Add basic validation error logging

Based on the draft whatwg/url#502

CHANGELOG.md

@@ -11,6 +11,8 @@
 ### Added
 
 - Performance improvements
+- Basic validation error logging
+
 ### Changed
 
 - Forbid C0 control code points and U+007F DEL code point in non-opaque domain names per [whatwg/url#685](https://github.com/whatwg/url/pull/685)

composer.json

@@ -15,6 +15,7 @@
         "ext-json": "*",
         "ext-mbstring": "*",
         "brick/math": "^0.8.13 || ^0.9",
+        "psr/log": "^3.0",
         "rowbot/idna": "^0.1.5"
     },
     "require-dev": {

phpstan.neon

@@ -24,3 +24,7 @@ parameters:
             path: src/String/Utf8String.php
 
         # Note to self: You can't escape single quotes in neon. Another possible alternative would be to replace single quotes with \x27.
+        -
+            message: '#Instanceof between Psr\\Log\\LoggerInterface and Psr\\Log\\LoggerInterface will always evaluate to true\.#'
+            path: src/URL.php
+            count: 1

src/BasicURLParser.php

@@ -4,12 +4,22 @@
 
 namespace Rowbot\URL;
 
+use Psr\Log\LoggerAwareInterface;
+use Psr\Log\LoggerAwareTrait;
+use Psr\Log\LoggerInterface;
 use Rowbot\URL\State\State;
 use Rowbot\URL\String\StringBuffer;
 use Rowbot\URL\String\USVStringInterface;
 
-class BasicURLParser
+class BasicURLParser implements LoggerAwareInterface
 {
+    use LoggerAwareTrait;
+
+    public function __construct(?LoggerInterface $logger = null)
+    {
+        $this->logger = $logger;
+    }
+
     /**
      * The parser can parse both absolute and relative URLs. If a relative URL is given, a base URL must also be given
      * so that an absolute URL can be resolved. It can also parse individual parts of a URL when the default starting
@@ -50,21 +60,32 @@ public function parse(
 
             if ($count !== 0) {
                 // Validation error.
+                $this->logger?->notice('unexpected-c0-control-or-space');
             }
         }
 
         $input = $input->replaceRegex('/[\x09\x0A\x0D]+/u', '', -1, $count);
 
         if ($count !== 0) {
             // Validation error.
+            $this->logger?->notice('unexpected-ascii-tab-or-newline');
         }
 
         $iter = $input->getIterator();
         $iter->rewind();
         // length + imaginary eof character
         $length = $input->length() + 1;
         $buffer = new StringBuffer();
-        $context = new ParserContext($input, $iter, $buffer, $url, $base, $stateOverride, $encodingOverride);
+        $context = new ParserContext(
+            $input,
+            $iter,
+            $buffer,
+            $url,
+            $base,
+            $stateOverride,
+            $encodingOverride,
+            $this->logger
+        );
 
         do {
             $status = $context->state->handle($context, $iter->current());

src/Component/Host/HostParser.php

@@ -5,6 +5,7 @@
 namespace Rowbot\URL\Component\Host;
 
 use Rowbot\Idna\Idna;
+use Rowbot\URL\ParserContext;
 use Rowbot\URL\String\CodePoint;
 use Rowbot\URL\String\EncodeSet;
 use Rowbot\URL\String\PercentEncodeTrait;
@@ -34,37 +35,46 @@ class HostParser
      *
      * @return \Rowbot\URL\Component\Host\HostInterface|false The returned Host can never be a null host.
      */
-    public function parse(USVStringInterface $input, bool $isNotSpecial = false): HostInterface|false
-    {
+    public function parse(
+        ParserContext $context,
+        USVStringInterface $input,
+        bool $isNotSpecial = false
+    ): HostInterface|false {
         if ($input->startsWith('[')) {
             if (!$input->endsWith(']')) {
                 // Validation error.
+                $context->logger?->warning('unclosed-ipv6-address');
+
                 return false;
             }
 
-            return IPv6AddressParser::parse($input->substr(1, -1));
+            return IPv6AddressParser::parse($context, $input->substr(1, -1));
         }
 
         if ($isNotSpecial) {
-            return $this->parseOpaqueHost($input);
+            return $this->parseOpaqueHost($context, $input);
         }
 
         assert(!$input->isEmpty());
         $domain = rawurldecode((string) $input);
-        $asciiDomain = $this->domainToAscii($domain);
+        $asciiDomain = $this->domainToAscii($context, $domain);
 
         if ($asciiDomain === false) {
             // Validation error.
+            $context->logger?->warning('domain-to-ascii-failure');
+
             return false;
         }
 
         if ($asciiDomain->matches('/[' . self::FORBIDDEN_DOMAIN_CODEPOINTS . ']/u')) {
             // Validation error.
+            $context->logger?->warning('domain-forbidden-code-point');
+
             return false;
         }
 
         if (IPv4AddressParser::endsInIPv4Number($asciiDomain)) {
-            return IPv4AddressParser::parse($asciiDomain);
+            return IPv4AddressParser::parse($context, $asciiDomain);
         }
 
         return $asciiDomain;
@@ -73,7 +83,7 @@ public function parse(USVStringInterface $input, bool $isNotSpecial = false): Ho
     /**
      * @see https://url.spec.whatwg.org/#concept-domain-to-ascii
      */
-    private function domainToAscii(string $domain, bool $beStrict = false): StringHost|false
+    private function domainToAscii(ParserContext $context, string $domain, bool $beStrict = false): StringHost|false
     {
         $result = Idna::toAscii($domain, [
             'CheckHyphens'            => false,
@@ -87,11 +97,15 @@ private function domainToAscii(string $domain, bool $beStrict = false): StringHo
 
         if ($convertedDomain === '') {
             // Validation error.
+            $context->logger?->warning('domain-to-ascii-empty-domain-failure');
+
             return false;
         }
 
         if ($result->hasErrors()) {
             // Validation error.
+            $context->logger?->warning('domain-to-ascii-failure');
+
             return false;
         }
 
@@ -103,20 +117,24 @@ private function domainToAscii(string $domain, bool $beStrict = false): StringHo
      *
      * @see https://url.spec.whatwg.org/#concept-opaque-host-parser
      */
-    private function parseOpaqueHost(USVStringInterface $input): HostInterface|false
+    private function parseOpaqueHost(ParserContext $context, USVStringInterface $input): HostInterface|false
     {
         if ($input->matches('/[' . self::FORBIDDEN_HOST_CODEPOINTS . ']/u')) {
             // Validation error.
+            $context->logger?->warning('opaque-host-forbidden-code-point');
+
             return false;
         }
 
         foreach ($input as $i => $codePoint) {
             if (!CodePoint::isUrlCodePoint($codePoint) && $codePoint !== '%') {
                 // Validation error.
+                $context->logger?->notice('invalid-url-code-point');
             }
 
             if ($codePoint === '%' && !$input->substr($i + 1)->startsWithTwoAsciiHexDigits()) {
                 // Validation error.
+                $context->logger?->notice('unescaped-percent-sign');
             }
         }
 

src/Component/Host/IPv4AddressParser.php

@@ -5,6 +5,7 @@
 namespace Rowbot\URL\Component\Host;
 
 use Rowbot\URL\Component\Host\Math\NumberFactory;
+use Rowbot\URL\ParserContext;
 use Rowbot\URL\String\CodePoint;
 use Rowbot\URL\String\USVStringInterface;
 
@@ -18,7 +19,7 @@
  */
 class IPv4AddressParser
 {
-    public static function parse(USVStringInterface $input): IPv4Address|false
+    public static function parse(ParserContext $context, USVStringInterface $input): IPv4Address|false
     {
         // 1. Let validationError be false.
         //
@@ -44,6 +45,8 @@ public static function parse(USVStringInterface $input): IPv4Address|false
 
         // 4. If parts’s size is greater than 4, validation error, return failure.
         if ($count > 4) {
+            $context->logger?->warning('ipv4-too-many-parts');
+
             return false;
         }
 
@@ -57,6 +60,8 @@ public static function parse(USVStringInterface $input): IPv4Address|false
 
             // 6.2. If result is failure, validation error, return failure.
             if ($result === false) {
+                $context->logger?->warning('ipv4-invalid-radix-digit');
+
                 return false;
             }
 
@@ -75,11 +80,14 @@ public static function parse(USVStringInterface $input): IPv4Address|false
         // And therefore error reporting resumes.
         if ($validationError) {
             // Validation error.
+            $context->logger?->notice('unexpected-non-decimal-number');
         }
 
         // 8. If any item in numbers is greater than 255, validation error.
         foreach ($numbers as $number) {
             if ($number->isGreaterThan(255)) {
+                $context->logger?->warning('ipv4-part-out-of-range');
+
                 break;
             }
         }
@@ -98,6 +106,8 @@ public static function parse(USVStringInterface $input): IPv4Address|false
         // 10. If the last item in numbers is greater than or equal to 256 ** (5 − numbers’s size), validation error,
         // return failure.
         if ($numbers[$size - 1]->isGreaterThanOrEqualTo($limit)) {
+            $context->logger?->warning('ipv4-part-out-of-range');
+
             return false;
         }
 

src/Component/Host/IPv6AddressParser.php

@@ -4,6 +4,7 @@
 
 namespace Rowbot\URL\Component\Host;
 
+use Rowbot\URL\ParserContext;
 use Rowbot\URL\String\CodePoint;
 use Rowbot\URL\String\StringIteratorInterface;
 use Rowbot\URL\String\USVStringInterface;
@@ -16,7 +17,7 @@
  */
 class IPv6AddressParser
 {
-    public static function parse(USVStringInterface $input): IPv6Address|false
+    public static function parse(ParserContext $context, USVStringInterface $input): IPv6Address|false
     {
         // 1. Let address be a new IPv6 address whose IPv6 pieces are all 0.
         $address = [0, 0, 0, 0, 0, 0, 0, 0];
@@ -35,6 +36,8 @@ public static function parse(USVStringInterface $input): IPv6Address|false
         if ($iter->current() === ':') {
             // 5.1. If remaining does not start with U+003A (:), validation error, return failure.
             if ($iter->peek() !== ':') {
+                $context->logger?->warning('invalid-compressed-ipv6-address');
+
                 return false;
             }
 
@@ -49,13 +52,17 @@ public static function parse(USVStringInterface $input): IPv6Address|false
         while ($iter->valid()) {
             // 6.1. If pieceIndex is 8, validation error, return failure.
             if ($pieceIndex === 8) {
+                $context->logger?->warning('ipv6-too-many-pieces');
+
                 return false;
             }
 
             // 6.2. If c is U+003A (:), then:
             if ($iter->current() === ':') {
                 // 6.2.1. If compress is non-null, validation error, return failure.
                 if ($compress !== null) {
+                    $context->logger?->warning('ipv6-multiple-compression');
+
                     return false;
                 }
 
@@ -84,6 +91,8 @@ public static function parse(USVStringInterface $input): IPv6Address|false
             if ($iter->current() === '.') {
                 // 6.5.1. If length is 0, validation error, return failure.
                 if ($length === 0) {
+                    $context->logger?->warning('ipv4-in-ipv6-empty-part');
+
                     return false;
                 }
 
@@ -92,10 +101,12 @@ public static function parse(USVStringInterface $input): IPv6Address|false
 
                 // 6.5.3. If pieceIndex is greater than 6, validation error, return failure.
                 if ($pieceIndex > 6) {
+                    $context->logger?->warning('ipv4-in-ipv6-too-many-pieces');
+
                     return false;
                 }
 
-                $result = self::parseIPv4Address($iter, $address, $pieceIndex);
+                $result = self::parseIPv4Address($context, $iter, $address, $pieceIndex);
 
                 if ($result === false) {
                     return false;
@@ -114,11 +125,15 @@ public static function parse(USVStringInterface $input): IPv6Address|false
 
                 // 6.2.2. If c is the EOF code point, validation error, return failure.
                 if (!$iter->valid()) {
+                    $context->logger?->warning('ipv6-unexpected-eof');
+
                     return false;
                 }
 
             // 6.7. Otherwise, if c is not the EOF code point, validation error, return failure.
             } elseif ($iter->valid()) {
+                $context->logger?->warning('ipv6-unexpected-delimiter');
+
                 return false;
             }
 
@@ -147,6 +162,8 @@ public static function parse(USVStringInterface $input): IPv6Address|false
 
         // Otherwise, if compress is null and pieceIndex is not 8, validation error, return failure.
         } elseif ($pieceIndex !== 8) {
+            $context->logger?->warning('ipv6-too-few-pieces');
+
             return false;
         }
 
@@ -160,6 +177,7 @@ public static function parse(USVStringInterface $input): IPv6Address|false
      * @return array{0: list<int>, 1: int}|false
      */
     private static function parseIPv4Address(
+        ParserContext $context,
         StringIteratorInterface $iter,
         array $address,
         int $pieceIndex
@@ -177,6 +195,8 @@ private static function parseIPv4Address(
                 // 6.5.5.2.2 Otherwise, validation error, return failure.
                 if ($iter->current() !== '.' || $numbersSeen >= 4) {
                     // Validation error.
+                    $context->logger?->warning('ipv4-in-ipv6-too-many-parts');
+
                     return false;
                 }
 
@@ -189,6 +209,8 @@ private static function parseIPv4Address(
             // 6.5.5.3. If c is not an ASCII digit, validation error, return failure.
             if (strpbrk($current, CodePoint::ASCII_DIGIT_MASK) !== $current) {
                 // Validation error.
+                $context->logger?->warning('ipv4-in-ipv6-unexpected-code-point');
+
                 return false;
             }
 
@@ -204,6 +226,8 @@ private static function parseIPv4Address(
                 // Otherwise, if ipv4Piece is 0, validation error, return failure.
                 } elseif ($ipv4Piece === 0) {
                     // Validation error.
+                    $context->logger?->warning('ipv4-in-ipv6-invalid-first-part');
+
                     return false;
 
                 // Otherwise, set ipv4Piece to ipv4Piece × 10 + number.
@@ -214,6 +238,8 @@ private static function parseIPv4Address(
                 // 6.5.5.4.3. If ipv4Piece is greater than 255, validation error, return failure.
                 if ($ipv4Piece > 255) {
                     // Validation error.
+                    $context->logger?->warning('ipv4-in-ipv6-part-out-of-range');
+
                     return false;
                 }
 
@@ -238,6 +264,8 @@ private static function parseIPv4Address(
         // 6.5.6. If numbersSeen is not 4, validation error, return failure.
         if ($numbersSeen !== 4) {
             // Validation error.
+            $context->logger?->warning('ipv4-in-ipv6-too-few-parts');
+
             return false;
         }