php-parser/bin/php-parse

#!/usr/bin/env php
<?php

foreach ([__DIR__ . '/../../../autoload.php', __DIR__ . '/../vendor/autoload.php'] as $file) {
    if (file_exists($file)) {
        require $file;
        break;
    }
}

ini_set('xdebug.max_nesting_level', 3000);

// Disable XDebug var_dump() output truncation
ini_set('xdebug.var_display_max_children', -1);
ini_set('xdebug.var_display_max_data', -1);
ini_set('xdebug.var_display_max_depth', -1);

list($operations, $files, $attributes) = parseArgs($argv);

/* Dump nodes by default */
if (empty($operations)) {
    $operations[] = 'dump';
}

if (empty($files)) {
    showHelp("Must specify at least one file.");
}

$lexer = new PhpParser\Lexer\Emulative(['usedAttributes' => [
    'startLine', 'endLine', 'startFilePos', 'endFilePos', 'comments'
]]);
$parser = (new PhpParser\ParserFactory)->create(
    PhpParser\ParserFactory::PREFER_PHP7,
    $lexer
);
$dumper = new PhpParser\NodeDumper([
    'dumpComments' => true,
    'dumpPositions' => $attributes['with-positions'],
]);
$prettyPrinter = new PhpParser\PrettyPrinter\Standard;

$traverser = new PhpParser\NodeTraverser();
$traverser->addVisitor(new PhpParser\NodeVisitor\NameResolver);

foreach ($files as $file) {
    if (strpos($file, '<?php') === 0) {
        $code = $file;
        echo "====> Code $code\n";
    } else {
        if (!file_exists($file)) {
            die("File $file does not exist.\n");
        }

        $code = file_get_contents($file);
        echo "====> File $file:\n";
    }

    if ($attributes['with-recovery']) {
        $errorHandler = new PhpParser\ErrorHandler\Collecting;
        $stmts = $parser->parse($code, $errorHandler);
        foreach ($errorHandler->getErrors() as $error) {
            $message = formatErrorMessage($error, $code, $attributes['with-column-info']);
            echo $message . "\n";
        }
        if (null === $stmts) {
            continue;
        }
    } else {
        try {
            $stmts = $parser->parse($code);
        } catch (PhpParser\Error $error) {
            $message = formatErrorMessage($error, $code, $attributes['with-column-info']);
            die($message . "\n");
        }
    }

    foreach ($operations as $operation) {
        if ('dump' === $operation) {
            echo "==> Node dump:\n";
            echo $dumper->dump($stmts, $code), "\n";
        } elseif ('pretty-print' === $operation) {
            echo "==> Pretty print:\n";
            echo $prettyPrinter->prettyPrintFile($stmts), "\n";
        } elseif ('json-dump' === $operation) {
            echo "==> JSON dump:\n";
            echo json_encode($stmts, JSON_PRETTY_PRINT), "\n";
        } elseif ('var-dump' === $operation) {
            echo "==> var_dump():\n";
            var_dump($stmts);
        } elseif ('resolve-names' === $operation) {
            echo "==> Resolved names.\n";
            $stmts = $traverser->traverse($stmts);
        }
    }
}

function formatErrorMessage(PhpParser\Error $e, $code, $withColumnInfo) {
    if ($withColumnInfo && $e->hasColumnInfo()) {
        return $e->getMessageWithColumnInfo($code);
    } else {
        return $e->getMessage();
    }
}

function showHelp($error = '') {
    if ($error) {
        echo $error . "\n\n";
    }
    die(<<<OUTPUT
Usage: php-parse [operations] file1.php [file2.php ...]
   or: php-parse [operations] "<?php code"
Turn PHP source code into an abstract syntax tree.

Operations is a list of the following options (--dump by default):

    -d, --dump              Dump nodes using NodeDumper
    -p, --pretty-print      Pretty print file using PrettyPrinter\Standard
    -j, --json-dump         Print json_encode() result
        --var-dump          var_dump() nodes (for exact structure)
    -N, --resolve-names     Resolve names using NodeVisitor\NameResolver
    -c, --with-column-info  Show column-numbers for errors (if available)
    -P, --with-positions    Show positions in node dumps
    -r, --with-recovery     Use parsing with error recovery
    -h, --help              Display this page

Example:
    php-parse -d -p -N -d file.php

    Dumps nodes, pretty prints them, then resolves names and dumps them again.


OUTPUT
    );
}

function parseArgs($args) {
    $operations = [];
    $files = [];
    $attributes = [
        'with-column-info' => false,
        'with-positions' => false,
        'with-recovery' => false,
    ];

    array_shift($args);
    $parseOptions = true;
    foreach ($args as $arg) {
        if (!$parseOptions) {
            $files[] = $arg;
            continue;
        }

        switch ($arg) {
            case '--dump':
            case '-d':
                $operations[] = 'dump';
                break;
            case '--pretty-print':
            case '-p':
                $operations[] = 'pretty-print';
                break;
            case '--json-dump':
            case '-j':
                $operations[] = 'json-dump';
                break;
            case '--var-dump':
                $operations[] = 'var-dump';
                break;
            case '--resolve-names':
            case '-N';
                $operations[] = 'resolve-names';
                break;
            case '--with-column-info':
            case '-c';
                $attributes['with-column-info'] = true;
                break;
            case '--with-positions':
            case '-P':
                $attributes['with-positions'] = true;
                break;
            case '--with-recovery':
            case '-r':
                $attributes['with-recovery'] = true;
                break;
            case '--help':
            case '-h';
                showHelp();
                break;
            case '--':
                $parseOptions = false;
                break;
            default:
                if ($arg[0] === '-') {
                    showHelp("Invalid operation $arg.");
                } else {
                    $files[] = $arg;
                }
        }
    }

    return [$operations, $files, $attributes];
}