C#中如何将字符串转换为SEPA格式？第三方工具集成问询

Great question! When integrating with SEPA-compliant tools, you don’t have to manually code every character replacement from the EPC conversion table. There’s a more efficient way to normalize and clean strings to meet SEPA’s strict character requirements. Let’s break down the approach:

Key Background: SEPA Allowed Characters #

SEPA only accepts a limited ASCII charset:

• Uppercase/lowercase letters (A-Z, a-z)
• Digits (0-9)
• Specific symbols: ', -, ?, :, (, ), /, \, ,, and spaces

Any other characters (like HTML entities, Unicode smart quotes, accented letters) need to be converted or removed.

Step-by-Step C# Implementation #

Here’s a reusable helper method that handles HTML entity decoding, Unicode normalization, and character filtering in one go:

using System;
using System.Globalization;
using System.Text;
using System.Text.RegularExpressions;
using System.Net; // For WebUtility in .NET Core/.NET 5+

public static class SepaStringConverter
{
    // Regex to catch any characters NOT in the SEPA allowed set
    private static readonly Regex _invalidSepaChars = new Regex(@"[^A-Za-z0-9'\-?:()/\\, ]", RegexOptions.Compiled);

    public static string ToSepaCompliantString(string input)
    {
        if (string.IsNullOrWhiteSpace(input))
            return input?.Trim() ?? string.Empty;

        // Step 1: Decode HTML entities (e.g., ’ → ')
        string decodedInput = WebUtility.HtmlDecode(input);

        // Step 2: Normalize Unicode characters to ASCII equivalents
        // FormKD breaks down compatibility characters (like smart apostrophes, accented letters)
        string normalized = decodedInput.Normalize(NormalizationForm.FormKD);
        StringBuilder cleanedBuilder = new StringBuilder();

        foreach (char c in normalized)
        {
            UnicodeCategory charCategory = CharUnicodeInfo.GetUnicodeCategory(c);
            
            // Keep allowed characters: letters, digits, approved symbols
            if (charCategory is UnicodeCategory.UppercaseLetter or UnicodeCategory.LowercaseLetter or UnicodeCategory.DecimalDigitNumber
                || "'-?:()/\\, ".Contains(c))
            {
                cleanedBuilder.Append(c);
            }
            // Replace non-breaking spaces with regular spaces
            else if (c == '\u00A0')
            {
                cleanedBuilder.Append(' ');
            }
            // All other characters are discarded (matches SEPA's requirement to remove invalid chars)
        }

        // Step 3: Final cleanup - remove any remaining invalid chars and normalize spaces
        string finalString = _invalidSepaChars.Replace(cleanedBuilder.ToString(), string.Empty);
        finalString = Regex.Replace(finalString, @"\s+", " ").Trim();

        return finalString;
    }
}

How It Works #

1. HTML Entity Decoding:
   Uses WebUtility.HtmlDecode to convert entities like ’ to their actual Unicode characters (in your example, O’CONNOR becomes O’CONNOR).

2. Unicode Normalization:
   NormalizationForm.FormKD decomposes complex Unicode characters into their basic ASCII equivalents. For example:

   • Smart apostrophes ’ → regular apostrophes '
   • Accented letters like é → e (since SEPA doesn’t allow accented characters)

3. Character Filtering:
   Explicitly keeps only SEPA-approved characters, discards everything else, and cleans up extra spaces for consistency.

Usage with EPPlus #

When writing to your XLSX file, just pass your raw string through the helper method first:

using OfficeOpenXml;

// Initialize your EPPlus package and worksheet
using (var package = new ExcelPackage(new FileInfo("sepa-data.xlsx")))
{
    var worksheet = package.Workbook.Worksheets.Add("SEPA Transactions");
    
    // Convert customer name to SEPA format before writing
    string rawCustomerName = "O’CONNOR";
    string sepaCompliantName = SepaStringConverter.ToSepaCompliantString(rawCustomerName);
    
    worksheet.Cells["A1"].Value = sepaCompliantName; // Writes "O'CONNOR"
    
    package.Save();
}

Customization Tips #

If you need to handle edge cases from the EPC conversion table (e.g., specific symbol mappings), you can add a manual replacement step before normalization:

// Add this before normalization in the helper method
string mappedInput = decodedInput
    .Replace("`", "'") // Map backticks to apostrophes
    .Replace("\"", string.Empty); // Remove double quotes (not allowed in SEPA)

内容的提问来源于stack exchange，提问作者korulis