Fix casing of unicode characters

Author: armano2

lib/utils/casing.js

@@ -1,18 +1,26 @@
 const assert = require('assert')
 
-const invalidChars = /[^a-zA-Z0-9:]+/g
+function parseWords (str) {
+  return str
+    .normalize()
+    .replace(/[^\p{L}0-9:]+/gu, ' ')
+    .split(/([\p{Upper}\s][0-9\p{Lower}:]*)/gu)
+    .map(word => word.trim())
+    .filter(Boolean)
+}
+
+function capitalizeFirstLetter (str) {
+  return str[0].toUpperCase() + str.slice(1)
+}
 
 /**
  * Convert text to kebab-case
  * @param {string} str Text to be converted
  * @return {string}
  */
 function kebabCase (str) {
-  return str
-    .replace(/[A-Z]/g, match => '-' + match)
-    .replace(/([^a-zA-Z])-([A-Z])/g, match => match[0] + match[2])
-    .replace(/^-/, '')
-    .replace(invalidChars, '-')
+  return parseWords(str)
+    .join('-')
     .toLowerCase()
 }
 
@@ -22,11 +30,8 @@ function kebabCase (str) {
  * @return {string}
  */
 function snakeCase (str) {
-  return str
-    .replace(/[A-Z]/g, match => '_' + match)
-    .replace(/([^a-zA-Z])_([A-Z])/g, match => match[0] + match[2])
-    .replace(/^_/, '')
-    .replace(invalidChars, '_')
+  return parseWords(str)
+    .join('_')
     .toLowerCase()
 }
 
@@ -36,12 +41,9 @@ function snakeCase (str) {
  * @return {string} Converted string
  */
 function camelCase (str) {
-  return str
-    .replace(/_/g, (_, index) => index === 0 ? _ : '-')
-    .replace(/(?:^\w|[A-Z]|\b\w)/g, (letter, index) =>
-      index === 0 ? letter.toLowerCase() : letter.toUpperCase()
-    )
-    .replace(invalidChars, '')
+  return parseWords(str)
+    .map((word, index) => index === 0 ? word.toLowerCase() : capitalizeFirstLetter(word))
+    .join('')
 }
 
 /**
@@ -50,10 +52,9 @@ function camelCase (str) {
  * @return {string} Converted string
  */
 function pascalCase (str) {
-  return str
-    .replace(/_/g, (_, index) => index === 0 ? _ : '-')
-    .replace(/(?:^\w|[A-Z]|\b\w)/g, (letter, index) => letter.toUpperCase())
-    .replace(invalidChars, '')
+  return parseWords(str)
+    .map((word) => capitalizeFirstLetter(word))
+    .join('')
 }
 
 const convertersMap = {

tests/lib/utils/casing.js

@@ -9,6 +9,7 @@ describe('getConverter()', () => {
   it('should convert string to camelCase', () => {
     const converter = casing.getConverter('camelCase')
 
+    assert.equal(converter('foo'), 'foo')
     assert.equal(converter('fooBar'), 'fooBar')
     assert.equal(converter('foo-bar'), 'fooBar')
     assert.equal(converter('foo_bar'), 'fooBar')
@@ -17,11 +18,14 @@ describe('getConverter()', () => {
     assert.equal(converter('FooBAR'), 'fooBAR')
     assert.equal(converter('Foo1BAZ'), 'foo1BAZ')
     assert.equal(converter('foo1b_a_z'), 'foo1bAZ')
+    assert.equal(converter('darИībaÊÊw'), 'darИībaÊÊw')
+    assert.equal(converter('    foo       Bar   '), 'fooBar')
   })
 
   it('should convert string to PascalCase', () => {
     const converter = casing.getConverter('PascalCase')
 
+    assert.equal(converter('foo'), 'Foo')
     assert.equal(converter('fooBar'), 'FooBar')
     assert.equal(converter('foo-bar'), 'FooBar')
     assert.equal(converter('foo_bar'), 'FooBar')
@@ -30,30 +34,56 @@ describe('getConverter()', () => {
     assert.equal(converter('FooBAR'), 'FooBAR')
     assert.equal(converter('Foo1BAZ'), 'Foo1BAZ')
     assert.equal(converter('foo1b_a_z'), 'Foo1bAZ')
+    assert.equal(converter('darИībaÊÊw'), 'DarИībaÊÊw')
+    assert.equal(converter('    foo       Bar   '), 'FooBar')
   })
 
   it('should convert string to kebab-case', () => {
     const converter = casing.getConverter('kebab-case')
 
+    assert.equal(converter('foo'), 'foo')
     assert.equal(converter('fooBar'), 'foo-bar')
     assert.equal(converter('foo-bar'), 'foo-bar')
     assert.equal(converter('foo_bar'), 'foo-bar')
     assert.equal(converter('FooBar'), 'foo-bar')
-    assert.equal(converter('Foo1Bar'), 'foo1bar')
+    assert.equal(converter('Foo1Bar'), 'foo1-bar')
     assert.equal(converter('FooBAR'), 'foo-b-a-r')
-    assert.equal(converter('Foo1BAZ'), 'foo1b-a-z')
+    assert.equal(converter('Foo1BAZ'), 'foo1-b-a-z')
     assert.equal(converter('foo1b_a_z'), 'foo1b-a-z')
+    assert.equal(converter('darИībaÊÊw'), 'dar-иība-ê-êw')
+    assert.equal(converter('    foo       Bar   '), 'foo-bar')
   })
 
   it('should convert string to snake_case', () => {
     const converter = casing.getConverter('snake_case')
 
+    assert.equal(converter('a'), 'a')
     assert.equal(converter('fooBar'), 'foo_bar')
     assert.equal(converter('foo-bar'), 'foo_bar')
     assert.equal(converter('FooBar'), 'foo_bar')
-    assert.equal(converter('Foo1Bar'), 'foo1bar')
+    assert.equal(converter('Foo1Bar'), 'foo1_bar')
     assert.equal(converter('FooBAR'), 'foo_b_a_r')
-    assert.equal(converter('Foo1BAZ'), 'foo1b_a_z')
+    assert.equal(converter('Foo1BAZ'), 'foo1_b_a_z')
     assert.equal(converter('foo1b_a_z'), 'foo1b_a_z')
+    assert.equal(converter('darИībaÊÊw'), 'dar_иība_ê_êw')
+    assert.equal(converter('    foo       Bar   '), 'foo_bar')
+  })
+
+  it('unicode tests', () => {
+    const words = [
+      ['ÊtreSîne', 'être_sîne'],
+      ['darbībaÊÊw', 'darbība_ê_êw'],
+      ['klâwen-ûf', 'klâwen_ûf'],
+      ['γλώσσαΤη', 'γλώσσα_τη'],
+      ['пустынныхИвдалП', 'пустынных_ивдал_п'],
+      ['kpłĄżć', 'kpł_ążć']
+    ]
+    for (const [word, snake] of words) {
+      assert.equal(casing.camelCase(word), casing.camelCase(casing.snakeCase(word)))
+      assert.equal(casing.pascalCase(word), casing.pascalCase(casing.kebabCase(word)))
+      assert.equal(casing.kebabCase(word), casing.kebabCase(casing.camelCase(word)))
+      assert.equal(casing.snakeCase(word), casing.snakeCase(casing.pascalCase(word)))
+      assert.equal(casing.snakeCase(word), snake)
+    }
   })
 })