Can't Serialize And Normalize the Receipt

Hello Hony Mohamed,

You can do a bit refactor for your code and achieve,

public static string NormalizeText(string text)
{
    if (string.IsNullOrWhiteSpace(text))
        return string.Empty;

    return Regex.Replace(text, @"[:]", "")
                .ApplyRegex(@"[a-z]", m => m.Value.ToUpper())
                .ApplyRegex(@"[\u0600-\u06FF]+", m => m.Value)
                .ApplyRegex(@"[\{\}\[\]]", "")
                .ApplyRegex(@"\s+", "");
}

private static string ApplyRegex(this string input, string pattern, MatchEvaluator evaluator)
{
    return Regex.Replace(input, pattern, evaluator);
}

private static string ApplyRegex(this string input, string pattern, string replacement)
{
    return Regex.Replace(input, pattern, replacement);
}

Dear friends, thank you all. Unfortunately, I did not get what I wanted. All the answers gave the following result

"HEADER""DATETIMEISSUED""""2024""-""12""-30T173749""RECEIPTNUMBER""HAZ002""UUID""""PREVIOUSUUID""""REFERENCEOLDUUID""""CURRENCY""EGP""EXCHANGERATE"""0"""DOCUMENTTYPE""RECEIPTTYPE""S""TYPEVERSION""""1.2""""SELLER""RIN""""332460843""""COMPANYTRADENAME""???? ???? ???? ???""BRANCHCODE""ABC""DEVICESERIALNUMBER""CWX02618""ACTIVITYCODE""""8620""""BRANCHADDRESS""COUNTRY""EG""GOVERNATE""CAIRO""REGIONCITY""CITYCENTER""STREET""16STREET""BUILDINGNUMBER""14BN""BUYER""TYPE""F""ID""""313717919""""NAME""TAXPAYER1""MOBILENUMBER""+""201020567462""""PAYMENTNUMBER""""987654""""ITEMDATA""INTERNALCODE""NA""DESCRIPTION""MEDICALCAREFORTHEELDERLY""ITEMTYPE""EGS""ITEMCODE""EG-""332460843""-M3""UNITTYPE""EA""QUANTITY"""1"""UNITPRICE"""200"""NETSALE"""200"""TOTALSALE"""200"""TOTAL"""200"""COMMERCIALDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""ITEMDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""TAXABLEITEMS""TAXTYPE""T1""AMOUNT"""28"""SUBTYPE""V009""RATE"""14"""TOTALSALES"""200"""TOTALCOMMERCIALDISCOUNT"""0"""TOTALITEMSDISCOUNT"""0"""EXTRARECEIPTDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""NETAMOUNT"""200"""FEESAMOUNT"""0"""TOTALAMOUNT"""228"""TAXTOTALS""TAXTYPE""T1""AMOUNT"""28"""PAYMENTMETHOD""C""ADJUSTMENT"""0"""CONTRACTOR""NAME""CONTRACTOR1""AMOUNT"""0"""RATE"""0""

Steps to Resolve:

1. Serialize Without Modifications to Numerical Fields: Ensure numbers are preserved correctly in their serialized form, including their quotes when required.

2. Normalize Text Without Breaking JSON Syntax: Adjust NormalizeText to only normalize keys and string values without affecting numerical fields.

Updated Code:

public static string SerializeAndNormalizeReceipt(object receipt)
{
    if (receipt == null)
        return string.Empty;
    string serializedData = JsonConvert.SerializeObject(receipt, Formatting.None);
    string normalizedData = NormalizeJson(serializedData);
    return normalizedData;
}

public static string NormalizeJson(string json)
{
    if (string.IsNullOrEmpty(json))
        return string.Empty;
    JObject parsedJson = JObject.Parse(json);
    NormalizeJToken(parsedJson);
    return parsedJson.ToString(Formatting.None);
}

private static void NormalizeJToken(JToken token)
{
    if (token is JObject obj)
    {
        foreach (var property in obj.Properties().ToList())
        {
            string normalizedKey = NormalizeText(property.Name);
            property.Replace(new JProperty(normalizedKey, property.Value));
            NormalizeJToken(property.Value);
        }
    }
    else if (token is JArray array)
    {
        foreach (var item in array)
        {
            NormalizeJToken(item);
        }
    }
    else if (token is JValue value)
    {
        if (value.Type == JTokenType.String)
        {
            value.Value = NormalizeText(value.ToString());
        }
    }
}

public static string NormalizeText(string text)
{
    if (string.IsNullOrEmpty(text))
        return string.Empty;
    string cleanedText = Regex.Replace(text, @"[:\,]", "");
    string normalizedText = Regex.Replace(cleanedText, @"[a-z]", m => m.Value.ToUpper());
    normalizedText = Regex.Replace(normalizedText, @"[{}[]]", "");
    normalizedText = Regex.Replace(normalizedText, @"\s+", "");
    return normalizedText;
}

Hello Hany,

Can you share the simple text which you are passing as parameter to NormalizeText method?

Try below method.

public static string NormalizeText(string text)
{
    if (string.IsNullOrEmpty(text))
        return string.Empty;

    // Remove colons and commas
    string cleanedText = Regex.Replace(text, @"[:\,]", "");

    // Convert lowercase letters to uppercase
    string normalizedText = Regex.Replace(cleanedText, @"[a-z]", m => m.Value.ToUpper());

    // Retain Arabic characters
    normalizedText = Regex.Replace(normalizedText, @"[\u0600-\u06FF]+", m => m.Value);

    // Remove curly braces and square brackets
    normalizedText = Regex.Replace(normalizedText, @"[\{\}\[\]]", "");

    // Remove spaces before alphanumeric characters
    normalizedText = Regex.Replace(normalizedText, @"\s+(?=[a-zA-Z0-9])", "");

    // Ensure numerical fields retain their quotation marks correctly
    normalizedText = Regex.Replace(normalizedText, @"(?<=\D)(\d+)(?=\D)", "\"$1\"");

    return normalizedText;
}

Thanks

Dear Mohamed this is the output 

"HEADER""DATETIMEISSUED""""2024""-""12""-30T173749""RECEIPTNUMBER""HAZ002""UUID""""PREVIOUSUUID""""REFERENCEOLDUUID""""CURRENCY""EGP""EXCHANGERATE"""0"""DOCUMENTTYPE""RECEIPTTYPE""S""TYPEVERSION""""1.2""""SELLER""RIN""""332460843""""COMPANYTRADENAME""???? ???? ???? ???""BRANCHCODE""ABC""DEVICESERIALNUMBER""CWX02618""ACTIVITYCODE""""8620""""BRANCHADDRESS""COUNTRY""EG""GOVERNATE""CAIRO""REGIONCITY""CITYCENTER""STREET""16STREET""BUILDINGNUMBER""14BN""BUYER""TYPE""F""ID""""313717919""""NAME""TAXPAYER1""MOBILENUMBER""+""201020567462""""PAYMENTNUMBER""""987654""""ITEMDATA""INTERNALCODE""NA""DESCRIPTION""MEDICALCAREFORTHEELDERLY""ITEMTYPE""EGS""ITEMCODE""EG-""332460843""-M3""UNITTYPE""EA""QUANTITY"""1"""UNITPRICE"""200"""NETSALE"""200"""TOTALSALE"""200"""TOTAL"""200"""COMMERCIALDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""ITEMDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""TAXABLEITEMS""TAXTYPE""T1""AMOUNT"""28"""SUBTYPE""V009""RATE"""14"""TOTALSALES"""200"""TOTALCOMMERCIALDISCOUNT"""0"""TOTALITEMSDISCOUNT"""0"""EXTRARECEIPTDISCOUNTDATA""AMOUNT"""0"""DESCRIPTION""NODISCOUNT""RATE"""0"""NETAMOUNT"""200"""FEESAMOUNT"""0"""TOTALAMOUNT"""228"""TAXTOTALS""TAXTYPE""T1""AMOUNT"""28"""PAYMENTMETHOD""C""ADJUSTMENT"""0"""CONTRACTOR""NAME""CONTRACTOR1""AMOUNT"""0"""RATE"""0"" as you see this is not right

Hello Hany,

You need to add quotation marks around numbers with an adittional Regex.Replace method. Here is an updated version of NormalizeText method:

public static string NormalizeText(string text)
{
	if (string.IsNullOrEmpty(text))
	 return string.Empty;

	string cleanedText = Regex.Replace(text, @"[:\,]", "");
	string normalizedText = Regex.Replace(cleanedText, @"[a-z]", m => m.Value.ToUpper());
	normalizedText = Regex.Replace(normalizedText, @"[\u0600-\u06FF]+", m => m.Value);
	normalizedText = Regex.Replace(normalizedText, @"[\{\}\[\]]", "");
	normalizedText = Regex.Replace(normalizedText, @"\s+(?=[a-zA-Z0-9])", "");
	normalizedText = Regex.Replace(normalizedText, @"\b\d+(\.\d+)?\b", "\"$0\"");

	return normalizedText;
}

This matches any integer or decimal number (\b\d+(\.\d+)?\b) and wraps it in quotation marks.

Thank you!