refactor(message-parser): optimize code point matchers (#20)

Author: ramsey

CHANGELOG.md

@@ -12,6 +12,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 - Provide `--add-missing-ids` extraction option to update source code with auto-generated identifiers
 - Add `Util\FormatHelper` that provides `getReader()` and `getWriter()` methods
 - Introduce `Format\Format` final static class for format constants
+- Port [@formatjs/icu-messageformat-parser](https://www.npmjs.com/package/@formatjs/icu-messageformat-parser) to FormatPHP (`FormatPHP\Icu\MessageFormat\Parser`)
 
 ### Changed
 

src/Icu/MessageFormat/Parser/Util/AbstractCodePointMatcher.php

@@ -30,6 +30,8 @@
 use function mb_ord;
 
 /**
+ * A helper for working with code points
+ *
  * @internal
  */
 class CodePointHelper
@@ -49,32 +51,61 @@ public function __construct()
         $this->isPotentialElementNameChar = new IsPotentialElementNameChar();
     }
 
+    /**
+     * Checks whether a code point is an alphabet character
+     *
+     * @see IsAlpha::matches()
+     */
     public function isAlpha(int $codepoint): bool
     {
         return $this->isAlpha->matches($codepoint);
     }
 
+    /**
+     * Checks whether a code point is an alphabet character or a forward slash ("/")
+     *
+     * @see IsAlpha::matches()
+     */
     public function isAlphaOrSlash(int $codepoint): bool
     {
-        return $this->isAlpha->matchesWithSlash($codepoint);
+        return $codepoint === 0x002f || $this->isAlpha->matches($codepoint);
     }
 
+    /**
+     * Checks whether a code point is in the Unicode Character Database
+     * White_Space and Pattern_White_Space groups
+     *
+     * @see IsWhiteSpace::matches()
+     */
     public function isWhiteSpace(int $codepoint): bool
     {
         return $this->isWhiteSpace->matches($codepoint);
     }
 
+    /**
+     * Checks whether a code point is in the Unicode Character Database
+     * Pattern_Syntax group
+     *
+     * @see IsPatternSyntax::matches()
+     */
     public function isPatternSyntax(int $codepoint): bool
     {
         return $this->isPatternSyntax->matches($codepoint);
     }
 
+    /**
+     * Checks whether the code point could be used in an HTML tag element name
+     *
+     * @see IsPotentialElementNameChar::matches()
+     */
     public function isPotentialElementNameChar(int $codepoint): bool
     {
         return $this->isPotentialElementNameChar->matches($codepoint);
     }
 
     /**
+     * Returns the code point for a character in a string array at a given offset
+     *
      * @param string[] $stringArray
      */
     public function charCodeAt(array $stringArray, int $offset): ?int
@@ -92,6 +123,9 @@ public function charCodeAt(array $stringArray, int $offset): ?int
         return null;
     }
 
+    /**
+     * Returns the string character for a given code point
+     */
     public function fromCharCode(int $code): ?string
     {
         $char = mb_chr($code, Parser::ENCODING);
@@ -104,6 +138,8 @@ public function fromCharCode(int $code): ?string
     }
 
     /**
+     * Returns a string of characters for the given code points
+     *
      * @throws InvalidUtf8CodePointException
      */
     public function fromCodePoint(int ...$codePoints): string

src/Icu/MessageFormat/Parser/Util/CodePointHelper.php

@@ -23,85 +23,19 @@
 namespace FormatPHP\Icu\MessageFormat\Parser\Util;
 
 /**
+ * Checks whether a code point is an alphabet character
+ *
  * @internal
  */
-class IsAlpha extends AbstractCodePointMatcher
+class IsAlpha implements CodePointMatcherInterface
 {
     /**
-     * Latin alphabet code points
-     *
-     * This array provides fast lookup of values.
-     */
-    protected const CODE_POINTS = [
-        0x0041,
-        0x0042,
-        0x0043,
-        0x0044,
-        0x0045,
-        0x0046,
-        0x0047,
-        0x0048,
-        0x0049,
-        0x004a,
-        0x004b,
-        0x004c,
-        0x004d,
-        0x004e,
-        0x004f,
-        0x0050,
-        0x0051,
-        0x0052,
-        0x0053,
-        0x0054,
-        0x0055,
-        0x0056,
-        0x0057,
-        0x0058,
-        0x0059,
-        0x005a,
-        0x0061,
-        0x0062,
-        0x0063,
-        0x0064,
-        0x0065,
-        0x0066,
-        0x0067,
-        0x0068,
-        0x0069,
-        0x006a,
-        0x006b,
-        0x006c,
-        0x006d,
-        0x006e,
-        0x006f,
-        0x0070,
-        0x0071,
-        0x0072,
-        0x0073,
-        0x0074,
-        0x0075,
-        0x0076,
-        0x0077,
-        0x0078,
-        0x0079,
-        0x007a,
-    ];
-
-    /**
-     * Checks whether the code point is a forward slash ("/") or an uppercase
-     * or lowercase alphabet character
-     */
-    public function matchesWithSlash(int $codepoint): bool
-    {
-        return $codepoint === 0x002f /* '/' */
-            || $this->matches($codepoint);
-    }
-
-    /**
-     * @inheritdoc
+     * Returns true if the code point is a Latin uppercase letter A-Z or
+     * lowercase letter a-z
      */
-    protected function getCodePoints(): array
+    public function matches(int $codepoint): bool
     {
-        return self::CODE_POINTS;
+        return ($codepoint >= 0x0041 && $codepoint <= 0x005a)
+            || ($codepoint >= 0x0061 && $codepoint <= 0x007a);
     }
 }