fix: pad with 0 when percent-encoding characters (via #1386)

* Fix lack of leading zero on low unicode values
* linter fixes
* logic fix
* more logic fixes
* drop unnecessary parens
* add necessary paren
* add tests

Author: adjenks

src/execute/oas3/style-serializer.js

@@ -33,7 +33,7 @@ export function encodeDisallowedCharacters(str, {escape} = {}, parse) {
     }
 
     const encoded = (toUTF8Bytes(char) || [])
-      .map(byte => byte.toString(16).toUpperCase())
+      .map(byte => `0${byte.toString(16).toUpperCase()}`.slice(-2))
       .map(encodedByte => `%${encodedByte}`)
       .join('')
 

test/oas3/execute/style-serializer.js

@@ -26,6 +26,7 @@ describe('OAS3 style serializer', () => {
       expect(tested('[')).toEqual('%5B')
       expect(tested(']')).toEqual('%5D')
       expect(tested('%')).toEqual('%25')
+      expect(tested('\n')).toEqual('%0A')
     })
 
     test('should correctly encode non-ASCII characters', () => {
@@ -58,6 +59,7 @@ describe('OAS3 style serializer', () => {
       expect(tested('@')).toEqual('@')
       expect(tested('[')).toEqual('[')
       expect(tested(']')).toEqual(']')
+      expect(tested('\n')).toEqual('\n')
 
       // Non-ASCII too!
       expect(tested('♥')).toEqual('♥')