package strategy;

import model.CrawlResult;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import exception.ParseException;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

public class CsdnBlogStrategy extends AbstractCrawlStrategy {
    private static final Logger logger = LoggerFactory.getLogger(CsdnBlogStrategy.class);

    private static final String SITE_NAME = "CSDN博客";

    private static final String[] CATEGORY_URLS = {
        "https://blog.csdn.net/nav/python?page=%d",
        "https://blog.csdn.net/nav/java?page=%d",
        "https://blog.csdn.net/nav/web?page=%d",
        "https://blog.csdn.net/nav/ai?page=%d",
        "https://blog.csdn.net/nav/database?page=%d",
        "https://blog.csdn.net/nav/ops?page=%d",
        "https://blog.csdn.net/nav/security?page=%d",
        "https://blog.csdn.net/nav/mobile?page=%d",
        "https://blog.csdn.net/nav/game?page=%d",
        "https://blog.csdn.net/nav/arch?page=%d"
    };

    private static final String[] CATEGORY_NAMES = {
        "Python", "Java", "Web", "人工智能", "数据库",
        "运维", "安全", "移动开发", "游戏", "架构"
    };

    @Override
    public String getBaseUrl() {
        return "https://blog.csdn.net/nav/python";
    }

    @Override
    public String getSiteName() {
        return SITE_NAME;
    }

    @Override
    public List<CrawlResult> crawlPage(int page) throws IOException, ParseException {
        List<CrawlResult> results = new ArrayList<>();

        int categoryIndex = (page - 1) / 5;
        int pageInCategory = (page - 1) % 5 + 1;

        if (categoryIndex >= CATEGORY_URLS.length) {
            logger.info("CSDN博客: 页码 {} 超出分类范围", page);
            return results;
        }

        String url = String.format(CATEGORY_URLS[categoryIndex], pageInCategory);
        String categoryName = CATEGORY_NAMES[categoryIndex];
        logger.info("正在爬取 CSDN {} 分类第 {} 页: {}", categoryName, pageInCategory, url);

        Document doc = fetchDocument(url);

        if (doc != null) {
            List<CrawlResult> parsed = parseCsdnArticles(doc, url);
            results.addAll(parsed);
            logger.info("CSDN {} 分类第 {} 页解析完成，获取 {} 条数据", categoryName, pageInCategory, parsed.size());
        }

        return results;
    }

    private List<CrawlResult> parseCsdnArticles(Document doc, String url) {
        List<CrawlResult> results = new ArrayList<>();

        Elements articleItems = doc.select(".article-item");
        if (articleItems.isEmpty()) {
            articleItems = doc.select(".feed_article");
        }
        if (articleItems.isEmpty()) {
            articleItems = doc.select(".list_article");
        }
        if (articleItems.isEmpty()) {
            articleItems = doc.select("div.article-item");
        }
        if (articleItems.isEmpty()) {
            articleItems = doc.select(".article-list");
        }
        if (articleItems.isEmpty()) {
            articleItems = doc.select("div.feed_article");
        }
        if (articleItems.isEmpty()) {
            Elements articleItems2 = doc.select("[class*=article]");
            for (Element item : articleItems2) {
                Element titleElem = item.selectFirst("h4 a");
                if (titleElem == null) titleElem = item.selectFirst("h3 a");
                if (titleElem == null) titleElem = item.selectFirst("a.article-title");
                if (titleElem == null) titleElem = item.selectFirst("a.title");
                if (titleElem != null) {
                    articleItems.add(item);
                }
            }
        }

        for (Element item : articleItems) {
            try {
                CrawlResult result = parseArticleItem(item, url);
                if (result != null && result.getTitle() != null && !result.getTitle().isEmpty()) {
                    results.add(result);
                }
            } catch (Exception e) {
                logger.debug("解析文章条目失败: {}", e.getMessage());
            }
        }

        if (results.isEmpty()) {
            Elements articleList = doc.select(".article-list li");
            for (Element li : articleList) {
                try {
                    Element titleElem = li.selectFirst("h2 a");
                    if (titleElem == null) titleElem = li.selectFirst("h3 a");
                    if (titleElem == null) titleElem = li.selectFirst(".title a");
                    if (titleElem == null) titleElem = li.selectFirst("a");

                    if (titleElem != null) {
                        String title = titleElem.text().trim();
                        String articleUrl = titleElem.attr("href");

                        Element descElem = li.selectFirst(".description");
                        if (descElem == null) descElem = li.selectFirst(".article-description");
                        if (descElem == null) descElem = li.selectFirst(".content");
                        String description = descElem != null ? descElem.text().trim() : "";

                        Element authorElem = li.selectFirst(".author");
                        if (authorElem == null) authorElem = li.selectFirst(".nick-name");
                        if (authorElem == null) authorElem = li.selectFirst("[class*=author]");
                        String author = authorElem != null ? authorElem.text().trim() : "CSDN用户";

                        if (!title.isEmpty()) {
                            CrawlResult result = new CrawlResult(
                                title,
                                0,
                                0,
                                10.0,
                                articleUrl,
                                author + " | " + description
                            );
                            results.add(result);
                        }
                    }
                } catch (Exception e) {
                    logger.debug("解析 li 文章失败: {}", e.getMessage());
                }
            }
        }

        Elements articles = doc.select(".article");
        for (Element article : articles) {
            try {
                Element titleElem = article.selectFirst("h4");
                if (titleElem == null) titleElem = article.selectFirst("h3");
                if (titleElem == null) titleElem = article.selectFirst(".article-title");
                if (titleElem == null) titleElem = article.selectFirst("a");

                String title = titleElem != null ? titleElem.text().trim() : "";
                String articleUrl = titleElem != null ? titleElem.attr("href") : "";

                Element descElem = article.selectFirst("p");
                if (descElem == null) descElem = article.selectFirst(".description");
                String description = descElem != null ? descElem.text().trim() : "";

                Element authorElem = article.selectFirst(".author");
                if (authorElem == null) authorElem = article.selectFirst(".nick-name");
                String author = authorElem != null ? authorElem.text().trim() : "CSDN用户";

                if (!title.isEmpty()) {
                    CrawlResult result = new CrawlResult(
                        title,
                        0,
                        0,
                        10.0,
                        articleUrl,
                        author + " | " + description
                    );
                    results.add(result);
                }
            } catch (Exception e) {
                logger.debug("解析 article 失败: {}", e.getMessage());
            }
        }

        return results;
    }

    private CrawlResult parseArticleItem(Element item, String url) {
        Element titleElem = item.selectFirst("h4 a");
        if (titleElem == null) titleElem = item.selectFirst("h3 a");
        if (titleElem == null) titleElem = item.selectFirst("a.article-title");
        if (titleElem == null) titleElem = item.selectFirst("a.title");
        if (titleElem == null) titleElem = item.selectFirst("a");

        if (titleElem == null) {
            return null;
        }

        String title = titleElem.text().trim();
        String articleUrl = titleElem.attr("href");

        if (title.isEmpty()) {
            return null;
        }

        Element descElem = item.selectFirst(".article-description");
        if (descElem == null) descElem = item.selectFirst(".description");
        if (descElem == null) descElem = item.selectFirst(".content");
        if (descElem == null) descElem = item.selectFirst("p");
        String description = descElem != null ? descElem.text().trim() : "";

        Element authorElem = item.selectFirst(".author");
        if (authorElem == null) authorElem = item.selectFirst(".nick-name");
        if (authorElem == null) authorElem = item.selectFirst(".user-name");
        if (authorElem == null) authorElem = item.selectFirst("[class*=author]");
        String author = authorElem != null ? authorElem.text().trim() : "CSDN用户";

        Element dateElem = item.selectFirst(".date");
        if (dateElem == null) dateElem = item.selectFirst(".time");
        if (dateElem == null) dateElem = item.selectFirst("[class*=date]");
        String date = dateElem != null ? dateElem.text().trim() : "";

        String extraInfo = author;
        if (!date.isEmpty()) {
            extraInfo += " | " + date;
        }
        if (!description.isEmpty()) {
            extraInfo += " | " + description;
        }

        return new CrawlResult(title, 0, 0, 10.0, articleUrl, extraInfo);
    }

    @Override
    public CrawlResult parseItem(Element element) throws ParseException {
        Element titleElem = element.selectFirst("h4 a");
        if (titleElem == null) titleElem = element.selectFirst("h3 a");
        if (titleElem == null) titleElem = element.selectFirst("a");

        String title = titleElem != null ? titleElem.text().trim() : "";
        String url = titleElem != null ? titleElem.attr("href") : "";

        if (title.isEmpty()) {
            return null;
        }

        Element authorElem = element.selectFirst(".author");
        String author = authorElem != null ? authorElem.text().trim() : "CSDN用户";

        return new CrawlResult(title, 0, 0, 10.0, url, author);
    }

    @Override
    public int getPageSize() {
        return 15;
    }

    @Override
    protected String getReferer(String url) {
        return "https://blog.csdn.net/";
    }
}