Maintenance work on library

.github/workflows/php-tests.yml

@@ -1,8 +1,8 @@
 name: PHP Tests
 
 on:
+  pull_request:
   push:
-    branches: [ "master" ]
 
 permissions:
   contents: read
@@ -14,7 +14,7 @@ jobs:
     strategy:
       matrix:
         operating-system: [ ubuntu-latest ]
-        php-version: [ '8.0', '8.1', '8.2' ]
+        php-version: ['8.2', '8.3']
 
     name: PHP ${{ matrix.php-version }} test on ${{ matrix.operating-system }}
 

composer.json

@@ -35,21 +35,22 @@
         "php": "^8.2.0"
     },
     "require-dev": {
-        "symfony/console": ">=6.0",
-        "phpunit/phpunit": "^11.0.3"
+        "phpunit/phpunit": "^11.0.3",
+        "rector/rector": "^1.0",
+        "symfony/console": ">=6.0"
     },
     "suggest": {
         "symfony/console": "for the command-line tool"
     },
     "autoload": {
-        "psr-0": {
+        "psr-4": {
             "Dissect\\": [
                 "src/"
             ]
         }
     },
     "autoload-dev": {
-        "psr-0": {
+        "psr-4": {
             "Dissect\\": [
                 "tests/"
             ]

docs/lexing.md

@@ -236,7 +236,7 @@ Now that we've demonstrated how to perform lexical analysis with
 Dissect, we can move onto syntactical analysis, commonly known as
 [parsing][parsing].
 
-[tokenstream]: ../src/Dissect/Lexer/TokenStream/TokenStream.php
+[tokenstream]: ../src/Lexer/TokenStream/TokenStream.php
 [parsing]: parsing.md
 [doctrinelexer]: https://github.com/doctrine/lexer/blob/master/lib/Doctrine/Common/Lexer/AbstractLexer.php
 [doctrine]: https://github.com/doctrine/lexer

rector.php

@@ -0,0 +1,37 @@
+<?php
+
+declare(strict_types=1);
+
+use Rector\Config\RectorConfig;
+use Rector\DeadCode\Rector\ClassMethod\RemoveUselessParamTagRector;
+use Rector\DeadCode\Rector\ClassMethod\RemoveUselessReturnTagRector;
+use Rector\DeadCode\Rector\Property\RemoveUselessVarTagRector;
+use Rector\Php71\Rector\ClassConst\PublicConstantVisibilityRector;
+use Rector\Php74\Rector\Closure\ClosureToArrowFunctionRector;
+use Rector\PHPUnit\Set\PHPUnitSetList;
+use Rector\TypeDeclaration\Rector\Class_\AddTestsVoidReturnTypeWhereNoReturnRector;
+use Rector\TypeDeclaration\Rector\ClassMethod\AddVoidReturnTypeWhereNoReturnRector;
+use Rector\TypeDeclaration\Rector\StmtsAwareInterface\DeclareStrictTypesRector;
+
+return RectorConfig::configure()
+    ->withPaths([
+        __DIR__ . '/src',
+        __DIR__ . '/tests',
+    ])
+    // uncomment to reach your current PHP version
+    // ->withPhpSets()
+    ->withRules([
+        // Dead-code
+        RemoveUselessParamTagRector::class,
+        RemoveUselessReturnTagRector::class,
+        RemoveUselessVarTagRector::class,
+        PublicConstantVisibilityRector::class,
+        ClosureToArrowFunctionRector::class,
+        AddVoidReturnTypeWhereNoReturnRector::class,
+        AddTestsVoidReturnTypeWhereNoReturnRector::class,
+        // DeclareStrictTypesRector::class,
+    ])
+    ->withSets([
+        PHPUnitSetList::ANNOTATIONS_TO_ATTRIBUTES,
+        PHPUnitSetList::PHPUNIT_CODE_QUALITY,
+    ]);

src/Dissect/Console/Application.php → src/Console/Application.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Console;
 
 use Symfony\Component\Console\Application as BaseApplication;

src/Dissect/Console/Command/DissectCommand.php → src/Console/Command/DissectCommand.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Console\Command;
 
 use Dissect\Parser\LALR1\Analysis\Exception\ConflictException;
@@ -27,28 +29,28 @@ protected function configure()
             ->addOption('state', 's', InputOption::VALUE_REQUIRED, 'Exports only the specified state instead of the entire DFA.')
             ->addOption('output-dir', 'o', InputOption::VALUE_REQUIRED, 'Overrides the default output directory.')
             ->setHelp(<<<EOT
-Analyzes the given grammar and, if successful, exports the parse table to a PHP
-file.
-
-By default, the output directory is taken to be the one in which the grammar is
-defined. You can change that with the <info>--output-dir</info> option:
-
- <info>--output-dir=../some/other/dir</info>
-
-The parse table is by default written with minimal whitespace to make it compact.
-If you wish to inspect the table manually, you can export it in a readable and
-well-commented way with the <info>--debug</info> option.
-
-If you wish to inspect the handle-finding automaton for your grammar (perhaps
-to aid with grammar debugging), use the <info>--dfa</info> option. When in use, Dissect
-will create a file with the automaton exported as a Graphviz graph
-in the output directory.
-
-Additionally, you can use the <info>--state</info> option to export only the specified
-state and any relevant transitions:
-
- <info>--dfa --state=5</info>
-EOT
+                Analyzes the given grammar and, if successful, exports the parse table to a PHP
+                file.
+                
+                By default, the output directory is taken to be the one in which the grammar is
+                defined. You can change that with the <info>--output-dir</info> option:
+                
+                 <info>--output-dir=../some/other/dir</info>
+                
+                The parse table is by default written with minimal whitespace to make it compact.
+                If you wish to inspect the table manually, you can export it in a readable and
+                well-commented way with the <info>--debug</info> option.
+                
+                If you wish to inspect the handle-finding automaton for your grammar (perhaps
+                to aid with grammar debugging), use the <info>--dfa</info> option. When in use, Dissect
+                will create a file with the automaton exported as a Graphviz graph
+                in the output directory.
+                
+                Additionally, you can use the <info>--state</info> option to export only the specified
+                state and any relevant transitions:
+                
+                 <info>--dfa --state=5</info>
+                EOT
             );
     }
 

src/Dissect/Lexer/AbstractLexer.php → src/Lexer/AbstractLexer.php

@@ -1,9 +1,12 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer;
 
 use Dissect\Lexer\Exception\RecognitionException;
 use Dissect\Lexer\TokenStream\ArrayTokenStream;
+use Dissect\Lexer\TokenStream\TokenStream;
 use Dissect\Parser\Parser;
 use Dissect\Util\Util;
 
@@ -13,18 +16,14 @@
  * SimpleLexer and StatefulLexer extend this class.
  *
  * @author Jakub Lédl <[email protected]>
+ * @see \Dissect\Lexer\AbstractLexerTest
  */
 abstract class AbstractLexer implements Lexer
 {
-    /**
-     * @var int
-     */
     private int $line = 1;
 
     /**
      * Returns the current line.
-     *
-     * @return int The current line.
      */
     protected function getCurrentLine(): int
     {
@@ -36,24 +35,18 @@ protected function getCurrentLine(): int
      * Returns the token on success or null on failure.
      *
      * @param string $string The string to extract the token from.
-     *
-     * @return Token|null The extracted token or null.
      */
     abstract protected function extractToken(string $string): ?Token;
 
     /**
      * Should given token be skipped?
-     *
-     * @param Token $token The token to evaluate.
-     *
-     * @return boolean Whether to skip the token.
      */
     abstract protected function shouldSkipToken(Token $token): bool;
 
     /**
      * {@inheritDoc}
      */
-    public function lex(string $string): TokenStream\TokenStream|ArrayTokenStream
+    public function lex(string $string): TokenStream
     {
         // normalize line endings
         $string = strtr($string, ["\r\n" => "\n", "\r" => "\n"]);
@@ -92,6 +85,6 @@ public function lex(string $string): TokenStream\TokenStream|ArrayTokenStream
 
         $tokens[] = new CommonToken(Parser::EOF_TOKEN_TYPE, '', $this->line);
 
-        return new ArrayTokenStream($tokens);
+        return new ArrayTokenStream(...$tokens);
     }
 }

src/Dissect/Lexer/CommonToken.php → src/Lexer/CommonToken.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer;
 
 /**
@@ -13,12 +15,12 @@ class CommonToken implements Token
      * Constructor.
      *
      * @param mixed $type The type of the token.
-     * @param string $value The token value.
+     * @param int|string $value The token value.
      * @param int $line The line.
      */
     public function __construct(
         protected mixed $type,
-        protected string $value,
+        protected int|string $value,
         protected int $line
     ) {}
 
@@ -33,7 +35,7 @@ public function getType(): mixed
     /**
      * {@inheritDoc}
      */
-    public function getValue(): string
+    public function getValue(): int|string
     {
         return $this->value;
     }

src/Dissect/Lexer/Exception/RecognitionException.php → src/Lexer/Exception/RecognitionException.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer\Exception;
 
 use RuntimeException;

src/Dissect/Lexer/Lexer.php → src/Lexer/Lexer.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer;
 
 use Dissect\Lexer\Exception\RecognitionException;
@@ -19,8 +21,6 @@ interface Lexer
      * @param string $string The string to lex.
      *
      * @throws RecognitionException When unable to extract more tokens from the string.
-     *
-     * @return TokenStream The resulting token stream.
      */
     public function lex(string $string): TokenStream;
 }

src/Dissect/Lexer/Recognizer/Recognizer.php → src/Lexer/Recognizer/Recognizer.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer\Recognizer;
 
 /**

src/Dissect/Lexer/Recognizer/RegexRecognizer.php → src/Lexer/Recognizer/RegexRecognizer.php

@@ -1,12 +1,15 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer\Recognizer;
 
 /**
  * The RegexRecognizer matches a string using a
  * regular expression.
  *
  * @author Jakub Lédl <[email protected]>
+ * @see \Dissect\Lexer\Recognizer\RegexRecognizerTest
  */
 class RegexRecognizer implements Recognizer
 {

src/Dissect/Lexer/Recognizer/SimpleRecognizer.php → src/Lexer/Recognizer/SimpleRecognizer.php

@@ -1,12 +1,15 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer\Recognizer;
 
 /**
  * SimpleRecognizer matches a string by a simple
  * strpos match.
  *
  * @author Jakub Lédl <[email protected]>
+ * @see \Dissect\Lexer\Recognizer\SimpleRecognizerTest
  */
 class SimpleRecognizer implements Recognizer
 {

src/Dissect/Lexer/RegexLexer.php → src/Lexer/RegexLexer.php

@@ -1,5 +1,7 @@
 <?php
 
+declare(strict_types=1);
+
 namespace Dissect\Lexer;
 
 use Dissect\Lexer\TokenStream\ArrayTokenStream;
@@ -13,6 +15,7 @@
  * @author Jonathan Wage <[email protected]>
  * @author Roman Borschel <[email protected]>
  * @author Jakub Lédl <[email protected]>
+ * @see \Dissect\Lexer\RegexLexerTest
  */
 abstract class RegexLexer implements Lexer
 {
@@ -52,29 +55,21 @@ public function lex(string $string): TokenStream
 
         $tokens[] = new CommonToken(Parser::EOF_TOKEN_TYPE, '', $line);
 
-        return new ArrayTokenStream($tokens);
+        return new ArrayTokenStream(...$tokens);
     }
 
     /**
      * The patterns corresponding to tokens.
-     *
-     * @return array
      */
     abstract protected function getCatchablePatterns(): array;
 
     /**
      * The patterns corresponding to tokens to be skipped.
-     *
-     * @return array
      */
     abstract protected function getNonCatchablePatterns(): array;
 
     /**
      * Retrieves the token type.
-     *
-     * @param string $value
-     *
-     * @return string $type
      */
     abstract protected function getType(string &$value): string;
 }