Support date formats with dot separator such as DD.MM.YYYY

[nicomni]

Is your feature request related to a problem? Please describe.
When using excelize to read an Excel workbook with dates formatted as DD.MM.YYYY it results in ..YYYY, e.g. 18.11.2023 is parsed as ..2023.

Describe the solution you'd like
Once the parser sees the dot ., it interprets this as a TokenTypeDecimal and discards the preceding datetime values, e.g. parsing MM. results in a . token.
I want to see an MM token of TokenTypeDatetime and a . of type TokenTypeLiteral being added to the tokens list.

Describe alternatives you've considered

Additional context

[xuri]

Thanks for your issue. I have tested number format code DD.MM.YYYY with the excelize master branch, and it works well. Please try to upgrade master branch code, if still have this problem, please let me know.

[nicomni]

Sorry for not adding more context. Let me rectify:

The problem occurs when the datetime is stored as a float in the specific cell rather than a shared string, and a custom format code is used. See this example:

package main

import (
	"fmt"
	"log"

	"github.com/xuri/excelize/v2"
)

func main() {
	f := excelize.NewFile()
	defer func() {
		if err := f.Close(); err != nil {
			log.Fatal(err)
		}
	}()
        activeSheetName := f.GetSheetName(f.GetActiveSheetIndex())
	f.SetCellValue(activeSheetName, "A1", 45186.0)
	exp := "DD.MM.YYYY"
	style, _ := f.NewStyle(&excelize.Style{CustomNumFmt: &exp})
	f.SetColStyle(activeSheetName, "A", style)
	val, _ := f.GetCellValue(activeSheetName, "A1")
	fmt.Println(val) // Getting '..2023'. Want '17.09.2023'
}

During execution, the circumstances lead to this line in the excelize codebase to be run, which ultimately lead to nfp.format() to be called.

I have addressed the issue in #8