import { absoluteUrl } from "./url.js";
import type { RobotsTxtInput, SiteSeoConfig } from "./types.js";

/** 站点默认 robots.txt：允许全站抓取 + 指向 sitemap。 */
export function buildDefaultRobotsInput(site: Pick<SiteSeoConfig, "siteUrl">): RobotsTxtInput {
    return {
        rules: [
            {
                userAgent: "*",
                allow: ["/"],
                disallow: ["/api/", "/dashboard"],
            },
        ],
        sitemap: [absoluteUrl(site.siteUrl, "/sitemap.xml")],
        host: site.siteUrl,
    };
}

/** 生成 robots.txt 纯文本。 */
export function buildRobotsTxt(input: RobotsTxtInput): string {
    const lines: string[] = [];

    if (input.host) {
        lines.push(`Host: ${input.host.replace(/\/+$/, "")}`);
    }

    const rules = input.rules ?? [];
    for (const rule of rules) {
        if (lines.length > 0) lines.push("");
        lines.push(`User-agent: ${rule.userAgent}`);
        for (const a of rule.allow ?? []) {
            lines.push(`Allow: ${a}`);
        }
        for (const d of rule.disallow ?? []) {
            // 空 Disallow 表示允许全部，跳过空串避免误导
            if (d === "") continue;
            lines.push(`Disallow: ${d}`);
        }
        if (rule.crawlDelay !== undefined) {
            lines.push(`Crawl-delay: ${rule.crawlDelay}`);
        }
    }

    lines.push("");
    for (const s of input.sitemap ?? []) {
        lines.push(`Sitemap: ${s}`);
    }
    lines.push("");
    return lines.join("\n");
}

/**
 * 抓取预算辅助：分页页默认 noindex + canonical 指向首页，
 * 避免无限分页稀释抓取配额。
 */
export function paginationRobots(
    pageNumber: number,
): Array<"index" | "noindex" | "follow" | "nofollow"> {
    if (pageNumber <= 1) return ["index", "follow"];
    return ["noindex", "follow"];
}

/** 内部/预发布页：noindex + nofollow。 */
export const DRAFT_ROBOTS = ["noindex", "nofollow", "noarchive"] as const;