<?php

declare(strict_types=1);

namespace AilysLens\Core;

use InvalidArgumentException;
use RuntimeException;

/**
 * crawlers.json を読み込み、スキーマ検証済みの定義一覧を提供する。
 */
final class DefinitionRepository
{
    /**
     * 例外メッセージへ埋め込む動的値の引用・エスケープ用フラグ。
     * プラグイン同梱コピーが管理画面に表示され得るため、HTML特殊文字を
     * \uXXXX形式に逃がしつつ引用符で値を区切る。coreのCMS非依存は維持(純PHPのみ)。
     */
    private const MESSAGE_VALUE_FLAGS = JSON_HEX_TAG | JSON_HEX_APOS | JSON_HEX_QUOT | JSON_HEX_AMP
        | JSON_UNESCAPED_SLASHES | JSON_INVALID_UTF8_SUBSTITUTE;

    /** @var array<int, CrawlerDefinition> */
    private array $crawlers;

    /** @var array<int, array<string, string>> */
    private array $robotsTxtTokens;

    private int $schemaVersion;

    private string $updatedAt;

    /**
     * @param array<int, CrawlerDefinition>     $crawlers
     * @param array<int, array<string, string>> $robotsTxtTokens
     */
    private function __construct(array $crawlers, array $robotsTxtTokens, int $schemaVersion, string $updatedAt)
    {
        $this->crawlers        = $crawlers;
        $this->robotsTxtTokens = $robotsTxtTokens;
        $this->schemaVersion   = $schemaVersion;
        $this->updatedAt       = $updatedAt;
    }

    public static function defaultFilePath(): string
    {
        return dirname(__DIR__) . '/data/crawlers.json';
    }

    public static function fromDefaultFile(): self
    {
        return self::fromFile(self::defaultFilePath());
    }

    public static function fromFile(string $path): self
    {
        if (!is_readable($path)) {
            throw new RuntimeException('Cannot read crawlers.json: ' . json_encode($path, self::MESSAGE_VALUE_FLAGS));
        }

        $raw = file_get_contents($path);
        if ($raw === false) {
            throw new RuntimeException('Failed to read crawlers.json: ' . json_encode($path, self::MESSAGE_VALUE_FLAGS));
        }

        $decoded = json_decode($raw, true);
        if (!is_array($decoded)) {
            throw new InvalidArgumentException('crawlers.json contains invalid JSON: ' . json_encode(json_last_error_msg(), self::MESSAGE_VALUE_FLAGS));
        }

        return self::fromArray($decoded);
    }

    /**
     * @param array<string, mixed> $data
     */
    public static function fromArray(array $data): self
    {
        if (!isset($data['schema_version']) || !is_int($data['schema_version'])) {
            throw new InvalidArgumentException('Missing or non-integer schema_version');
        }
        if ($data['schema_version'] !== 1) {
            throw new InvalidArgumentException('Unsupported schema_version: ' . json_encode($data['schema_version'], self::MESSAGE_VALUE_FLAGS));
        }
        if (!isset($data['crawlers']) || !is_array($data['crawlers'])) {
            throw new InvalidArgumentException('Missing crawlers array');
        }

        $crawlers = [];
        $seenIds  = [];
        foreach ($data['crawlers'] as $index => $entry) {
            if (!is_array($entry)) {
                throw new InvalidArgumentException('crawlers[' . json_encode($index, self::MESSAGE_VALUE_FLAGS) . '] is not an object');
            }
            $crawler = self::buildCrawler($entry, (string) $index);
            if (isset($seenIds[$crawler->id])) {
                throw new InvalidArgumentException('Duplicate crawler id: ' . json_encode($crawler->id, self::MESSAGE_VALUE_FLAGS));
            }
            $seenIds[$crawler->id] = true;
            $crawlers[]            = $crawler;
        }

        $tokens = [];
        foreach ($data['robots_txt_tokens'] ?? [] as $index => $entry) {
            if (!is_array($entry) || !isset($entry['id'], $entry['vendor'], $entry['token'])) {
                throw new InvalidArgumentException('robots_txt_tokens[' . json_encode($index, self::MESSAGE_VALUE_FLAGS) . '] is missing required fields (id, vendor, token)');
            }
            $tokens[] = array_map('strval', $entry);
        }

        return new self(
            $crawlers,
            $tokens,
            $data['schema_version'],
            isset($data['updated_at']) && is_string($data['updated_at']) ? $data['updated_at'] : ''
        );
    }

    /**
     * @param array<string, mixed> $entry
     */
    private static function buildCrawler(array $entry, string $where): CrawlerDefinition
    {
        foreach (['id', 'vendor', 'ua_pattern', 'category', 'verify'] as $key) {
            if (!isset($entry[$key])) {
                throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] is missing ' . json_encode($key, self::MESSAGE_VALUE_FLAGS));
            }
        }
        if (!array_key_exists('docs_url', $entry)) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] is missing docs_url (use an explicit empty string if unpublished)');
        }

        $category = (string) $entry['category'];
        if (!in_array($category, CrawlerDefinition::CATEGORIES, true)) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] has an invalid category: ' . json_encode($category, self::MESSAGE_VALUE_FLAGS));
        }

        $verify = $entry['verify'];
        if (!is_array($verify) || !isset($verify['method'])) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] is missing verify.method');
        }
        if (!in_array($verify['method'], CrawlerDefinition::VERIFY_METHODS, true)) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] has an invalid verify.method: ' . json_encode($verify['method'], self::MESSAGE_VALUE_FLAGS));
        }
        if ($verify['method'] === CrawlerDefinition::VERIFY_RDNS && empty($verify['domains'])) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] uses rdns verification but has no domains');
        }
        if ($verify['method'] === CrawlerDefinition::VERIFY_IP_LIST && empty($verify['ip_list_url'])) {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] uses ip_list verification but has no ip_list_url');
        }

        $uaPattern = (string) $entry['ua_pattern'];
        if ($uaPattern === '') {
            throw new InvalidArgumentException('crawlers[' . json_encode($where, self::MESSAGE_VALUE_FLAGS) . '] has an empty ua_pattern');
        }

        /** @var array{method: string, ip_list_url?: string, domains?: array<int, string>} $verify */
        return new CrawlerDefinition(
            (string) $entry['id'],
            (string) $entry['vendor'],
            $uaPattern,
            $category,
            $verify,
            (string) $entry['docs_url'],
            isset($entry['note']) ? (string) $entry['note'] : null
        );
    }

    /**
     * @return array<int, CrawlerDefinition>
     */
    public function crawlers(): array
    {
        return $this->crawlers;
    }

    public function findById(string $id): ?CrawlerDefinition
    {
        foreach ($this->crawlers as $crawler) {
            if ($crawler->id === $id) {
                return $crawler;
            }
        }

        return null;
    }

    /**
     * @return array<int, array<string, string>>
     */
    public function robotsTxtTokens(): array
    {
        return $this->robotsTxtTokens;
    }

    public function schemaVersion(): int
    {
        return $this->schemaVersion;
    }

    public function updatedAt(): string
    {
        return $this->updatedAt;
    }
}