Mark letter case changes to fix "camelCaseText" input

Author: VojtechVitek

camel.go

@@ -9,11 +9,13 @@ import (
 // Converts input string to "camelCase" (lower camel case) naming convention.
 // Removes all whitespace and special characters. Supports Unicode characters.
 func CamelCase(input string) string {
+	str := markLetterCaseChanges(input)
+
 	var b strings.Builder
 
 	state := idle
-	for i := 0; i < len(input); {
-		r, size := utf8.DecodeRuneInString(input[i:])
+	for i := 0; i < len(str); {
+		r, size := utf8.DecodeRuneInString(str[i:])
 		i += size
 		state = state.next(r)
 		switch state {
@@ -27,5 +29,6 @@ func CamelCase(input string) string {
 			b.WriteRune(unicode.ToLower(r))
 		}
 	}
+
 	return b.String()
 }

kebab.go

@@ -8,7 +8,9 @@ import (
 
 // Converts input string to "kebab-case" naming convention.
 // Removes all whitespace and special characters. Supports Unicode characters.
-func KebabCase(str string) string {
+func KebabCase(input string) string {
+	str := markLetterCaseChanges(input)
+
 	var b bytes.Buffer
 
 	state := idle

parser.go

@@ -1,20 +1,22 @@
 package textcase
 
 import (
+	"strings"
 	"unicode"
+	"unicode/utf8"
 )
 
-type parserStateMachine int
+type parser int
 
 const (
-	_             parserStateMachine = iota //   _$$_This is some text, OK?!
-	idle                                    // 1 ↑↑↑↑                  ↑   ↑
-	firstAlphaNum                           // 2     ↑    ↑  ↑    ↑     ↑
-	alphaNum                                // 3      ↑↑↑  ↑  ↑↑↑  ↑↑↑   ↑
-	delimiter                               // 4         ↑  ↑    ↑    ↑   ↑
+	_             parser = iota //   _$$_This is some text, OK?!
+	idle                        // 1 ↑↑↑↑                  ↑   ↑
+	firstAlphaNum               // 2     ↑    ↑  ↑    ↑     ↑
+	alphaNum                    // 3      ↑↑↑  ↑  ↑↑↑  ↑↑↑   ↑
+	delimiter                   // 4         ↑  ↑    ↑    ↑   ↑
 )
 
-func (s parserStateMachine) next(r rune) parserStateMachine {
+func (s parser) next(r rune) parser {
 	switch s {
 	case idle:
 		if isAlphaNum(r) {
@@ -41,3 +43,36 @@ func (s parserStateMachine) next(r rune) parserStateMachine {
 func isAlphaNum(r rune) bool {
 	return unicode.IsLetter(r) || unicode.IsNumber(r)
 }
+
+// Mark letter case changes, ie. "camelCaseTEXT" -> "camel_Case_TEXT".
+func markLetterCaseChanges(input string) string {
+	var b strings.Builder
+
+	wasLetter := false
+	countConsecutiveUpperLetters := 0
+
+	for i := 0; i < len(input); {
+		r, size := utf8.DecodeRuneInString(input[i:])
+		i += size
+
+		if unicode.IsLetter(r) {
+			if wasLetter && countConsecutiveUpperLetters > 1 && !unicode.IsUpper(r) {
+				b.WriteString("_")
+			}
+			if wasLetter && countConsecutiveUpperLetters == 0 && unicode.IsUpper(r) {
+				b.WriteString("_")
+			}
+		}
+
+		wasLetter = unicode.IsLetter(r)
+		if unicode.IsUpper(r) {
+			countConsecutiveUpperLetters++
+		} else {
+			countConsecutiveUpperLetters = 0
+		}
+
+		b.WriteRune(r)
+	}
+
+	return b.String()
+}

pascal.go

@@ -9,11 +9,13 @@ import (
 // Converts input string to "PascalCase" (upper camel case) naming convention.
 // Removes all whitespace and special characters. Supports Unicode characters.
 func PascalCase(input string) string {
+	str := markLetterCaseChanges(input)
+
 	var b strings.Builder
 
 	state := idle
-	for i := 0; i < len(input); {
-		r, size := utf8.DecodeRuneInString(input[i:])
+	for i := 0; i < len(str); {
+		r, size := utf8.DecodeRuneInString(str[i:])
 		i += size
 		state = state.next(r)
 		switch state {
@@ -23,5 +25,6 @@ func PascalCase(input string) string {
 			b.WriteRune(unicode.ToLower(r))
 		}
 	}
+
 	return b.String()
 }

snake.go

@@ -8,7 +8,9 @@ import (
 
 // Converts input string to "snake_case" naming convention.
 // Removes all whitespace and special characters. Supports Unicode characters.
-func SnakeCase(str string) string {
+func SnakeCase(input string) string {
+	str := markLetterCaseChanges(input)
+
 	var b bytes.Buffer
 
 	state := idle

textcase_test.go

@@ -6,49 +6,74 @@ import (
 )
 
 func TestTextCases(t *testing.T) {
-	t.Parallel()
-
 	tt := []struct {
 		in    string
 		camel string
 		snake string
 	}{
-		{in: "Add updated_at to users table", camel: "addUpdatedAtToUsersTable", snake: "add_updated_at_to_users_table"},
-		{in: "$()&^%(_--crazy__--input$)", camel: "crazyInput", snake: "crazy_input"},
-		{in: "Hey, this TEXT will have to obey some rules!!", camel: "heyThisTextWillHaveToObeySomeRules", snake: "hey_this_text_will_have_to_obey_some_rules"},
-		{in: "_$$_This is some text, OK?!", camel: "thisIsSomeTextOk", snake: "this_is_some_text_ok"},
-		{in: "_", camel: "", snake: ""},
-		{in: "$(((*&^%$#@!)))#$%^&*", camel: "", snake: ""},
 		{in: "", camel: "", snake: ""},
+		{in: "_", camel: "", snake: ""},
 		{in: "a", camel: "a", snake: "a"},
 		{in: "a___", camel: "a", snake: "a"},
+		{in: "___a", camel: "a", snake: "a"},
+		{in: "a_b", camel: "aB", snake: "a_b"},
 		{in: "a___b", camel: "aB", snake: "a_b"},
 		{in: "ax___by", camel: "axBy", snake: "ax_by"},
+		{in: "someText", camel: "someText", snake: "some_text"},
+		{in: "someTEXT", camel: "someText", snake: "some_text"},
+		{in: "NeXT", camel: "neXt", snake: "ne_xt"},
+		{in: "Add updated_at to users table", camel: "addUpdatedAtToUsersTable", snake: "add_updated_at_to_users_table"},
+		{in: "Hey, this TEXT will have to obey some rules!!", camel: "heyThisTextWillHaveToObeySomeRules", snake: "hey_this_text_will_have_to_obey_some_rules"},
 		{in: "Háčky, čárky. Příliš žluťoučký kůň úpěl ďábelské ódy.", camel: "háčkyČárkyPřílišŽluťoučkýKůňÚpělĎábelskéÓdy", snake: "háčky_čárky_příliš_žluťoučký_kůň_úpěl_ďábelské_ódy"},
 		{in: "here comes O'Brian", camel: "hereComesOBrian", snake: "here_comes_o_brian"},
+		{in: "thisIsCamelCase", camel: "thisIsCamelCase", snake: "this_is_camel_case"},
+		{in: "this_is_snake_case", camel: "thisIsSnakeCase", snake: "this_is_snake_case"},
+		{in: "__snake_case__", camel: "snakeCase", snake: "snake_case"},
+		{in: "fromCamelCaseToCamelCase", camel: "fromCamelCaseToCamelCase", snake: "from_camel_case_to_camel_case"},
+		{in: "$()&^%(_--crazy__--input$)", camel: "crazyInput", snake: "crazy_input"},
+		{in: "_$$_This is some text, OK?!", camel: "thisIsSomeTextOk", snake: "this_is_some_text_ok"},
+		{in: "$(((*&^%$#@!)))#$%^&*", camel: "", snake: ""},
 	}
 
 	for _, test := range tt {
 		// camelCase
 		if got := CamelCase(test.in); got != test.camel {
-			t.Errorf("unexpected camelCase for input(%q), got %q, want %q", test.in, got, test.camel)
+			t.Errorf("unexpected camelCase for %q: got %q, want %q", test.in, got, test.camel)
 		}
 
 		// PascalCase
 		testPascal := strings.Title(test.camel)
 		if got := PascalCase(test.in); got != testPascal {
-			t.Errorf("unexpected PascalCase for input(%q), got %q, want %q", test.in, got, testPascal)
+			t.Errorf("unexpected PascalCase for %q: got %q, want %q", test.in, got, testPascal)
 		}
 
 		// snake_case
 		if got := SnakeCase(test.in); got != test.snake {
-			t.Errorf("unexpected snake_case for input(%q), got %q, want %q", test.in, got, test.snake)
+			t.Errorf("unexpected snake_case for %q: got %q, want %q", test.in, got, test.snake)
 		}
 
 		// kebab-case
 		testKebab := strings.ReplaceAll(test.snake, "_", "-")
 		if got := KebabCase(test.in); got != testKebab {
-			t.Errorf("unexpected kebab-case for input(%q), got %q, want %q", test.in, got, testKebab)
+			t.Errorf("unexpected kebab-case for %q: got %q, want %q", test.in, got, testKebab)
+		}
+	}
+}
+
+func TestMarkLetterCaseChanges(t *testing.T) {
+	tt := []struct {
+		in  string
+		out string
+	}{
+		{in: "detectUpperLowerChanges", out: "detect_Upper_Lower_Changes"},
+		{in: "detectUPPERchange", out: "detect_UPPER_change"},
+		{in: "detect_UPPER_change", out: "detect_UPPER_change"},
+		{in: "Some camelCase and PascalCase text, OK?", out: "Some camel_Case and Pascal_Case text, OK?"},
+	}
+
+	for _, test := range tt {
+		if got := markLetterCaseChanges(test.in); got != test.out {
+			t.Errorf("unexpected markLowerUpperChanges for %q: got %q, want %q", test.in, got, test.out)
 		}
 	}
 }