    : {}\r\n\r\n  const { sheets, epoch1904 } = parseSpreadsheetInfo(getFileContent('xl/workbook.xml'), xml)\r\n\r\n  const sheetIdsToRead = options.sheets && options.sheets.map(sheet => getSheetId(sheet, sheets))\r\n\r\n  // Parse sheets data.\r\n\r\n  const sheetsData = []\r\n\r\n  for (const sheetId of Object.keys(filePaths.sheets)) {\r\n    if (sheetIdsToRead && !sheetIdsToRead.includes(sheetId)) {\r\n      continue\r\n    }\r\n\r\n    sheetsData.push({\r\n      sheet: getSheetNameById(sheetId, sheets),\r\n      data: parseSheet(\r\n        getFileContent(filePaths.sheets[sheetId]),\r\n        xml,\r\n        sharedStrings,\r\n        styles,\r\n        epoch1904,\r\n        options\r\n      )\r\n    })\r\n  }\r\n\r\n  // Return spreadsheet data.\r\n  return sheetsData\r\n}\r\n\r\nfunction getSheetId(sheet, sheets) {\r\n  if (typeof sheet === 'string') {\r\n    for (const _sheet of sheets) {\r\n      if (_sheet.name === sheet) {\r\n        return _sheet.relationId\r\n      }\r\n    }\r\n\t\tthrow new Error(`Sheet \"${sheet}\" not found. Available sheets: ${sheets.map(({ name }) => `\"${name}\"`).join(', ')}`)\r\n  } else {\r\n\t\tif (sheet <= sheets.length) {\r\n      return sheets[sheet - 1].relationId\r\n    }\r\n    throw new Error(`Sheet number out of bounds: ${sheet}. Available sheets count: ${sheets.length}`)\r\n  }\r\n}\r\n\r\nfunction getSheetNameById(sheetId, sheets) {\r\n  for (const sheet of sheets) {\r\n    if (sheet.relationId === sheetId) {\r\n      return sheet.name\r\n    }\r\n  }\r\n  throw new Error(`Sheet ID not found: ${sheetId}`)\r\n}"],"mappings":";;;AAAA,OAAOA,oBAAoB,MAAM,2BAA2B;AAC5D,OAAOC,cAAc,MAAM,qBAAqB;AAChD,OAAOC,WAAW,MAAM,kBAAkB;AAC1C,OAAOC,kBAAkB,MAAM,yBAAyB;AACxD,OAAOC,UAAU,MAAM,iBAAiB;;AAExC;AACA;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASC,wBAAwBA,CAACC,QAAQ,EAAEC,GAAG,EAAgB;EAAA,IAAdC,OAAO,GAAAC,SAAA,CAAAC,MAAA,QAAAD,SAAA,QAAAE,SAAA,GAAAF,SAAA,MAAG,CAAC,CAAC;EAC1E,IAAMG,cAAc,GAAG,SAAjBA,cAAcA,CAAIC,QAAQ,EAAK;IACnC,IAAI,CAACP,QAAQ,CAACO,QAAQ,CAAC,EAAE;MACvB,MAAM,IAAIC,KAAK,MAAAC,MAAA,CAAKF,QAAQ,yDAAqD,CAAC;IACpF;IACA,OAAOP,QAAQ,CAACO,QAAQ,CAAC;EAC3B,CAAC;;EAED;EACA;EACA;EACA,IAAMG,SAAS,GAAGf,cAAc,CAACW,cAAc,CAAC,4BAA4B,CAAC,EAAEL,GAAG,CAAC;;EAEnF;EACA,IAAMU,aAAa,GAAGD,SAAS,CAACC,aAAa,GACzCd,kBAAkB,CAACS,cAAc,CAACI,SAAS,CAACC,aAAa,CAAC,EAAEV,GAAG,CAAC,GAChE,EAAE;;EAEN;EACA,IAAMW,MAAM,GAAGF,SAAS,CAACE,MAAM,GAC3BhB,WAAW,CAACU,cAAc,CAACI,SAAS,CAACE,MAAM,CAAC,EAAEX,GAAG,CAAC,GAClD,CAAC,CAAC;EAEN,IAAAY,qBAAA,GAA8BnB,oBAAoB,CAACY,cAAc,CAAC,iBAAiB,CAAC,EAAEL,GAAG,CAAC;IAAlFa,MAAM,GAAAD,qBAAA,CAANC,MAAM;IAAEC,SAAS,GAAAF,qBAAA,CAATE,SAAS;EAEzB,IAAMC,cAAc,GAAGd,OAAO,CAACY,MAAM,IAAIZ,OAAO,CAACY,MAAM,CAACG,GAAG,CAAC,UAAAC,KAAK;IAAA,OAAIC,UAAU,CAACD,KAAK,EAAEJ,MAAM,CAAC;EAAA,EAAC;;EAE/F;;EAEA,IAAMM,UAAU,GAAG,EAAE;EAErB,SAAAC,EAAA,MAAAC,YAAA,GAAsBC,MAAM,CAACC,IAAI,CAACd,SAAS,CAACI,MAAM,CAAC,EAAAO,EAAA,GAAAC,YAAA,CAAAlB,MAAA,EAAAiB,EAAA,IAAE;IAAhD,IAAMI,OAAO,GAAAH,YAAA,CAAAD,EAAA;IAChB,IAAIL,cAAc,IAAI,CAACA,cAAc,CAACU,QAAQ,CAACD,OAAO,CAAC,EAAE;MACvD;IACF;IAEAL,UAAU,CAACO,IAAI,CAAC;MACdT,KAAK,EAAEU,gBAAgB,CAACH,OAAO,EAAEX,MAAM,CAAC;MACxCe,IAAI,EAAE/B,UAAU,CACdQ,cAAc,CAACI,SAAS,CAACI,MAAM,CAACW,OAAO,CAAC,CAAC,EACzCxB,GAAG,EACHU,aAAa,EACbC,MAAM,EACNG,SAAS,EACTb,OACF;IACF,CAAC,CAAC;EACJ;;EAEA;EACA,OAAOkB,UAAU;AACnB;AAEA,SAASD,UAAUA,CAACD,KAAK,EAAEJ,MAAM,EAAE;EACjC,IAAI,OAAOI,KAAK,KAAK,QAAQ,EAAE;IAC7B,SAAAY,SAAA,GAAAC,+BAAA,CAAqBjB,MAAM,GAAAkB,KAAA,IAAAA,KAAA,GAAAF,SAAA,IAAAG,IAAA,GAAE;MAAA,IAAlBC,MAAM,GAAAF,KAAA,CAAAG,KAAA;MACf,IAAID,MAAM,CAACE,IAAI,KAAKlB,KAAK,EAAE;QACzB,OAAOgB,MAAM,CAACG,UAAU;MAC1B;IACF;IACF,MAAM,IAAI7B,KAAK,YAAAC,MAAA,CAAWS,KAAK,sCAAAT,MAAA,CAAkCK,MAAM,CAACG,GAAG,CAAC,UAAAqB,IAAA;MAAA,IAAGF,IAAI,GAAAE,IAAA,CAAJF,IAAI;MAAA,YAAA3B,MAAA,CAAW2B,IAAI;IAAA,CAAG,CAAC,CAACG,IAAI,CAAC,IAAI,CAAC,CAAE,CAAC;EACpH,CAAC,MAAM;IACP,IAAIrB,KAAK,IAAIJ,MAAM,CAACV,MAAM,EAAE;MACxB,OAAOU,MAAM,CAACI,KAAK,GAAG,CAAC,CAAC,CAACmB,UAAU;IACrC;IACA,MAAM,IAAI7B,KAAK,gCAAAC,MAAA,CAAgCS,KAAK,gCAAAT,MAAA,CAA6BK,MAAM,CAACV,MAAM,CAAE,CAAC;EACnG;AACF;AAEA,SAASwB,gBAAgBA,CAACH,OAAO,EAAEX,MAAM,EAAE;EACzC,SAAA0B,UAAA,GAAAT,+BAAA,CAAoBjB,MAAM,GAAA2B,MAAA,IAAAA,MAAA,GAAAD,UAAA,IAAAP,IAAA,GAAE;IAAA,IAAjBf,KAAK,GAAAuB,MAAA,CAAAN,KAAA;IACd,IAAIjB,KAAK,CAACmB,UAAU,KAAKZ,OAAO,EAAE;MAChC,OAAOP,KAAK,CAACkB,IAAI;IACnB;EACF;EACA,MAAM,IAAI5B,KAAK,wBAAAC,MAAA,CAAwBgB,OAAO,CAAE,CAAC;AACnD"}
+{"version":3,"file":"parseSpreadsheetContents.js","names":["parseSpreadsheetInfo","parseFilePaths","parseStyles","parseSharedStrings","parseSheet","parseNumberDefault","checkpoint","latestCheckpointTimestamp","isPromise","parseSpreadsheetContents","parseXmlStream","contents","options","arguments","length","undefined","readFiles","getXmlFilesAtFixedPaths","then","_ref","spreadsheetInfo","filePaths","getXmlFilesAtNonFixedPaths","_ref2","sharedStrings","styles","sheetRelationIdsToRead","sheets","map","sheet","getSheetRelationId","_","relationId","concat","getSheetDataXmlFiles","epoch1904","sheetsData","sheetRelationId","getSheetNameByRelationId","data","parseSpreadsheetContentsInWorker","createWorkerFunction","parseNumber","_objectSpread","_options","optionsJson","_objectWithoutProperties","_excluded","workerFn","addDependencies","callOnce","_iterator","_createForOfIteratorHelperLoose","_step","done","_sheet","value","name","Error","_ref3","join","_iterator2","_step2","parse","_ref4","_defineProperty","fallback","sheetDataParserParameters","Object","keys","filter","includes","reduce","filesInfo","content","results","_loop","filePath","_Object$keys","_i","fileInfo","promises","_loop2","_Object$keys2","_i2","push","result","Promise","all"],"sources":["../../source/xlsx/parseSpreadsheetContents.js"],"sourcesContent":["import parseSpreadsheetInfo from './parseSpreadsheetInfo.js'\r\nimport parseFilePaths from './parseFilePaths.js'\r\nimport parseStyles from './parseStyles.js'\r\nimport parseSharedStrings from './parseSharedStrings.js'\r\nimport parseSheet from './parseSheet.js'\r\nimport parseNumberDefault from './parseNumber.js'\r\n\r\nimport checkpoint, { latestCheckpointTimestamp } from '../utility/checkpoint.js'\r\nimport isPromise from '../utility/isPromise.js'\r\n\r\n// For an introduction in reading `.xlsx` files see \"The minimum viable XLSX reader\":\r\n// https://www.brendanlong.com/the-minimum-viable-xlsx-reader.html\r\n\r\n/**\r\n * Reads data from an `.xlsx` file.\r\n * @param  {function} parseXmlStream — SAX XML parser.\r\n * @param  {Record<string,string>} contents - A map of `.xml` files inside the `.xlsx` file (which itself is just a zipped directory).\r\n * @param  {object} [options]\r\n * @return {Promise<Sheet[]>}\r\n */\r\nfunction parseSpreadsheetContents(parseXmlStream, contents, options = {}) {\r\n  // Because of how `.xlsx` file contents are defined in the specification,\r\n  // it will have to be read in 3 passes:\r\n  // * First pass — read the actual file paths\r\n  // * Second pass — read \"shared strings\" and \"styles\"\r\n  // * Thirs pass — read the sheets data\r\n\r\n  checkpoint('parse spreadsheet info and file paths')\r\n\r\n  // Get spreadsheet info and the paths to files.\r\n  return readFiles(\r\n    getXmlFilesAtFixedPaths(),\r\n    contents,\r\n    parseXmlStream\r\n  ).then(({ spreadsheetInfo, filePaths }) => {\r\n    checkpoint('parse \"shared strings\" and \"styles\"')\r\n\r\n    // Parse \"shared strings\" and \"styles\".\r\n    return readFiles(\r\n      getXmlFilesAtNonFixedPaths(filePaths),\r\n      contents,\r\n      parseXmlStream\r\n    ).then(({ sharedStrings, styles }) => {\r\n      const sheetRelationIdsToRead = options.sheets\r\n        ? options.sheets.map(sheet => getSheetRelationId(sheet, spreadsheetInfo.sheets))\r\n        : spreadsheetInfo.sheets.map(_ => _.relationId)\r\n\r\n      checkpoint(`parse sheet${sheetRelationIdsToRead.length === 1 ? '' : 's'} data`)\r\n\r\n      // Parse sheets data.\r\n      return readFiles(\r\n        getSheetDataXmlFiles(filePaths, sheetRelationIdsToRead, {\r\n          sharedStrings,\r\n          styles,\r\n          epoch1904: spreadsheetInfo.epoch1904,\r\n          options\r\n        }),\r\n        contents,\r\n        parseXmlStream\r\n      ).then((sheetsData) => {\r\n        checkpoint('end')\r\n        // Return sheets data.\r\n        return sheetRelationIdsToRead.map((sheetRelationId) => ({\r\n          sheet: getSheetNameByRelationId(sheetRelationId, spreadsheetInfo.sheets),\r\n          data: sheetsData[sheetRelationId]\r\n        }))\r\n      })\r\n    })\r\n  })\r\n}\r\n\r\n/**\r\n * Reads data from an `.xlsx` file in a worker.\r\n * @param  {function} [createWorkerFunction] — Creates a worker function.\r\n * @param  {function} parseXmlStream — SAX XML parser.\r\n * @param  {Record<string,string>} contents - A map of `.xml` files inside the `.xlsx` file (which itself is just a zipped directory).\r\n * @param  {object} [options]\r\n * @return {Promise<Sheet[]>}\r\n */\r\nexport default function parseSpreadsheetContentsInWorker(createWorkerFunction, parseXmlStream, contents, options) {\r\n  // Assign the default `parseNumber()` function in the `options`.\r\n  // The reason is that the worker code requires it to be non-`undefined`.\r\n  // Otherwise, it would throw \"parseNumber is not defined\".\r\n  if (!(options && options.parseNumber)) {\r\n    options = {\r\n      ...options,\r\n      parseNumber: parseNumberDefault\r\n    }\r\n  }\r\n\r\n  // Using a worker requires specifying all the top-level variables or functions that it uses.\r\n  // Currently, that list looks a little bit too long so for now workers aren't used for parsing sheet data.\r\n  // See the comment in `createWorkerFunction()` call for more details.\r\n  // createWorkerFunction = undefined\r\n\r\n  // If the environment doesn't support \"workers\", parse spreadsheet contents \"synchronously\".\r\n  // This will \"block\" the main thread while parsing.\r\n  if (!createWorkerFunction) {\r\n    return parseSpreadsheetContents(parseXmlStream, contents, options)\r\n  }\r\n\r\n  // Any functions have to be removed from the `options` in order for them to be \"serializable\"\r\n  // before sending them to the worker thread.\r\n  const { parseNumber, ...optionsJson } = options\r\n\r\n  // Create a worker.\r\n  const workerFn = createWorkerFunction(\r\n    (data) => {\r\n      // Reconstruct the `options`.\r\n      const options = {\r\n        ...data.optionsJson,\r\n        parseNumber\r\n      }\r\n      // Parse sheet data from the `.xml` files.\r\n      return parseSpreadsheetContents(parseXmlStream, data.contents, options)\r\n    }\r\n  )\r\n\r\n  workerFn.addDependencies(\r\n    // Any \"outside\" dependencies that're referenced in the function (below).\r\n    () => [\r\n      parseXmlStream,\r\n      parseNumber,\r\n      parseSpreadsheetContents,\r\n      getSheetRelationId,\r\n      getSheetNameByRelationId,\r\n      getXmlFilesAtFixedPaths,\r\n      getXmlFilesAtNonFixedPaths,\r\n      getSheetDataXmlFiles,\r\n      readFiles,\r\n      checkpoint,\r\n      latestCheckpointTimestamp,\r\n      isPromise,\r\n      // parseSheet,\r\n      //\r\n      // This is not the full list by any means. There's more. Quite a lot more of them.\r\n      // It started looking a bit too much so I stopped adding the dependencies here.\r\n      // Now I understand why `fflate`'s source code is all written in a single\r\n      // multi-thousand-line `index.ts` file. The thing is, once one starts splitting\r\n      // the code into modules, they'd have to manually export any top-level variables\r\n      // or functions from those modules and import them here in order to specify them\r\n      // in the list of dependencies. And even if doing so for every top-level variable\r\n      // or function in every imported module doesn't seem like an impossible task,\r\n      // imagine someone refactoring the code later and extracting new top-level\r\n      // variables or functions. Without 100% code coverage requirement, it won't be caught\r\n      // at build time and can only be caught in production, which isn't ideal to say the least.\r\n      // So it seems like users of this package will have to just deal with the \"blocking\"\r\n      // nature of the sheet data parser, because I won't follow into `fflate`'s steps\r\n      // and rewrite this package as a single `index.js` file.\r\n      // Users of this package will have to manually put their code in a worker\r\n      // in case they'd prefer it to run in a separate thread to prevent \"blocking\"\r\n      // the main thread when parsing sheet data.\r\n    ]\r\n  )\r\n\r\n  // Post a message with some data to the worker\r\n  // so that it starts processing the data\r\n  // and later posts a message back to the main thread\r\n  // with the result of the calculation.\r\n  return workerFn.callOnce({ optionsJson, contents }) // (optional) add `transferList` argument.\r\n}\r\n\r\nfunction getSheetRelationId(sheet, sheets) {\r\n  if (typeof sheet === 'string') {\r\n    for (const _sheet of sheets) {\r\n      if (_sheet.name === sheet) {\r\n        return _sheet.relationId\r\n      }\r\n    }\r\n\t\tthrow new Error(`Sheet \"${sheet}\" not found. Available sheets: ${sheets.map(({ name }) => `\"${name}\"`).join(', ')}`)\r\n  } else {\r\n\t\tif (sheet <= sheets.length) {\r\n      return sheets[sheet - 1].relationId\r\n    }\r\n    throw new Error(`Sheet number out of bounds: ${sheet}. Available sheets count: ${sheets.length}`)\r\n  }\r\n}\r\n\r\nfunction getSheetNameByRelationId(sheetRelationId, sheets) {\r\n  for (const sheet of sheets) {\r\n    if (sheet.relationId === sheetRelationId) {\r\n      return sheet.name\r\n    }\r\n  }\r\n  throw new Error(`Sheet relation ID not found: ${sheetRelationId}`)\r\n}\r\n\r\nfunction getXmlFilesAtFixedPaths() {\r\n  return {\r\n    // Read the paths to certain files inside the `.xlsx` file, which is itself just a `.zip` archive.\r\n    // These paths aren't standardized between different spreadsheet editors.\r\n    // https://github.com/tidyverse/readxl/issues/104\r\n    'xl/_rels/workbook.xml.rels': {\r\n      name: 'filePaths',\r\n      parse: parseFilePaths\r\n    },\r\n\r\n    // General info on the spreadsheet.\r\n    'xl/workbook.xml': {\r\n      name: 'spreadsheetInfo',\r\n      parse: parseSpreadsheetInfo\r\n    }\r\n  }\r\n}\r\n\r\nfunction getXmlFilesAtNonFixedPaths(filePaths) {\r\n  return {\r\n    // The usual file path for \"shared strings\" is \"xl/sharedStrings.xml\".\r\n    [filePaths.sharedStrings || 'xl/sharedStrings.xml']: {\r\n      name: 'sharedStrings',\r\n      // `parseSharedStrings()` returns a `Promise`.\r\n      parse: parseSharedStrings,\r\n      // It seems that \"sharedStrings.xml\" is not required to exist.\r\n      // For example, that could be the case when a spreadsheet doesn't contain any strings.\r\n      // https://github.com/catamphetamine/read-excel-file/issues/85\r\n      fallback: []\r\n    },\r\n\r\n    // The usual file path for \"styles\" is \"xl/styles.xml\".\r\n    [filePaths.styles || 'xl/styles.xml']: {\r\n      name: 'styles',\r\n      parse: parseStyles,\r\n      fallback: {}\r\n    }\r\n  }\r\n}\r\n\r\n// Returns the list of sheet data `.xml` files.\r\nfunction getSheetDataXmlFiles(filePaths, sheetRelationIdsToRead, sheetDataParserParameters) {\r\n  return Object.keys(filePaths.sheets)\r\n    .filter((sheetRelationId) => sheetRelationIdsToRead.includes(sheetRelationId))\r\n    .reduce((filesInfo, sheetRelationId) => ({\r\n      ...filesInfo,\r\n      [filePaths.sheets[sheetRelationId]]: {\r\n        name: sheetRelationId,\r\n        // `parseSheet()` returns a `Promise`.\r\n        parse: (content, parseXmlStream) => parseSheet(content, parseXmlStream, sheetDataParserParameters)\r\n      }\r\n    }), {})\r\n}\r\n\r\n// In case of converting `.zip` file reader from a \"read-and-return\" one to a \"streaming\" one,\r\n// this function could be modified to process the files as they come rather than all-at-once.\r\n\r\n// Reads files from inside an `.xlsx` archive by file paths.\r\n//\r\n// In case of converting `.zip` file reader from a \"read-and-return\" one to a \"streaming\" one,\r\n// this function could be modified to process the files as they come rather than all-at-once.\r\n//\r\n// But there's a catch: inside an `.xlsx` file, some file paths are not fixed\r\n// and are instead defined in \"xl/_rels/workbook.xml.rels\" file,\r\n// which presents a \"chicken and an egg\" dilemma: how could one possibly\r\n// read an `.xlsx` file in one go when the order of the files inside it isn't fixed\r\n// and could be random. Most likely, in the majority of cases, \"xl/_rels/workbook.xml.rels\"\r\n// file is gonna be one of the first in a given `.xlsx` archive, but still it's not guaranteed.\r\n// A solution would be reading an `.xlsx` file in two passes: one pass would be just to read\r\n// the \"xl/_rels/workbook.xml.rels\" and ignore decompressing anything else,\r\n// and then the second pass would be to read all other files whose paths are now known.\r\n//\r\n// Returns:\r\n// * If none of the `parse()` functions returned a `Promise`, it returns a map of files' contents.\r\n// * If any of the `parse()` functions returned a `Promise`, it returns a `Promise` that resolves to a map of files' contents.\r\n//\r\nfunction readFiles(filesInfo, contents, parseXmlStream) {\r\n  // Get files' contents.\r\n  const results = {}\r\n  for (const filePath of Object.keys(filesInfo)) {\r\n    const fileInfo = filesInfo[filePath]\r\n    results[fileInfo.name] = contents[filePath] === undefined\r\n      ? (\r\n        fileInfo.fallback === undefined\r\n          ? (() => { throw new Error(`\"${filePath}\" file not found inside the \\`.xlsx\\` file`) })()\r\n          : fileInfo.fallback\r\n      )\r\n      : fileInfo.parse(contents[filePath], parseXmlStream)\r\n  }\r\n  // Resolve any `Promise`s.\r\n  const promises = []\r\n  for (const name of Object.keys(results)) {\r\n    if (isPromise(results[name])) {\r\n      promises.push(results[name].then((result) => {\r\n        results[name] = result\r\n      }))\r\n    }\r\n  }\r\n  if (promises.length > 0) {\r\n    return Promise.all(promises).then(() => results)\r\n  }\r\n  return results\r\n}"],"mappings":";;;;;;;;;;;;AAAA,OAAOA,oBAAoB,MAAM,2BAA2B;AAC5D,OAAOC,cAAc,MAAM,qBAAqB;AAChD,OAAOC,WAAW,MAAM,kBAAkB;AAC1C,OAAOC,kBAAkB,MAAM,yBAAyB;AACxD,OAAOC,UAAU,MAAM,iBAAiB;AACxC,OAAOC,kBAAkB,MAAM,kBAAkB;AAEjD,OAAOC,UAAU,IAAIC,yBAAyB,QAAQ,0BAA0B;AAChF,OAAOC,SAAS,MAAM,yBAAyB;;AAE/C;AACA;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASC,wBAAwBA,CAACC,cAAc,EAAEC,QAAQ,EAAgB;EAAA,IAAdC,OAAO,GAAAC,SAAA,CAAAC,MAAA,QAAAD,SAAA,QAAAE,SAAA,GAAAF,SAAA,MAAG,CAAC,CAAC;EACtE;EACA;EACA;EACA;EACA;;EAEAP,UAAU,CAAC,uCAAuC,CAAC;;EAEnD;EACA,OAAOU,SAAS,CACdC,uBAAuB,CAAC,CAAC,EACzBN,QAAQ,EACRD,cACF,CAAC,CAACQ,IAAI,CAAC,UAAAC,IAAA,EAAoC;IAAA,IAAjCC,eAAe,GAAAD,IAAA,CAAfC,eAAe;MAAEC,SAAS,GAAAF,IAAA,CAATE,SAAS;IAClCf,UAAU,CAAC,qCAAqC,CAAC;;IAEjD;IACA,OAAOU,SAAS,CACdM,0BAA0B,CAACD,SAAS,CAAC,EACrCV,QAAQ,EACRD,cACF,CAAC,CAACQ,IAAI,CAAC,UAAAK,KAAA,EAA+B;MAAA,IAA5BC,aAAa,GAAAD,KAAA,CAAbC,aAAa;QAAEC,MAAM,GAAAF,KAAA,CAANE,MAAM;MAC7B,IAAMC,sBAAsB,GAAGd,OAAO,CAACe,MAAM,GACzCf,OAAO,CAACe,MAAM,CAACC,GAAG,CAAC,UAAAC,KAAK;QAAA,OAAIC,kBAAkB,CAACD,KAAK,EAAET,eAAe,CAACO,MAAM,CAAC;MAAA,EAAC,GAC9EP,eAAe,CAACO,MAAM,CAACC,GAAG,CAAC,UAAAG,CAAC;QAAA,OAAIA,CAAC,CAACC,UAAU;MAAA,EAAC;MAEjD1B,UAAU,eAAA2B,MAAA,CAAeP,sBAAsB,CAACZ,MAAM,KAAK,CAAC,GAAG,EAAE,GAAG,GAAG,UAAO,CAAC;;MAE/E;MACA,OAAOE,SAAS,CACdkB,oBAAoB,CAACb,SAAS,EAAEK,sBAAsB,EAAE;QACtDF,aAAa,EAAbA,aAAa;QACbC,MAAM,EAANA,MAAM;QACNU,SAAS,EAAEf,eAAe,CAACe,SAAS;QACpCvB,OAAO,EAAPA;MACF,CAAC,CAAC,EACFD,QAAQ,EACRD,cACF,CAAC,CAACQ,IAAI,CAAC,UAACkB,UAAU,EAAK;QACrB9B,UAAU,CAAC,KAAK,CAAC;QACjB;QACA,OAAOoB,sBAAsB,CAACE,GAAG,CAAC,UAACS,eAAe;UAAA,OAAM;YACtDR,KAAK,EAAES,wBAAwB,CAACD,eAAe,EAAEjB,eAAe,CAACO,MAAM,CAAC;YACxEY,IAAI,EAAEH,UAAU,CAACC,eAAe;UAClC,CAAC;QAAA,CAAC,CAAC;MACL,CAAC,CAAC;IACJ,CAAC,CAAC;EACJ,CAAC,CAAC;AACJ;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASG,gCAAgCA,CAACC,oBAAoB,EAAE/B,cAAc,EAAEC,QAAQ,EAAEC,OAAO,EAAE;EAChH;EACA;EACA;EACA,IAAI,EAAEA,OAAO,IAAIA,OAAO,CAAC8B,WAAW,CAAC,EAAE;IACrC9B,OAAO,GAAA+B,aAAA,CAAAA,aAAA,KACF/B,OAAO;MACV8B,WAAW,EAAErC;IAAkB,EAChC;EACH;;EAEA;EACA;EACA;EACA;;EAEA;EACA;EACA,IAAI,CAACoC,oBAAoB,EAAE;IACzB,OAAOhC,wBAAwB,CAACC,cAAc,EAAEC,QAAQ,EAAEC,OAAO,CAAC;EACpE;;EAEA;EACA;EACA,IAAAgC,QAAA,GAAwChC,OAAO;IAAvC8B,WAAW,GAAAE,QAAA,CAAXF,WAAW;IAAKG,WAAW,GAAAC,wBAAA,CAAAF,QAAA,EAAAG,SAAA;;EAEnC;EACA,IAAMC,QAAQ,GAAGP,oBAAoB,CACnC,UAACF,IAAI,EAAK;IACR;IACA,IAAM3B,OAAO,GAAA+B,aAAA,CAAAA,aAAA,KACRJ,IAAI,CAACM,WAAW;MACnBH,WAAW,EAAXA;IAAW,EACZ;IACD;IACA,OAAOjC,wBAAwB,CAACC,cAAc,EAAE6B,IAAI,CAAC5B,QAAQ,EAAEC,OAAO,CAAC;EACzE,CACF,CAAC;EAEDoC,QAAQ,CAACC,eAAe;EACtB;EACA;IAAA,OAAM,CACJvC,cAAc,EACdgC,WAAW,EACXjC,wBAAwB,EACxBqB,kBAAkB,EAClBQ,wBAAwB,EACxBrB,uBAAuB,EACvBK,0BAA0B,EAC1BY,oBAAoB,EACpBlB,SAAS,EACTV,UAAU,EACVC,yBAAyB,EACzBC;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IAAA,CACD;EAAA,CACH,CAAC;;EAED;EACA;EACA;EACA;EACA,OAAOwC,QAAQ,CAACE,QAAQ,CAAC;IAAEL,WAAW,EAAXA,WAAW;IAAElC,QAAQ,EAARA;EAAS,CAAC,CAAC,EAAC;AACtD;;AAEA,SAASmB,kBAAkBA,CAACD,KAAK,EAAEF,MAAM,EAAE;EACzC,IAAI,OAAOE,KAAK,KAAK,QAAQ,EAAE;IAC7B,SAAAsB,SAAA,GAAAC,+BAAA,CAAqBzB,MAAM,GAAA0B,KAAA,IAAAA,KAAA,GAAAF,SAAA,IAAAG,IAAA,GAAE;MAAA,IAAlBC,MAAM,GAAAF,KAAA,CAAAG,KAAA;MACf,IAAID,MAAM,CAACE,IAAI,KAAK5B,KAAK,EAAE;QACzB,OAAO0B,MAAM,CAACvB,UAAU;MAC1B;IACF;IACF,MAAM,IAAI0B,KAAK,YAAAzB,MAAA,CAAWJ,KAAK,sCAAAI,MAAA,CAAkCN,MAAM,CAACC,GAAG,CAAC,UAAA+B,KAAA;MAAA,IAAGF,IAAI,GAAAE,KAAA,CAAJF,IAAI;MAAA,YAAAxB,MAAA,CAAWwB,IAAI;IAAA,CAAG,CAAC,CAACG,IAAI,CAAC,IAAI,CAAC,CAAE,CAAC;EACpH,CAAC,MAAM;IACP,IAAI/B,KAAK,IAAIF,MAAM,CAACb,MAAM,EAAE;MACxB,OAAOa,MAAM,CAACE,KAAK,GAAG,CAAC,CAAC,CAACG,UAAU;IACrC;IACA,MAAM,IAAI0B,KAAK,gCAAAzB,MAAA,CAAgCJ,KAAK,gCAAAI,MAAA,CAA6BN,MAAM,CAACb,MAAM,CAAE,CAAC;EACnG;AACF;AAEA,SAASwB,wBAAwBA,CAACD,eAAe,EAAEV,MAAM,EAAE;EACzD,SAAAkC,UAAA,GAAAT,+BAAA,CAAoBzB,MAAM,GAAAmC,MAAA,IAAAA,MAAA,GAAAD,UAAA,IAAAP,IAAA,GAAE;IAAA,IAAjBzB,KAAK,GAAAiC,MAAA,CAAAN,KAAA;IACd,IAAI3B,KAAK,CAACG,UAAU,KAAKK,eAAe,EAAE;MACxC,OAAOR,KAAK,CAAC4B,IAAI;IACnB;EACF;EACA,MAAM,IAAIC,KAAK,iCAAAzB,MAAA,CAAiCI,eAAe,CAAE,CAAC;AACpE;AAEA,SAASpB,uBAAuBA,CAAA,EAAG;EACjC,OAAO;IACL;IACA;IACA;IACA,4BAA4B,EAAE;MAC5BwC,IAAI,EAAE,WAAW;MACjBM,KAAK,EAAE9D;IACT,CAAC;IAED;IACA,iBAAiB,EAAE;MACjBwD,IAAI,EAAE,iBAAiB;MACvBM,KAAK,EAAE/D;IACT;EACF,CAAC;AACH;AAEA,SAASsB,0BAA0BA,CAACD,SAAS,EAAE;EAAA,IAAA2C,KAAA;EAC7C,OAAAA,KAAA,OAAAC,eAAA,CAAAD,KAAA,EAEG3C,SAAS,CAACG,aAAa,IAAI,sBAAsB,EAAG;IACnDiC,IAAI,EAAE,eAAe;IACrB;IACAM,KAAK,EAAE5D,kBAAkB;IACzB;IACA;IACA;IACA+D,QAAQ,EAAE;EACZ,CAAC,GAAAD,eAAA,CAAAD,KAAA,EAGA3C,SAAS,CAACI,MAAM,IAAI,eAAe,EAAG;IACrCgC,IAAI,EAAE,QAAQ;IACdM,KAAK,EAAE7D,WAAW;IAClBgE,QAAQ,EAAE,CAAC;EACb,CAAC,GAAAF,KAAA;AAEL;;AAEA;AACA,SAAS9B,oBAAoBA,CAACb,SAAS,EAAEK,sBAAsB,EAAEyC,yBAAyB,EAAE;EAC1F,OAAOC,MAAM,CAACC,IAAI,CAAChD,SAAS,CAACM,MAAM,CAAC,CACjC2C,MAAM,CAAC,UAACjC,eAAe;IAAA,OAAKX,sBAAsB,CAAC6C,QAAQ,CAAClC,eAAe,CAAC;EAAA,EAAC,CAC7EmC,MAAM,CAAC,UAACC,SAAS,EAAEpC,eAAe;IAAA,OAAAM,aAAA,CAAAA,aAAA,KAC9B8B,SAAS,OAAAR,eAAA,KACX5C,SAAS,CAACM,MAAM,CAACU,eAAe,CAAC,EAAG;MACnCoB,IAAI,EAAEpB,eAAe;MACrB;MACA0B,KAAK,EAAE,SAAAA,MAACW,OAAO,EAAEhE,cAAc;QAAA,OAAKN,UAAU,CAACsE,OAAO,EAAEhE,cAAc,EAAEyD,yBAAyB,CAAC;MAAA;IACpG,CAAC;EAAA,CACD,EAAE,CAAC,CAAC,CAAC;AACX;;AAEA;AACA;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASnD,SAASA,CAACyD,SAAS,EAAE9D,QAAQ,EAAED,cAAc,EAAE;EACtD;EACA,IAAMiE,OAAO,GAAG,CAAC,CAAC;EAAA,IAAAC,KAAA,YAAAA,MAAA,EAC6B;IAA1C,IAAMC,QAAQ,GAAAC,YAAA,CAAAC,EAAA;IACjB,IAAMC,QAAQ,GAAGP,SAAS,CAACI,QAAQ,CAAC;IACpCF,OAAO,CAACK,QAAQ,CAACvB,IAAI,CAAC,GAAG9C,QAAQ,CAACkE,QAAQ,CAAC,KAAK9D,SAAS,GAErDiE,QAAQ,CAACd,QAAQ,KAAKnD,SAAS,GAC1B,YAAM;MAAE,MAAM,IAAI2C,KAAK,MAAAzB,MAAA,CAAK4C,QAAQ,8CAA4C,CAAC;IAAC,CAAC,CAAE,CAAC,GACvFG,QAAQ,CAACd,QAAQ,GAErBc,QAAQ,CAACjB,KAAK,CAACpD,QAAQ,CAACkE,QAAQ,CAAC,EAAEnE,cAAc,CAAC;EACxD,CAAC;EATD,SAAAqE,EAAA,MAAAD,YAAA,GAAuBV,MAAM,CAACC,IAAI,CAACI,SAAS,CAAC,EAAAM,EAAA,GAAAD,YAAA,CAAAhE,MAAA,EAAAiE,EAAA;IAAAH,KAAA;EAAA;EAU7C;EACA,IAAMK,QAAQ,GAAG,EAAE;EAAA,IAAAC,MAAA,YAAAA,OAAA,EACsB;IAApC,IAAMzB,IAAI,GAAA0B,aAAA,CAAAC,GAAA;IACb,IAAI5E,SAAS,CAACmE,OAAO,CAAClB,IAAI,CAAC,CAAC,EAAE;MAC5BwB,QAAQ,CAACI,IAAI,CAACV,OAAO,CAAClB,IAAI,CAAC,CAACvC,IAAI,CAAC,UAACoE,MAAM,EAAK;QAC3CX,OAAO,CAAClB,IAAI,CAAC,GAAG6B,MAAM;MACxB,CAAC,CAAC,CAAC;IACL;EACF,CAAC;EAND,SAAAF,GAAA,MAAAD,aAAA,GAAmBf,MAAM,CAACC,IAAI,CAACM,OAAO,CAAC,EAAAS,GAAA,GAAAD,aAAA,CAAArE,MAAA,EAAAsE,GAAA;IAAAF,MAAA;EAAA;EAOvC,IAAID,QAAQ,CAACnE,MAAM,GAAG,CAAC,EAAE;IACvB,OAAOyE,OAAO,CAACC,GAAG,CAACP,QAAQ,CAAC,CAAC/D,IAAI,CAAC;MAAA,OAAMyD,OAAO;IAAA,EAAC;EAClD;EACA,OAAOA,OAAO;AAChB"}

package/modules/xlsx/parseSpreadsheetInfo.js

@@ -1,42 +1,72 @@
-function _createForOfIteratorHelperLoose(o, allowArrayLike) { var it = typeof Symbol !== "undefined" && o[Symbol.iterator] || o["@@iterator"]; if (it) return (it = it.call(o)).next.bind(it); if (Array.isArray(o) || (it = _unsupportedIterableToArray(o)) || allowArrayLike && o && typeof o.length === "number") { if (it) o = it; var i = 0; return function () { if (i >= o.length) return { done: true }; return { done: false, value: o[i++] }; }; } throw new TypeError("Invalid attempt to iterate non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method."); }
-function _unsupportedIterableToArray(o, minLen) { if (!o) return; if (typeof o === "string") return _arrayLikeToArray(o, minLen); var n = Object.prototype.toString.call(o).slice(8, -1); if (n === "Object" && o.constructor) n = o.constructor.name; if (n === "Map" || n === "Set") return Array.from(o); if (n === "Arguments" || /^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(n)) return _arrayLikeToArray(o, minLen); }
-function _arrayLikeToArray(arr, len) { if (len == null || len > arr.length) len = arr.length; for (var i = 0, arr2 = new Array(len); i < len; i++) arr2[i] = arr[i]; return arr2; }
-import { getWorkbookProperties, getSheets } from '../xml/xlsx.js';
-
-// I guess `xl/workbook.xml` file should always be present inside the *.xlsx archive.
-export default function parseSpreadsheetInfo(content, xml) {
-  var book = xml.createDocument(content);
-
-  // Read `<workbookPr/>` element to detect whether dates are 1900-based or 1904-based.
-  // https://support.microsoft.com/en-gb/help/214330/differences-between-the-1900-and-the-1904-date-system-in-excel
-  // http://webapp.docx4java.org/OnlineDemo/ecma376/SpreadsheetML/workbookPr.html
-  var workbookProperties = getWorkbookProperties(book);
-  var epoch1904 = Boolean(workbookProperties) && workbookProperties.getAttribute('date1904') === '1';
-
-  // Example of `<sheets/>` element:
-  //
-  // <sheets>
-  //   <sheet
-  //     xmlns:ns="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
-  //     name="Sheet1"
-  //     sheetId="1"
-  //     ns:id="rId3"/>
-  // </sheets>
-
-  var sheets = [];
-  for (var _iterator = _createForOfIteratorHelperLoose(getSheets(book)), _step; !(_step = _iterator()).done;) {
-    var sheet = _step.value;
-    if (sheet.getAttribute('name')) {
-      sheets.push({
-        id: sheet.getAttribute('sheetId'),
-        name: sheet.getAttribute('name'),
-        relationId: sheet.getAttribute('r:id')
+/**
+ * Parses spreadsheet info.
+ * @param {string} content
+ * @param  {function} parseXmlStream — SAX XML parser.
+ * @returns {object} An object of shape `{ epoch1904: boolean, sheets: Sheet[] }`.
+ */
+export default function parseSpreadsheetInfo(content, parseXmlStream) {
+  return parseXmlStream(content, {
+    createInitialState: createInitialStateInWorkbookXml,
+    onOpenTag: onOpenTagInWorkbookXml
+  }).then(function (state) {
+    return {
+      epoch1904: state.workbookPr ? state.workbookPr.epoch1904 : false,
+      sheets: state.sheets
+    };
+  });
+}
+function createInitialStateInWorkbookXml() {
+  return {
+    workbookPr: undefined,
+    sheets: []
+  };
+}
+function onOpenTagInWorkbookXml(tagName, attributes, state) {
+  if (tagName === 'workbookPr') {
+    // The official Open XML specification dictates that the top-level `<workbook/>` element
+    // should only contain exactly one `<workbookPr/>` element. In case of multiple ones,
+    // only the first one is being read and the rest are simply ignored.
+    //
+    // There's a `1904-based-dates.xlsx` test case that happens to include multiple `<workbookPr/>`
+    // elements by accident, most likely because I edited its `.xml` files by hand and didn't notice
+    // the already-existing `<workbookPr/>` tag.
+    //
+    // At first I thought about just fixing that one `.xlsx` test file but later
+    // I saw that another library called `SheetJS` specifically handles such cases
+    // by only reading the first `<workbookPr/>` element and ignoring the rest,
+    // so I though that perhaps it would make some sense to do same thing here.
+    // https://github.com/SheetJS/sheetjs/issues/1619
+    //
+    if (!state.workbookPr) {
+      // Read `<workbookPr/>` element attributes to detect whether dates are 1900-based or 1904-based.
+      // https://support.microsoft.com/en-gb/help/214330/differences-between-the-1900-and-the-1904-date-system-in-excel
+      // http://webapp.docx4java.org/OnlineDemo/ecma376/SpreadsheetML/workbookPr.html
+      state.workbookPr = {
+        epoch1904: attributes.date1904 === '1'
+      };
+    }
+  } else if (tagName === 'sheet') {
+    // Example of `<sheets/>` element:
+    // ```
+    // <sheets>
+    //   <sheet
+    //     xmlns:ns="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
+    //     name="Sheet1"
+    //     sheetId="1"
+    //     ns:id="rId3"/>
+    // </sheets>
+    // ```
+    if (attributes.name) {
+      state.sheets.push({
+        // `sheetId` attribute value is an arbitrary, `1`-based unique positive integer
+        // assigned to a worksheet, typically starting at `1` for the first sheet.
+        //  Deleting and adding new sheets might cause the sheetId values to become non-sequential.
+        // For example, `sheetId`s could be `1`, `2`, `4`, if sheet `3` was deleted.
+        id: Number(attributes.sheetId),
+        name: attributes.name,
+        relationId: attributes['r:id']
       });
     }
   }
-  return {
-    epoch1904: epoch1904,
-    sheets: sheets
-  };
 }
 //# sourceMappingURL=parseSpreadsheetInfo.js.map

package/modules/xlsx/parseSpreadsheetInfo.js.map

@@ -1 +1 @@
-{"version":3,"file":"parseSpreadsheetInfo.js","names":["getWorkbookProperties","getSheets","parseSpreadsheetInfo","content","xml","book","createDocument","workbookProperties","epoch1904","Boolean","getAttribute","sheets","_iterator","_createForOfIteratorHelperLoose","_step","done","sheet","value","push","id","name","relationId"],"sources":["../../source/xlsx/parseSpreadsheetInfo.js"],"sourcesContent":["import {\r\n  getWorkbookProperties,\r\n  getSheets\r\n} from '../xml/xlsx.js'\r\n\r\n// I guess `xl/workbook.xml` file should always be present inside the *.xlsx archive.\r\nexport default function parseSpreadsheetInfo(content, xml) {\r\n  const book = xml.createDocument(content)\r\n\r\n  // Read `<workbookPr/>` element to detect whether dates are 1900-based or 1904-based.\r\n  // https://support.microsoft.com/en-gb/help/214330/differences-between-the-1900-and-the-1904-date-system-in-excel\r\n  // http://webapp.docx4java.org/OnlineDemo/ecma376/SpreadsheetML/workbookPr.html\r\n  const workbookProperties = getWorkbookProperties(book)\r\n  const epoch1904 = Boolean(workbookProperties) && workbookProperties.getAttribute('date1904') === '1'\r\n\r\n  // Example of `<sheets/>` element:\r\n  //\r\n  // <sheets>\r\n  //   <sheet\r\n  //     xmlns:ns=\"http://schemas.openxmlformats.org/officeDocument/2006/relationships\"\r\n  //     name=\"Sheet1\"\r\n  //     sheetId=\"1\"\r\n  //     ns:id=\"rId3\"/>\r\n  // </sheets>\r\n\r\n  const sheets = []\r\n  for (const sheet of getSheets(book)) {\r\n    if (sheet.getAttribute('name')) {\r\n      sheets.push({\r\n        id: sheet.getAttribute('sheetId'),\r\n        name: sheet.getAttribute('name'),\r\n        relationId: sheet.getAttribute('r:id')\r\n      })\r\n    }\r\n  }\r\n\r\n  return {\r\n    epoch1904,\r\n    sheets\r\n  }\r\n}"],"mappings":";;;AAAA,SACEA,qBAAqB,EACrBC,SAAS,QACJ,gBAAgB;;AAEvB;AACA,eAAe,SAASC,oBAAoBA,CAACC,OAAO,EAAEC,GAAG,EAAE;EACzD,IAAMC,IAAI,GAAGD,GAAG,CAACE,cAAc,CAACH,OAAO,CAAC;;EAExC;EACA;EACA;EACA,IAAMI,kBAAkB,GAAGP,qBAAqB,CAACK,IAAI,CAAC;EACtD,IAAMG,SAAS,GAAGC,OAAO,CAACF,kBAAkB,CAAC,IAAIA,kBAAkB,CAACG,YAAY,CAAC,UAAU,CAAC,KAAK,GAAG;;EAEpG;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA,IAAMC,MAAM,GAAG,EAAE;EACjB,SAAAC,SAAA,GAAAC,+BAAA,CAAoBZ,SAAS,CAACI,IAAI,CAAC,GAAAS,KAAA,IAAAA,KAAA,GAAAF,SAAA,IAAAG,IAAA,GAAE;IAAA,IAA1BC,KAAK,GAAAF,KAAA,CAAAG,KAAA;IACd,IAAID,KAAK,CAACN,YAAY,CAAC,MAAM,CAAC,EAAE;MAC9BC,MAAM,CAACO,IAAI,CAAC;QACVC,EAAE,EAAEH,KAAK,CAACN,YAAY,CAAC,SAAS,CAAC;QACjCU,IAAI,EAAEJ,KAAK,CAACN,YAAY,CAAC,MAAM,CAAC;QAChCW,UAAU,EAAEL,KAAK,CAACN,YAAY,CAAC,MAAM;MACvC,CAAC,CAAC;IACJ;EACF;EAEA,OAAO;IACLF,SAAS,EAATA,SAAS;IACTG,MAAM,EAANA;EACF,CAAC;AACH"}
+{"version":3,"file":"parseSpreadsheetInfo.js","names":["parseSpreadsheetInfo","content","parseXmlStream","createInitialState","createInitialStateInWorkbookXml","onOpenTag","onOpenTagInWorkbookXml","then","state","epoch1904","workbookPr","sheets","undefined","tagName","attributes","date1904","name","push","id","Number","sheetId","relationId"],"sources":["../../source/xlsx/parseSpreadsheetInfo.js"],"sourcesContent":["/**\r\n * Parses spreadsheet info.\r\n * @param {string} content\r\n * @param  {function} parseXmlStream — SAX XML parser.\r\n * @returns {object} An object of shape `{ epoch1904: boolean, sheets: Sheet[] }`.\r\n */\r\nexport default function parseSpreadsheetInfo(content, parseXmlStream) {\r\n  return parseXmlStream(content, {\r\n    createInitialState: createInitialStateInWorkbookXml,\r\n    onOpenTag: onOpenTagInWorkbookXml\r\n  }).then((state) => {\r\n    return {\r\n      epoch1904: state.workbookPr ? state.workbookPr.epoch1904 : false,\r\n      sheets: state.sheets\r\n    }\r\n  })\r\n}\r\n\r\nfunction createInitialStateInWorkbookXml() {\r\n  return {\r\n    workbookPr: undefined,\r\n    sheets: []\r\n  }\r\n}\r\n\r\nfunction onOpenTagInWorkbookXml(tagName, attributes, state) {\r\n  if (tagName === 'workbookPr') {\r\n    // The official Open XML specification dictates that the top-level `<workbook/>` element\r\n    // should only contain exactly one `<workbookPr/>` element. In case of multiple ones,\r\n    // only the first one is being read and the rest are simply ignored.\r\n    //\r\n    // There's a `1904-based-dates.xlsx` test case that happens to include multiple `<workbookPr/>`\r\n    // elements by accident, most likely because I edited its `.xml` files by hand and didn't notice\r\n    // the already-existing `<workbookPr/>` tag.\r\n    //\r\n    // At first I thought about just fixing that one `.xlsx` test file but later\r\n    // I saw that another library called `SheetJS` specifically handles such cases\r\n    // by only reading the first `<workbookPr/>` element and ignoring the rest,\r\n    // so I though that perhaps it would make some sense to do same thing here.\r\n    // https://github.com/SheetJS/sheetjs/issues/1619\r\n    //\r\n    if (!state.workbookPr) {\r\n      // Read `<workbookPr/>` element attributes to detect whether dates are 1900-based or 1904-based.\r\n      // https://support.microsoft.com/en-gb/help/214330/differences-between-the-1900-and-the-1904-date-system-in-excel\r\n      // http://webapp.docx4java.org/OnlineDemo/ecma376/SpreadsheetML/workbookPr.html\r\n      state.workbookPr = {\r\n        epoch1904: attributes.date1904 === '1'\r\n      }\r\n    }\r\n  } else if (tagName === 'sheet') {\r\n    // Example of `<sheets/>` element:\r\n    // ```\r\n    // <sheets>\r\n    //   <sheet\r\n    //     xmlns:ns=\"http://schemas.openxmlformats.org/officeDocument/2006/relationships\"\r\n    //     name=\"Sheet1\"\r\n    //     sheetId=\"1\"\r\n    //     ns:id=\"rId3\"/>\r\n    // </sheets>\r\n    // ```\r\n    if (attributes.name) {\r\n      state.sheets.push({\r\n        // `sheetId` attribute value is an arbitrary, `1`-based unique positive integer\r\n        // assigned to a worksheet, typically starting at `1` for the first sheet.\r\n        //  Deleting and adding new sheets might cause the sheetId values to become non-sequential.\r\n        // For example, `sheetId`s could be `1`, `2`, `4`, if sheet `3` was deleted.\r\n        id: Number(attributes.sheetId),\r\n        name: attributes.name,\r\n        relationId: attributes['r:id']\r\n      })\r\n    }\r\n  }\r\n}"],"mappings":"AAAA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASA,oBAAoBA,CAACC,OAAO,EAAEC,cAAc,EAAE;EACpE,OAAOA,cAAc,CAACD,OAAO,EAAE;IAC7BE,kBAAkB,EAAEC,+BAA+B;IACnDC,SAAS,EAAEC;EACb,CAAC,CAAC,CAACC,IAAI,CAAC,UAACC,KAAK,EAAK;IACjB,OAAO;MACLC,SAAS,EAAED,KAAK,CAACE,UAAU,GAAGF,KAAK,CAACE,UAAU,CAACD,SAAS,GAAG,KAAK;MAChEE,MAAM,EAAEH,KAAK,CAACG;IAChB,CAAC;EACH,CAAC,CAAC;AACJ;AAEA,SAASP,+BAA+BA,CAAA,EAAG;EACzC,OAAO;IACLM,UAAU,EAAEE,SAAS;IACrBD,MAAM,EAAE;EACV,CAAC;AACH;AAEA,SAASL,sBAAsBA,CAACO,OAAO,EAAEC,UAAU,EAAEN,KAAK,EAAE;EAC1D,IAAIK,OAAO,KAAK,YAAY,EAAE;IAC5B;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA,IAAI,CAACL,KAAK,CAACE,UAAU,EAAE;MACrB;MACA;MACA;MACAF,KAAK,CAACE,UAAU,GAAG;QACjBD,SAAS,EAAEK,UAAU,CAACC,QAAQ,KAAK;MACrC,CAAC;IACH;EACF,CAAC,MAAM,IAAIF,OAAO,KAAK,OAAO,EAAE;IAC9B;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA,IAAIC,UAAU,CAACE,IAAI,EAAE;MACnBR,KAAK,CAACG,MAAM,CAACM,IAAI,CAAC;QAChB;QACA;QACA;QACA;QACAC,EAAE,EAAEC,MAAM,CAACL,UAAU,CAACM,OAAO,CAAC;QAC9BJ,IAAI,EAAEF,UAAU,CAACE,IAAI;QACrBK,UAAU,EAAEP,UAAU,CAAC,MAAM;MAC/B,CAAC,CAAC;IACJ;EACF;AACF"}