Slow Performance when there are lots of merge cells & calculations

[nathj07]

Description

I am using this library to read text out of XLSX files, and generally it works incredibly well. Typically the performance is fantastic, well, well under 1 second. However, I have on file, which sadly I cannot share, where the time to extract the text is ~10minutes. What I have been able to do isprofile that file and see that the issue is in mergeCellsParser. The exact sheet in question does have merged cells, among other things.

Steps to reproduce the issue:

Using a file I can share you can still see the issue. The file,
merge_test.xlsx, has two columns with ~5,000 rows each, but these columns are merged A-B, and merged C-D.

This file takes 1m50s to extract text from.

A snippet of the code I am using, though not the full code:

    f, err := excelize.OpenFile(xlsxFile, excelOpts)
	if err != nil {
		return nil, fmt.Errorf("excelize failed to open file %s: %w", xlsxFile, err)
	}
	defer func() {
		// Close the spreadsheet.
		if err := f.Close(); err != nil {
			logger.WithError(err).Error("failed to close XLSX file")
		}
	}()

	// Get extras that we need, comments first
	comments, err := f.GetComments()
	if err != nil {
		return nil, fmt.Errorf("failed to GetComments: %w", err)
	}

	sheetRels, err := msx.findRels(arc, ExcelRelsXML, "xl/")
	if err != nil {
		return nil, err
	}

	// Now we can process the main data from the file using excelize
	for i, sheet := range f.GetSheetList() {
		rows, err := f.GetRows(sheet, excelOpts)
		if err != nil {
			return nil, fmt.Errorf("failed to GetRows for %s: %w", xlsxFile, err)
		}
		if _, err := sb.WriteString(sheet + "\n"); err != nil {
			return nil, fmt.Errorf("failed to write the sheet title %q: %s", sheet, err)
		}
		for j, row := range rows {
			for k, colCell := range row {
				txt := colCell
				// cellName e.g. "A1" is needed to get cell hyperlinks.
				cellName, err := excelize.CoordinatesToCellName(k+1, j+1) // this is not 0 based
				if err != nil {
					return nil, fmt.Errorf("failed to get cell name for %d-%d: %w", k+1, j+1, err)
				}
				hasLink, link, err := f.GetCellHyperLink(sheet, cellName)
				if err != nil {
					// log this, it is not worth aborting the whole extraction for
					logger.WithError(err).WithField("cellName", cellName).Warn("failed to get link for cell")
				}
				if hasLink {
					txt = colCell + " " + link
				}
				if _, err := sb.WriteString(txt + " "); err != nil {
					return nil, fmt.Errorf("failed to write string to builder: %w", err)
				}
			}
		}

As I say go's profiling shows the issue is.

This profile was generated using the worst case scenario file that takes 10mins to read

Describe the results you received:
The text is extracted correctly, however it takes ~10minutes

Describe the results you expected:
I would expect it to be faster, even if it were longer than normal, I think under 30s is reasonable

Output of go version:

go version go1.19.4 darwin/amd64

Excelize version or commit ID:

github.com/xuri/excelize/v2 v2.7.0

Environment details (OS, Microsoft Excel™ version, physical, etc.):
This performance is particularly

[xuri]

This issue has been fixed, please try to upgrade master branch code. This patch will be released in the next version.