Splitting CamelCase

Question

This is all asp.net c#.

I have an enum

public enum ControlSelectionType 
{
    NotApplicable = 1,
    SingleSelectRadioButtons = 2,
    SingleSelectD         


        

Answer 1

Here's an extension method that handles numbers and multiple uppercase characters sanely, and also allows for upper-casing specific acronyms in the final string:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Globalization;
using System.Text.RegularExpressions;
using System.Web.Configuration;

namespace System
{
    /// <summary>
    /// Extension methods for the string data type
    /// </summary>
    public static class ConventionBasedFormattingExtensions
    {
        /// <summary>
        /// Turn CamelCaseText into Camel Case Text.
        /// </summary>
        /// <param name="input"></param>
        /// <returns></returns>
        /// <remarks>Use AppSettings["SplitCamelCase_AllCapsWords"] to specify a comma-delimited list of words that should be ALL CAPS after split</remarks>
        /// <example>
        /// wordWordIDWord1WordWORDWord32Word2
        /// Word Word ID Word 1 Word WORD Word 32 Word 2
        /// 
        /// wordWordIDWord1WordWORDWord32WordID2ID
        /// Word Word ID Word 1 Word WORD Word 32 Word ID 2 ID
        /// 
        /// WordWordIDWord1WordWORDWord32Word2Aa
        /// Word Word ID Word 1 Word WORD Word 32 Word 2 Aa
        /// 
        /// wordWordIDWord1WordWORDWord32Word2A
        /// Word Word ID Word 1 Word WORD Word 32 Word 2 A
        /// </example>
        public static string SplitCamelCase(this string input)
        {
            if (input == null) return null;
            if (string.IsNullOrWhiteSpace(input)) return "";

            var separated = input;

            separated = SplitCamelCaseRegex.Replace(separated, @" $1").Trim();

            //Set ALL CAPS words
            if (_SplitCamelCase_AllCapsWords.Any())
                foreach (var word in _SplitCamelCase_AllCapsWords)
                    separated = SplitCamelCase_AllCapsWords_Regexes[word].Replace(separated, word.ToUpper());

            //Capitalize first letter
            var firstChar = separated.First(); //NullOrWhiteSpace handled earlier
            if (char.IsLower(firstChar))
                separated = char.ToUpper(firstChar) + separated.Substring(1);

            return separated;
        }

        private static readonly Regex SplitCamelCaseRegex = new Regex(@"
            (
                (?<=[a-z])[A-Z0-9] (?# lower-to-other boundaries )
                |
                (?<=[0-9])[a-zA-Z] (?# number-to-other boundaries )
                |
                (?<=[A-Z])[0-9] (?# cap-to-number boundaries; handles a specific issue with the next condition )
                |
                (?<=[A-Z])[A-Z](?=[a-z]) (?# handles longer strings of caps like ID or CMS by splitting off the last capital )
            )"
            , RegexOptions.Compiled | RegexOptions.IgnorePatternWhitespace
        );

        private static readonly string[] _SplitCamelCase_AllCapsWords =
            (WebConfigurationManager.AppSettings["SplitCamelCase_AllCapsWords"] ?? "")
                .Split(new[] { ',' }, StringSplitOptions.RemoveEmptyEntries)
                .Select(a => a.ToLowerInvariant().Trim())
                .ToArray()
                ;

        private static Dictionary<string, Regex> _SplitCamelCase_AllCapsWords_Regexes;
        private static Dictionary<string, Regex> SplitCamelCase_AllCapsWords_Regexes
        {
            get
            {
                if (_SplitCamelCase_AllCapsWords_Regexes == null)
                {
                    _SplitCamelCase_AllCapsWords_Regexes = new Dictionary<string,Regex>();
                    foreach(var word in _SplitCamelCase_AllCapsWords)
                        _SplitCamelCase_AllCapsWords_Regexes.Add(word, new Regex(@"\b" + word + @"\b", RegexOptions.Compiled | RegexOptions.IgnoreCase));
                }

                return _SplitCamelCase_AllCapsWords_Regexes;
            }
        }
    }
}

Answer 2

The solution from Eoin Campbell works good except if you have a Web Service.

You would need to do the Following as the Description Attribute is not serializable.

[DataContract]
public enum ControlSelectionType
{
    [EnumMember(Value = "Not Applicable")]
    NotApplicable = 1,
    [EnumMember(Value = "Single Select Radio Buttons")]
    SingleSelectRadioButtons = 2,
    [EnumMember(Value = "Completely Different Display Text")]
    SingleSelectDropDownList = 3,
}


public static string GetDescriptionFromEnumValue(Enum value)
{
    EnumMemberAttribute attribute = value.GetType()
        .GetField(value.ToString())
        .GetCustomAttributes(typeof(EnumMemberAttribute), false)
        .SingleOrDefault() as EnumMemberAttribute;
    return attribute == null ? value.ToString() : attribute.Value;
}

Answer 3

And if you don't fancy using regex - try this:

public static string SeperateByCamelCase(this string text, char splitChar = ' ') {

        var output = new StringBuilder();

        for (int i = 0; i < text.Length; i++)
        {
            var c = text[i];

            //if not the first and the char is upper
            if (i > 0 && char.IsUpper(c)) {

                var wasLastLower = char.IsLower(text[i - 1]);

                if (i + 1 < text.Length) //is there a next
                {
                    var isNextUpper = char.IsUpper(text[i + 1]);

                    if (!isNextUpper) //if next is not upper (start of a word).
                    {
                        output.Append(splitChar);
                    }
                    else if (wasLastLower) //last was lower but i'm upper and my next is an upper (start of an achromin). 'abcdHTTP' 'abcd HTTP'
                    {
                        output.Append(splitChar);
                    }
                }
                else
                {
                    //last letter - if its upper and the last letter was lower 'abcd' to 'abcd A'
                    if (wasLastLower)
                    {
                        output.Append(splitChar);
                    }
                }
            }

            output.Append(c);
        }


        return output.ToString();

    }

Passes these tests, it doesn't like numbers but i didn't need it to.

    [TestMethod()]
    public void ToCamelCaseTest()
    {

        var testData = new string[] { "AAACamel", "AAA", "SplitThisByCamel", "AnA", "doesnothing", "a", "A", "aasdasdAAA" };
        var expectedData = new string[] { "AAA Camel", "AAA", "Split This By Camel", "An A", "doesnothing", "a", "A", "aasdasd AAA" };

        for (int i = 0; i < testData.Length; i++)
        {
            var actual = testData[i].SeperateByCamelCase();
            var expected = expectedData[i];
            Assert.AreEqual(actual, expected);
        }

    }

Answer 4

I also have an enum which I had to separate. In my case this method solved the problem-

string SeparateCamelCase(string str)
{
    for (int i = 1; i < str.Length; i++)
    {
        if (char.IsUpper(str[i]))
        {
            str = str.Insert(i, " ");
            i++;
        }
    }
    return str;
}

Answer 5

I used:

    public static string SplitCamelCase(string input)
    {
        return System.Text.RegularExpressions.Regex.Replace(input, "([A-Z])", " $1", System.Text.RegularExpressions.RegexOptions.Compiled).Trim();
    }

Taken from http://weblogs.asp.net/jgalloway/archive/2005/09/27/426087.aspx

vb.net:

Public Shared Function SplitCamelCase(ByVal input As String) As String
    Return System.Text.RegularExpressions.Regex.Replace(input, "([A-Z])", " $1", System.Text.RegularExpressions.RegexOptions.Compiled).Trim()
End Function

Answer 6

#JustSayNoToRegex

Takes a C# identifier, with uderscores and numbers, and converts it to space-separated string.

public static class StringExtensions
{
    public static string SplitOnCase(this string identifier)
    {
        if (identifier == null || identifier.Length == 0) return string.Empty;
        var sb = new StringBuilder();

        if (identifier.Length == 1) sb.Append(char.ToUpperInvariant(identifier[0]));

        else if (identifier.Length == 2) sb.Append(char.ToUpperInvariant(identifier[0])).Append(identifier[1]);

        else {
            if (identifier[0] != '_') sb.Append(char.ToUpperInvariant(identifier[0]));
            for (int i = 1; i < identifier.Length; i++) {
                var current = identifier[i];
                var previous = identifier[i - 1];

                if (current == '_' && previous == '_') continue;

                else if (current == '_') {
                    sb.Append(' ');
                }

                else if (char.IsLetter(current) && previous == '_') {
                    sb.Append(char.ToUpperInvariant(current));
                }

                else if (char.IsDigit(current) && char.IsLetter(previous)) {
                    sb.Append(' ').Append(current);
                }

                else if (char.IsLetter(current) && char.IsDigit(previous)) {
                    sb.Append(' ').Append(char.ToUpperInvariant(current));
                }

                else if (char.IsUpper(current) && char.IsLower(previous) 
                    && (i < identifier.Length - 1 && char.IsUpper(identifier[i + 1]) || i == identifier.Length - 1)) {
                        sb.Append(' ').Append(current);
                }

                else if (char.IsUpper(current) && i < identifier.Length - 1 && char.IsLower(identifier[i + 1])) {
                    sb.Append(' ').Append(current);
                }

                else {
                    sb.Append(current);
                }
            }
        }
        return sb.ToString();
    }

}

Tests:

[TestFixture]
static class HelpersTests
{
    [Test]
    public static void Basic()
    {
        Assert.AreEqual("Foo", "foo".SplitOnCase());
        Assert.AreEqual("Foo", "_foo".SplitOnCase());
        Assert.AreEqual("Foo", "__foo".SplitOnCase());
        Assert.AreEqual("Foo", "___foo".SplitOnCase());
        Assert.AreEqual("Foo 2", "foo2".SplitOnCase());
        Assert.AreEqual("Foo 23", "foo23".SplitOnCase());
        Assert.AreEqual("Foo 23 A", "foo23A".SplitOnCase());
        Assert.AreEqual("Foo 23 Ab", "foo23Ab".SplitOnCase());
        Assert.AreEqual("Foo 23 Ab", "foo23_ab".SplitOnCase());
        Assert.AreEqual("Foo 23 Ab", "foo23___ab".SplitOnCase());
        Assert.AreEqual("Foo 23", "foo__23".SplitOnCase());
        Assert.AreEqual("Foo Bar", "Foo_bar".SplitOnCase());
        Assert.AreEqual("Foo Bar", "Foo____bar".SplitOnCase());
        Assert.AreEqual("AAA", "AAA".SplitOnCase());
        Assert.AreEqual("Foo A Aa", "fooAAa".SplitOnCase());
        Assert.AreEqual("Foo AAA", "fooAAA".SplitOnCase());
        Assert.AreEqual("Foo Bar", "FooBar".SplitOnCase());
        Assert.AreEqual("Mn M", "MnM".SplitOnCase());
        Assert.AreEqual("AS", "aS".SplitOnCase());
        Assert.AreEqual("As", "as".SplitOnCase());
        Assert.AreEqual("A", "a".SplitOnCase());
        Assert.AreEqual("_", "_".SplitOnCase());

    }
}