<?php
/**
 * SeekQuarry/Yioop --
 * Open Source Pure PHP Search Engine, Crawler, and Indexer
 *
 * Copyright (C) 2009 - 2026  Chris Pollett chris@pollett.org
 *
 * LICENSE:
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <https://www.gnu.org/licenses/>.
 *
 * END LICENSE
 *
 * @author Chris Pollett chris@pollett.org
 * @license https://www.gnu.org/licenses/ GPL3
 * @link https://www.seekquarry.com/
 * @copyright 2009 - 2026
 * @filesource
 */
namespace seekquarry\yioop\library;

/**
 * Turns the handful of yes/no choices an administrator makes for one
 * domain into the text of that domain's robots.txt file. The choices
 * say whether to keep all crawlers out, how long crawlers should wait
 * between requests, and which kinds of pages (search results, group
 * feeds, wiki pages, and a wiki page's source/history, discussion, and
 * page listing) crawlers are allowed to visit. This class only builds
 * the text from values handed to it; it never reads the database, so a
 * controller or the front controller looks up a domain's saved choices
 * and passes them in.
 *
 * @author Chris Pollett
 */
class RobotsTxtGenerator
{
    /**
     * For each kind of page, the addresses that lead to it. Both the
     * clean-URL path and the plain query-string form are listed for
     * every kind, so a disallowed kind is kept out whether the site is
     * serving clean URLs or index.php query strings.
     */
    const CATEGORY_PATHS = [
        'search' => ['/s', '/*?*q=', '/*?*c=search'],
        'feeds' => ['/group', '/user', '/*?*c=group'],
        'wiki' => ['/p', '/*?*a=wiki'],
        'page_source' => ['/*?*arg=source', '/*?*arg=history'],
        'page_feed' => ['/thread', '/*?*just_thread='],
        'page_list' => ['/*?*arg=pages'],
    ];
    /**
     * The choices to assume for a domain that has never been
     * configured, so an unconfigured domain still serves a sensible
     * robots.txt. Everything is crawlable except search-result pages
     * (so crawlers do not walk every possible query), matching the
     * intent of the old shipped robots.txt.
     */
    const DEFAULT_CONFIG = [
        'disallow_all' => false,
        'crawl_delay' => 0,
        'search' => false,
        'feeds' => true,
        'wiki' => true,
        'page_source' => true,
        'page_feed' => true,
        'page_list' => true,
        'additional' => "",
    ];
    /**
     * Builds the complete robots.txt body for one domain from its saved
     * choices. When the choices say to keep all crawlers out, the body
     * is just a blanket disallow. Otherwise the body opens for all
     * crawlers, optionally asks for a crawl delay, adds a disallow line
     * for every address that leads to a kind of page the administrator
     * turned off, and finally appends any extra directives the
     * administrator typed in by hand.
     *
     * @param array $config the saved yes/no choices for the domain;
     *      missing choices fall back to the unconfigured defaults
     * @return string the complete robots.txt body, ending in a newline
     */
    public static function generate($config)
    {
        $config = array_merge(self::DEFAULT_CONFIG, is_array($config) ?
            $config : []);
        if (!empty($config['disallow_all'])) {
            return "User-agent: *\nDisallow: /\n";
        }
        $lines = ["User-agent: *"];
        $crawl_delay = intval($config['crawl_delay']);
        if ($crawl_delay > 0) {
            $lines[] = "Crawl-delay: " . $crawl_delay;
        }
        $blocked = [];
        if (empty($config['search'])) {
            $blocked[] = 'search';
        }
        if (empty($config['feeds'])) {
            $blocked[] = 'feeds';
        }
        if (empty($config['wiki'])) {
            $blocked[] = 'wiki';
        } else {
            if (empty($config['page_source'])) {
                $blocked[] = 'page_source';
            }
            if (empty($config['page_feed'])) {
                $blocked[] = 'page_feed';
            }
            if (empty($config['page_list'])) {
                $blocked[] = 'page_list';
            }
        }
        foreach ($blocked as $category) {
            foreach (self::CATEGORY_PATHS[$category] as $path) {
                $lines[] = "Disallow: " . $path;
            }
        }
        $text = implode("\n", $lines) . "\n";
        $additional = trim($config['additional'] ?? "");
        if ($additional !== "") {
            $text .= $additional . "\n";
        }
        return $text;
    }
}