Tuesday, October 13, 2020

C# .NET How to get integer from hexadecimal string, many hex formats supported















There are many hex formats in which the hex prefixes must be removed, such as "0x" in order for C# TryParse function with NumberStyles.HexNumber to work. 


UInt32.TryParse("0x20", 
             NumberStyles.HexNumber, // I've tried also AllowHexSpecifier
             CultureInfo.InvariantCulture,  // I've also tried CurrentCulture
             out number));


The code below removes many hex formats prefixes as listed here - https://en.wikipedia.org/wiki/Hexadecimal

As per feedback, I improved the algo to first quarantine likely regex hex matches with prefixes then run another regex to extract the hex within that match.



Source Code
using System;
using System.Globalization;
using System.Text.RegularExpressions;

public static class Program
{
	const string strRegGetHexNumber = @"\\u[0-9a-f]{2,}|0x[0-9a-f]{2,}|%[0-9a-f]{2,}|\u0023([0-9a-f]){1,6}|&\u0023x([0-9a-f]){1,6};|\s[0-9a-f]{2,}|[0-9a-f]{2,}|[«‹»›„‚“‟‘‛”’""""❛❜❝❞〝〞〟"""""'‘][0-9a-f]{2,}[’'""""«‹»›„‚“‟‘‛”’""""❛❜❝❞〝〞〟"]";
	private static readonly Regex rgxGetHex = new Regex(strRegGetHexNumber, RegexOptions.IgnoreCase | RegexOptions.CultureInvariant);
	public static void Main()
	{
		string unicodeText = "UTF-16 (hex)	0x0023 (0023)";
		//https://en.wikipedia.org/wiki/Hexadecimal --remove possible prefixes
		string prefixfree = string.Empty;
		string[] prefixHexs = new string[]{"0x", "\\u", "#", "&#", "\\x", "\\s", "U+", "X'", "16#", "#x", "#16r", "&H", "0h"};
		foreach (var pre in prefixHexs)
		{
			if (unicodeText.IndexOf(pre) > -1)
			{
				prefixfree = unicodeText.Substring(unicodeText.IndexOf(pre) + pre.Length);
				break;
			}
		}

		if (string.IsNullOrEmpty(prefixfree))
			prefixfree = unicodeText;
		Match m = rgxGetHex.Match(prefixfree);
		bool success = false;
		ulong number = 0;
		string hex_value = string.Empty;
		if (m.Success)
		{
			try
			{
				hex_value = m.Value;
				//long number = Convert.ToInt64(hex_value, 16); //base 16 - hex....
				//https://stackoverflow.com/questions/2801509/uint32-tryparse-hex-number-not-working -> remove prefixes
				success = ulong.TryParse(hex_value, NumberStyles.HexNumber, CultureInfo.InvariantCulture, out number);
			}
			catch (FormatException)
			{
				unicodeText = string.Format("{0} is not in the correct format for a hexadecimal number.", m.Value);
			}
			catch (OverflowException)
			{
				unicodeText = string.Format("{0} is outside the range of an Int64.", m.Value);
			}
			catch (ArgumentException)
			{
				unicodeText = string.Format("{0} is invalid in base 16.", m.Value);
			}
			catch (Exception ex)
			{
				unicodeText = string.Format("{0} return error\r\n{2}", m.Value, ex.Message);
			}
		}
		else
			unicodeText = "Could not find a hex number in \"" + unicodeText + "\". Select the hex number only.";
		if (success)
			unicodeText = string.Format("{0} from found {1:N0} hex number in string {2}", number, hex_value, unicodeText);
		else
			unicodeText = "Could not find a hex number in \"" + unicodeText + "\". Select the hex number only.";
		Console.WriteLine(unicodeText);
	}
}

No comments:

Post a Comment