Added logic to properly treat underscores, dashes and spaces. Fixes #318

src/Humanizer.Tests/StringHumanizeTests.cs

@@ -11,6 +11,7 @@ public class StringHumanizeTests
         [InlineData("10IsInTheBegining", "10 is in the begining")]
         [InlineData("NumberIsAtTheEnd100", "Number is at the end 100")]
         [InlineData("XIsFirstWordInTheSentence", "X is first word in the sentence")]
+        [InlineData("XIsFirstWordInTheSentence ThenThereIsASpace", "X is first word in the sentence then there is a space")]
         public void CanHumanizeStringInPascalCase(string input, string expectedResult)
         {
             Assert.Equal(expectedResult, input.Humanize());
@@ -19,6 +20,14 @@ public void CanHumanizeStringInPascalCase(string input, string expectedResult)
         [Theory]
         [InlineData("Underscored_input_string_is_turned_into_sentence", "Underscored input string is turned into sentence")]
         [InlineData("Underscored_input_String_is_turned_INTO_sentence", "Underscored input String is turned INTO sentence")]
+        [InlineData("TEST 1 - THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1 -THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1- THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1_ THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1 _THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1 _ THIS IS A TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1 - THIS_IS_A_TEST", "TEST 1 THIS IS A TEST")]
+        [InlineData("TEST 1 - THIS is A Test", "TEST 1 THIS is A test")]
         public void CanHumanizeStringWithUnderscores(string input, string expectedReseult)
         {
             Assert.Equal(expectedReseult, input.Humanize());
@@ -28,6 +37,7 @@ public void CanHumanizeStringWithUnderscores(string input, string expectedReseul
         [InlineData("HTML", "HTML")]
         [InlineData("TheHTMLLanguage", "The HTML language")]
         [InlineData("HTMLIsTheLanguage", "HTML is the language")]
+        [InlineData("TheLanguage IsHTML", "The language is HTML")]
         [InlineData("TheLanguageIsHTML", "The language is HTML")]
         [InlineData("HTML5", "HTML 5")]
         [InlineData("1HTML", "1 HTML")]

src/Humanizer/StringHumanizeExtensions.cs

@@ -24,15 +24,17 @@ static string FromPascalCase(string input)
 (?# acronym to number)
 (?<=[A-Z])(?=[0-9])|
 (?# acronym to word)
-(?<=[A-Z])(?=[A-Z][a-z])
+(?<=[A-Z])(?=[A-Z][a-z])|
+(?# words/acronyms/numbers separated by space)
+(?<=[^\s])(?=[\s])
 ", RegexOptions.IgnorePatternWhitespace);
 
             var result = pascalCaseWordBoundaryRegex
                 .Split(input)
                 .Select(word =>
-                    word.ToCharArray().All(Char.IsUpper) && word.Length > 1
-                        ? word
-                        : word.ToLower())
+                    word.Trim().ToCharArray().All(Char.IsUpper) && word.Length > 1
+                        ? word.Trim()
+                        : word.Trim().ToLower())
                 .Aggregate((res, word) => res + " " + word);
 
             result = Char.ToUpper(result[0]) +
@@ -51,6 +53,12 @@ public static string Humanize(this string input)
             if (input.ToCharArray().All(Char.IsUpper))
                 return input;
 
+            // if input contains a dash or hyphen which preceeds or follows a space (or both, i.g. free-standing)
+            // remove the dash/hyphen and run it through FromPascalCase
+            Regex r = new Regex(@"[\s]{1}[-_][\s]{0}|[\s]{0}[-_][\s]{1}", RegexOptions.IgnoreCase);
+            if (r.IsMatch(input))
+                return FromPascalCase(FromUnderscoreDashSeparatedWords(input));
+
             if (input.Contains("_") || input.Contains("-"))
                 return FromUnderscoreDashSeparatedWords(input);