Slower rows iterator on server than local machine

[a-yahaha]

本地测试代码

测试代码如下：

func main() {
	fmt.Println("hello")
	readExcel("/Users/bytedance/Downloads/excel/测试数据100列4500行.xlsx")
}
// 模拟excel读取数据
func readExcel(path string) {
    options := excelize.Options{
       RawCellValue:     false,
       ShortDatePattern: "yyyy-mm-dd",
       LongTimePattern:  "HH:mm:ss",
       LongDatePattern:  "yyyy-mm-dd HH:mm:ss",
       CultureInfo:      excelize.CultureNameZhCN,
    }
    excel, err := excelize.OpenFile(path, options)
    if err != nil {
       fmt.Println("open excel failed", err)
       return
    }
    totalRows, _ := GetRowsCountIterator(excel, "Sheet1")
    if totalRows == 0 {
       fmt.Println("GetRowsCountIterator is empty")
       return
    }
}
func GetRowsCountIterator(excel *excelize.File, sheetName string) (int64, error) {
	// 获取时间才 1.1s
	startTime := time.Now()
	defer func() {
		if err := excel.Close(); err != nil {
			fmt.Printf("excel GetRowsCount file close error:%+v", err)
		}
		fmt.Printf("GetRowsCountIterator cost: %v \n", time.Since(startTime))
	}()
	rows, err := excel.Rows(sheetName)
	if err != nil {
		fmt.Println("GetRowsCountIterator failed", err)
		return 0, err
	}
	var totalRows int64
	for rows.Next() {
		row, err := rows.Columns(excelize.Options{
			RawCellValue: false,
		})
		if err != nil {
			break
		}
		if len(row) > 0 {
			totalRows++
		}
	}
	if err = rows.Close(); err != nil {
		fmt.Printf("excel GetRowsCountIterator rows close  error:%+v", err)
		return 0, err
	}
	fmt.Printf("GetRowsCountIterator total: %d \n", totalRows)
	return totalRows, nil
}

耗时

GetRowsCountIterator cost: 1.156624625s 

服务器上测试

代码如下

// GetRowsCount retrieves the total rows count from the excelize iterator.
func GetRowsCount(f *excelize.File, sheetName string) (int64, error) {
	defer func() {
		if err := f.Close(); err != nil {
			fmt.Printf("excel GetRowsCount file close error:%+v", err)
		}
	}()
	var totalRows int64
	startTime := time.Now()
	rows, err := f.Rows(sheetName)
	fmt.Printf("[costTime]excel GetRowsCount Rows cost: %v", time.Since(startTime))
	if err != nil {
		fmt.Printf("excel GetRowsCount getRows error:%+v", err)
		return 0, err
	}
	startTime = time.Now()
	for rows.Next() {
		row, err := rows.Columns(excelize.Options{
			RawCellValue: false,
		})
		if err != nil {
			break
		}
		if len(row) > 0 {
			totalRows++
		}
	}
	fmt.Printf("[costTime]excel GetRowsCount Rows.next cost: %v", time.Since(startTime))
	startTime = time.Now()
	if err = rows.Close(); err != nil {
		fmt.Printf("excel GetRowsCount rows close  error:%+v", err)
		return 0, err
	}
	fmt.Printf("[costTime]excel GetRowsCount Rows.close cost: %v", time.Since(startTime))
	return totalRows, nil
}

耗时

[costTime]excel GetRowsCount Rows.next cost: 6.809400376s

服务器的配置 8G 16G

问题

初步排查是 rows.Columns在服务端的性能很差，基本耗时都在这里。

有几个疑问

1. 是否只能通过行迭代器获取excel中sheet的行总数
2. Columns的性能为什么有这么大的差异，是否有其他优化的方法

测试文件

测试文件： 100列4500行文本

测试数据100列4500行.xlsx

[xuri]

Thanks for your issue. Is your local machine environment different from your server? The performance is also impacted by these factors: OS, Go language version, Excelize library version, hardware such as CPU, memory, storage equipment (HDD or SSD), etc. I don't have enough information to reproduce this issue, but here are some performance figures for your reference: Performance Figures. I'll close this issue, if you have any questions, please let me know to reopen this anytime.