Are there any guarantees that the workbookReader would return worksheets in the order they appear in the workbook? I ran a test and it seems like the order is not guaranteed. The total number of worksheets returned appear correct.

In addition, each worksheet returned doesn't appear to return the correct row data if I iterate through the rows and add rows to a new worksheet. If you add some sample rows to a few worksheets in the workbook and run this code, it should reproduce.

async function printWorkbook() {
    const outputWorkbook = new ExcelJS.Workbook();
    let combinedWorksheet = outputWorkbook.addWorksheet('Test worksheet');    
    const filePath = 'WORKBOOK_WITH_188_SHEETS.xlsx';
    const workbookReader = new ExcelJS.stream.xlsx.WorkbookReader(filePath, {
      worksheets: 'emit',
    });

    for await (const worksheet of workbookReader) {
    console.log('===> worksheet:', worksheet.name);
        for await (const row of worksheet) {
            combinedWorksheet.addRow(row.values).commit();
        }
    }
    await outputWorkbook.xlsx.writeFile('combined.xlsx');
}

printWorkbook().catch(console.error);

There are no guarantee that the worksheets are returned in the way they appear in the workbook. I believe the order is effectively set by how unzipper extracts the xlsx. I believe the order on this branch is the same as on master.

Also regarding the rows not return the correct data. I think that is not an issue introduced by my changes. I am able to replicate the same results on master as on my branch.