Introduce the Encoder class

* Introduce the Encoder class to normalize encoding/decoding in all packages * Introduce the KeyValuePairConverter class to normalize key/value parsing and building * Rewrite QueryString parser/builder class * Add new methods to QueryString
thephpleague · Aug 19, 2023 · 2b57d2a · 2b57d2a
1 parent c1cf91e
commit 2b57d2a
Show file tree

Hide file tree

Showing 10 changed files with 97 additions and 117 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -2,6 +2,24 @@
 
 All Notable changes to `League\Uri\Components` will be documented in this file
 
+## Next - TBD
+
+### Added
+
+- `Modifier::encodeQuery`
+
+### Fixed
+
+- Using the `Encoder` class to normalize encoding and decoding in all packages
+
+### Deprecated
+
+- None
+
+### Removed
+
+- None
+
 ## [7.0.0](https://github.com/thephpleague/uri-components/compare/2.4.1...7.0.0) - 2023-08-10
 
 ### Added

diff --git a/Components/Component.php b/Components/Component.php
@@ -16,31 +16,17 @@
 use League\Uri\Contracts\UriAccess;
 use League\Uri\Contracts\UriComponentInterface;
 use League\Uri\Contracts\UriInterface;
+use League\Uri\Encoder;
 use League\Uri\Exceptions\SyntaxError;
 use League\Uri\Uri;
 use Psr\Http\Message\UriInterface as Psr7UriInterface;
 use Stringable;
 use function preg_match;
-use function preg_replace_callback;
-use function rawurldecode;
-use function rawurlencode;
 use function sprintf;
-use function strtoupper;
 
 abstract class Component implements UriComponentInterface
 {
-    protected const REGEXP_ENCODED_CHARS = ',%[A-Fa-f0-9]{2},';
     protected const REGEXP_INVALID_URI_CHARS = '/[\x00-\x1f\x7f]/';
-    protected const REGEXP_NO_ENCODING = '/[^A-Za-z0-9_\-.~]/';
-    protected const REGEXP_NON_ASCII_PATTERN = '/[^\x20-\x7f]/';
-    protected const REGEXP_PREVENTS_DECODING = ',%
-     	2[A-F|1-2|4-9]|
-        3[0-9|B|D]|
-        4[1-9|A-F]|
-        5[0-9|A|F]|
-        6[1-9|A-F]|
-        7[0-9|E]
-    ,ix';
 
     abstract public function value(): ?string;
 
@@ -78,7 +64,7 @@ final protected static function filterUri(Stringable|string $uri): UriInterface|
      */
     protected function validateComponent(Stringable|int|string|null $component): ?string
     {
-        return $this->decodeComponent(self::filterComponent($component));
+        return Encoder::decodePartial($component);
     }
 
     /**
@@ -95,45 +81,4 @@ final protected static function filterComponent(Stringable|int|string|null $comp
             default => (string) $component,
         };
     }
-
-    /**
-     * Filter the URI password component.
-     */
-    protected function decodeComponent(?string $str): ?string
-    {
-        return match (true) {
-            null === $str => null,
-            default => preg_replace_callback(self::REGEXP_ENCODED_CHARS, $this->decodeMatches(...), $str),
-        };
-    }
-
-    /**
-     * Decodes Matches sequence.
-     */
-    protected function decodeMatches(array $matches): string
-    {
-        return match (true) {
-            1 === preg_match(static::REGEXP_PREVENTS_DECODING, $matches[0]) => strtoupper($matches[0]),
-            default => rawurldecode($matches[0]),
-        };
-    }
-
-    /**
-     * Returns the component as converted for RFC3986.
-     */
-    protected function encodeComponent(?string $str, string $regexp): ?string
-    {
-        return match (true) {
-            null === $str || 1 !== preg_match(self::REGEXP_NO_ENCODING, $str) => $str,
-            default => preg_replace_callback($regexp, $this->encodeMatches(...), $str) ?? rawurlencode($str),
-        };
-    }
-
-    /**
-     * Encode Matches sequence.
-     */
-    protected function encodeMatches(array $matches): string
-    {
-        return rawurlencode($matches[0]);
-    }
 }
diff --git a/Components/DataPath.php b/Components/DataPath.php
@@ -103,15 +103,11 @@ private function filterPath(string $path): string
      */
     private function filterMimeType(string $mimetype): string
     {
-        if ('' == $mimetype) {
-            return self::DEFAULT_MIMETYPE;
-        }
-
-        if (1 === preg_match(self::REGEXP_MIMETYPE, $mimetype)) {
-            return $mimetype;
-        }
-
-        throw new SyntaxError(sprintf('Invalid mimeType, `%s`.', $mimetype));
+        return match (true) {
+            '' == $mimetype => self::DEFAULT_MIMETYPE,
+            1 === preg_match(self::REGEXP_MIMETYPE, $mimetype) =>  $mimetype,
+            default => throw new SyntaxError(sprintf('Invalid mimeType, `%s`.', $mimetype)),
+        };
     }
 
     /**
@@ -302,7 +298,11 @@ private function formatComponent(
 
         $path = $mimetype.$parameters.','.$data;
 
-        return preg_replace_callback(self::REGEXP_DATAPATH_ENCODING, $this->encodeMatches(...), $path) ?? $path;
+        return preg_replace_callback(
+            self::REGEXP_DATAPATH_ENCODING,
+            static fn (array $matches): string => rawurlencode($matches[0]),
+            $path
+        ) ?? $path;
     }
 
     public function toAscii(): DataPathInterface

diff --git a/Components/Fragment.php b/Components/Fragment.php
@@ -15,13 +15,12 @@
 
 use League\Uri\Contracts\FragmentInterface;
 use League\Uri\Contracts\UriInterface;
+use League\Uri\Encoder;
 use Psr\Http\Message\UriInterface as Psr7UriInterface;
 use Stringable;
 
 final class Fragment extends Component implements FragmentInterface
 {
-    private const REGEXP_FRAGMENT_ENCODING = '/[^A-Za-z0-9_\-.~!$&\'()*+,;=%:\/@?]+|%(?![A-Fa-f0-9]{2})/';
-
     private readonly ?string $fragment;
 
     /**
@@ -53,7 +52,7 @@ public static function fromUri(Stringable|string $uri): self
 
     public function value(): ?string
     {
-        return $this->encodeComponent($this->fragment, self::REGEXP_FRAGMENT_ENCODING);
+        return Encoder::encodeQueryOrFragment($this->fragment);
     }
 
     public function getUriComponent(): string

diff --git a/Components/HierarchicalPath.php b/Components/HierarchicalPath.php
@@ -17,6 +17,7 @@
 use League\Uri\Contracts\PathInterface;
 use League\Uri\Contracts\SegmentedPathInterface;
 use League\Uri\Contracts\UriInterface;
+use League\Uri\Encoder;
 use League\Uri\Exceptions\OffsetOutOfBounds;
 use League\Uri\Exceptions\SyntaxError;
 use Psr\Http\Message\UriInterface as Psr7UriInterface;
@@ -59,7 +60,7 @@ private function __construct(Stringable|string $path)
         }
 
         $this->path = $path;
-        $segments = $this->decodeComponent($this->path->value()) ?? '';
+        $segments = $this->path->decoded();
         if ($this->path->isAbsolute()) {
             $segments = substr($segments, 1);
         }
@@ -153,7 +154,7 @@ public function decoded(): string
 
     public function getDirname(): string
     {
-        $path = (string) $this->decodeComponent($this->path->toString());
+        $path = $this->path->decoded();
 
         return str_replace(
             ['\\', "\0"],
@@ -297,7 +298,7 @@ public function withSegment(int $key, Stringable|string $segment): SegmentedPath
             $segment = new self($segment);
         }
 
-        $segment = $this->decodeComponent((string) $segment);
+        $segment = Encoder::decodeAll($segment);
         if ($segment === $this->segments[$key]) {
             return $this;
         }

diff --git a/Components/Host.php b/Components/Host.php
@@ -41,6 +41,8 @@
 
 final class Host extends Component implements IpHostInterface
 {
+    protected const REGEXP_NON_ASCII_PATTERN = '/[^\x20-\x7f]/';
+
     /**
      * @see https://tools.ietf.org/html/rfc3986#section-3.2.2
      *

diff --git a/Components/Path.php b/Components/Path.php
@@ -15,6 +15,7 @@
 
 use League\Uri\Contracts\PathInterface;
 use League\Uri\Contracts\UriInterface;
+use League\Uri\Encoder;
 use Psr\Http\Message\UriInterface as Psr7UriInterface;
 use Stringable;
 use function array_pop;
@@ -27,7 +28,6 @@
 final class Path extends Component implements PathInterface
 {
     private const DOT_SEGMENTS = ['.' => 1, '..' => 1];
-    private const REGEXP_PATH_ENCODING = '/[^A-Za-z0-9_\-.!$&\'()*+,;=%:\/@]+|%(?![A-Fa-f0-9]{2})/';
     private const SEPARATOR = '/';
 
     private readonly string $path;
@@ -76,7 +76,7 @@ public static function fromUri(Stringable|string $uri): self
 
     public function value(): ?string
     {
-        return $this->encodeComponent($this->path, self::REGEXP_PATH_ENCODING);
+        return Encoder::encodePath($this->path);
     }
 
     public function decoded(): string

diff --git a/Components/QueryTest.php b/Components/QueryTest.php
@@ -37,10 +37,10 @@ protected function setUp(): void
     public function testSeparator(): void
     {
         $query = Query::new('foo=bar&kingkong=toto');
-        $newQuery = $query->withSeparator('|');
+        $newQuery = $query->withSeparator(';');
         self::assertSame('&', $query->getSeparator());
-        self::assertSame('|', $newQuery->getSeparator());
-        self::assertSame('foo=bar|kingkong=toto', $newQuery->value());
+        self::assertSame(';', $newQuery->getSeparator());
+        self::assertSame('foo=bar;kingkong=toto', $newQuery->value());
 
         $this->expectException(SyntaxError::class);
         $newQuery->withSeparator('');
@@ -682,11 +682,17 @@ public static function getURIProvider(): iterable
         ];
     }
 
-    public function testCreateFromRFCSpecification(): void
+    public function testItFailsToCreateFromRFCSpecificationWithInvalidSeparator(): void
     {
-        self::assertEquals(
-            Query::fromRFC3986('foo=b%20ar|foo=baz', '|'),
-            Query::fromRFC1738('foo=b+ar|foo=baz', '|')
-        );
+        $this->expectException(SyntaxError::class);
+
+        Query::fromRFC3986('foo=b%20ar;foo=baz', ''); /* @phpstan-ignore-line */
+    }
+
+    public function testItFailsToCreateFromRFCSpecificationWithEmptySeparator(): void
+    {
+        $this->expectException(SyntaxError::class);
+
+        Query::fromRFC1738('foo=b%20ar;foo=baz', ''); /* @phpstan-ignore-line */
     }
 }
diff --git a/Components/UserInfo.php b/Components/UserInfo.php
@@ -17,20 +17,15 @@
 use League\Uri\Contracts\UriComponentInterface;
 use League\Uri\Contracts\UriInterface;
 use League\Uri\Contracts\UserInfoInterface;
+use League\Uri\Encoder;
 use League\Uri\Exceptions\SyntaxError;
 use Psr\Http\Message\UriInterface as Psr7UriInterface;
 use SensitiveParameter;
 use Stringable;
 use function explode;
-use function preg_replace_callback;
-use function rawurldecode;
 
 final class UserInfo extends Component implements UserInfoInterface
 {
-    private const REGEXP_USER_ENCODING = '/[^A-Za-z0-9_\-.~!$&\'()*+,;=%]+|%(?![A-Fa-f0-9]{2})/x';
-    private const REGEXP_PASS_ENCODING = '/[^A-Za-z0-9_\-.~!$&\'()*+,;=%:]+|%(?![A-Fa-f0-9]{2})/x';
-    private const REGEXP_ENCODED_CHAR = ',%[A-Fa-f0-9]{2},';
-
     private readonly ?string $username;
     private readonly ?string $password;
 
@@ -69,11 +64,10 @@ public static function fromUri(Stringable|string $uri): self
      */
     public static function fromAuthority(Stringable|string|null $authority): self
     {
-        if (!$authority instanceof AuthorityInterface) {
-            $authority = Authority::new($authority);
-        }
-
-        return self::new($authority->getUserInfo());
+        return match (true) {
+            $authority instanceof AuthorityInterface => self::new($authority->getUserInfo()),
+            default => self::new(Authority::new($authority)->getUserInfo()),
+        };
     }
 
     /**
@@ -111,33 +105,16 @@ public static function new(Stringable|string|null $value = null): self
 
         [$user, $pass] = explode(':', $value, 2) + [1 => null];
 
-        return new self(self::decode($user), self::decode($pass));
-    }
-
-    /**
-     * Decodes an encoded string.
-     */
-    private static function decode(?string $str): ?string
-    {
-        return null === $str ? null : preg_replace_callback(
-            self::REGEXP_ENCODED_CHAR,
-            static fn (array $matches): string => rawurldecode($matches[0]),
-            $str
-        );
+        return new self(Encoder::decodeAll($user), Encoder::decodeAll($pass));
     }
 
     public function value(): ?string
     {
-        if (null === $this->username) {
-            return null;
-        }
-
-        $userInfo = $this->encodeComponent($this->username, self::REGEXP_USER_ENCODING);
-        if (null === $this->password) {
-            return $userInfo;
-        }
-
-        return $userInfo.':'.$this->encodeComponent($this->password, self::REGEXP_PASS_ENCODING);
+        return match (true) {
+            null === $this->username => null,
+            null === $this->password => Encoder::encodeUser($this->username),
+            default => Encoder::encodeUser($this->username).':'.Encoder::encodePassword($this->password),
+        };
     }
 
     public function getUriComponent(): string