Merge pull request #121 from bogdancondorachi/python-language

Add python support

src/Highlighter.php

@@ -17,6 +17,7 @@
 use Tempest\Highlight\Languages\JavaScript\JavaScriptLanguage;
 use Tempest\Highlight\Languages\Json\JsonLanguage;
 use Tempest\Highlight\Languages\Php\PhpLanguage;
+use Tempest\Highlight\Languages\Python\PythonLanguage;
 use Tempest\Highlight\Languages\Sql\SqlLanguage;
 use Tempest\Highlight\Languages\Text\TextLanguage;
 use Tempest\Highlight\Languages\Twig\TwigLanguage;
@@ -49,6 +50,7 @@ public function __construct(
             ->addLanguage(new JavaScriptLanguage())
             ->addLanguage(new JsonLanguage())
             ->addLanguage(new PhpLanguage())
+            ->addLanguage(new PythonLanguage())
             ->addLanguage(new SqlLanguage())
             ->addLanguage(new XmlLanguage())
             ->addLanguage(new YamlLanguage())

src/Languages/Python/Patterns/PyArgumentPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyArgumentPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '(?<=,|\()\s*(?<match>\w+)s*=';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::VARIABLE;
+    }
+}

src/Languages/Python/Patterns/PyBooleanPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyBooleanPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '\b(?<match>(?:False|None|True))\b';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::BOOLEAN;
+    }
+}

src/Languages/Python/Patterns/PyBuiltinPattern.php

@@ -0,0 +1,40 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyBuiltinPattern implements Pattern
+{
+    use IsPattern;
+
+    public function __construct(private array $builtinFunctions = [
+        '__import__', 'abs', 'aiter', 'all', 'any', 'anext', 'ascii', 'bin', 'bool',
+        'breakpoint', 'bytearray', 'bytes', 'callable', 'chr', 'classmethod', 'compile',
+        'complex', 'delattr', 'dict', 'dir', 'divmod', 'enumerate', 'eval', 'exec',
+        'filter', 'float', 'format', 'frozenset', 'getattr', 'globals', 'hasattr',
+        'hash', 'help', 'hex', 'id', 'input', 'int', 'isinstance', 'issubclass', 'iter',
+        'len', 'list', 'locals', 'map', 'max', 'memoryview', 'min', 'next', 'object',
+        'oct', 'open', 'ord', 'pow', 'print', 'property', 'range', 'repr', 'reversed',
+        'round', 'set', 'setattr', 'slice', 'sorted', 'staticmethod', 'str', 'sum',
+        'super', 'tuple', 'type', 'vars', 'zip',
+    ])
+    {
+    }
+
+    public function getPattern(): string
+    {
+        $builtinFunctions = implode('|', $this->builtinFunctions);
+
+        return "\b(?<match>(?:{$builtinFunctions}))\b";
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::TYPE;
+    }
+}

src/Languages/Python/Patterns/PyClassNamePattern.php

@@ -0,0 +1,27 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\PatternTest;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+#[PatternTest(input: 'class MyClass:', output: 'MyClass')]
+#[PatternTest(input: 'class HisClass(MyClass):', output: 'HisClass')]
+final readonly class PyClassNamePattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '\bclass\s+(?<match>\w*)(?=[\s*\:(])';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::PROPERTY;
+    }
+}

src/Languages/Python/Patterns/PyCommentPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyCommentPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '(^|[^\\\\])(?<match>#.*)';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::COMMENT;
+    }
+}

src/Languages/Python/Patterns/PyDecoratorPattern.php

@@ -0,0 +1,27 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\PatternTest;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+#[PatternTest(input: '@decorator', output: '@decorator')]
+#[PatternTest(input: '@decorator.chained', output: '@decorator.chained')]
+final readonly class PyDecoratorPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '(^|\n)\s*(?<match>@\s*\w*(?:\.\w+)*)';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::PROPERTY;
+    }
+}

src/Languages/Python/Patterns/PyFunctionPattern.php

@@ -0,0 +1,26 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\PatternTest;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+#[PatternTest(input: 'def fibonacci(n)', output: 'fibonacci')]
+final readonly class PyFunctionPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '\bdef\s+(?<match>\w*)(?=\s*\()';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::PROPERTY;
+    }
+}

src/Languages/Python/Patterns/PyKeywordPattern.php

@@ -0,0 +1,35 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyKeywordPattern implements Pattern
+{
+    use IsPattern;
+
+    public function __construct(private array $keywords = [
+        'and', 'as', 'assert', 'async', 'await', 'break', 'class', 'continue',
+        'def', 'del', 'elif', 'else', 'except', 'finally', 'for', 'from', 'global',
+        'if', 'import', 'in', 'is', 'lambda', 'nonlocal', 'not', 'or', 'pass',
+        'raise', 'return', 'try', 'while', 'with', 'yield',
+    ])
+    {
+    }
+
+    public function getPattern(): string
+    {
+        $keywords = implode('|', $this->keywords);
+
+        return "\b(?<match>(?:_(?=\s*:){$keywords}))\b";
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::KEYWORD;
+    }
+}

src/Languages/Python/Patterns/PyNumberPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyNumberPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '(?<match>\b0(?:[bB](?:_?[01])+|[oO](?:_?[0-7])+|[xX](?:_?[a-fA-F0-9])+)\b|(?:\b\d+(?:_\d+)*(?:\.(?:\d+(?:_\d+)*)?)?|\B\.\d+(?:_\d+)*)(?:[eE][+-]?\d+(?:_\d+)*)?j?(?!\w))';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::NUMBER;
+    }
+}

src/Languages/Python/Patterns/PyOperatorPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyOperatorPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return "(?<match>([-+&%=]=?|!=|:=|>>=|<<=|\|=|\^=|\*\*?=?|\/\/?=?|<[<=]?|>[=>]?|[\|^~]))";
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::OPERATOR;
+    }
+}

src/Languages/Python/Patterns/PyTripleDoubleQuoteStringPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyTripleDoubleQuoteStringPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '/(?<match>"""(.|\n)*?""")/m';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::VALUE;
+    }
+}

src/Languages/Python/Patterns/PyTripleSingleQuoteStringPattern.php

@@ -0,0 +1,24 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tempest\Highlight\Languages\Python\Patterns;
+
+use Tempest\Highlight\IsPattern;
+use Tempest\Highlight\Pattern;
+use Tempest\Highlight\Tokens\TokenTypeEnum;
+
+final readonly class PyTripleSingleQuoteStringPattern implements Pattern
+{
+    use IsPattern;
+
+    public function getPattern(): string
+    {
+        return '/(?<match>\'\'\'(.|\n)*?\'\'\')/m';
+    }
+
+    public function getTokenType(): TokenTypeEnum
+    {
+        return TokenTypeEnum::VALUE;
+    }
+}