long-article-manage/long-article-server/src/main/java/com/tzld/piaoquan/longarticle/utils/XiGuaUtils.java

package com.tzld.piaoquan.longarticle.utils;

import cn.hutool.http.HttpRequest;
import cn.hutool.http.HttpResponse;
import com.alibaba.fastjson.JSONArray;
import com.alibaba.fastjson.JSONObject;
import com.fasterxml.jackson.databind.JsonNode;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.select.Elements;

import javax.crypto.Cipher;
import javax.crypto.spec.IvParameterSpec;
import javax.crypto.spec.SecretKeySpec;
import java.net.URLEncoder;
import java.nio.charset.StandardCharsets;
import java.util.*;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class XiGuaUtils {

    public static List<JSONObject> xiguaSearchV2(String keyword, List<String> sensitiveWords) {
        String url = "https://www.ixigua.com/api/searchv2/complex/" + URLEncoder.encode(keyword, StandardCharsets.UTF_8) + "/10";
        Map<String, String> headers = new HashMap<>();
        headers.put("accept", "application/json, text/plain, */*");
        headers.put("accept-language", "en,zh;q=0.9,zh-CN;q=0.8");
        headers.put("cookie", "_tea_utm_cache_2285=undefined;");
        headers.put("priority", "u=1, i");
        headers.put("referer", "https://www.ixigua.com/search/" + URLEncoder.encode(keyword, StandardCharsets.UTF_8));
        headers.put("user-agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3");

        HttpResponse response = HttpRequest.get(url)
                .addHeaders(headers)
                .execute();

        try {
            JSONObject jsonResponse = JSONObject.parseObject(response.body());
            JSONArray recallList = jsonResponse.getJSONObject("data").getJSONArray("data");

            if (recallList != null) {
                for (int i = 0; i < recallList.size(); i++) {
                    JSONObject videoObj = recallList.getJSONObject(i);
                    if ("video".equals(videoObj.getString("type"))) {
                        String title = videoObj.getJSONObject("data").getString("title");
                        String groupId = videoObj.getJSONObject("data").getString("group_id");
                        int duration = videoObj.getJSONObject("data").getInteger("video_time");

                        // 假设 sensitiveFlag 是一个方法用于检查敏感词
                        if (sensitiveFlag(sensitiveWords, title) && duration <= 300) {
                            System.out.println(groupId);
                            JSONObject videoInfo = getVideoInfo(groupId);
                            if (videoInfo != null) {
                                return List.of(videoInfo);
                            }
                        }
                    }
                }
            }
            return List.of();
        } catch (Exception e) {
            e.printStackTrace();
            return List.of();
        }
    }

    public static JSONObject getVideoInfo(String itemId) {
        String url = "https://www.ixigua.com/" + itemId;

        // 设置请求头
        HttpRequest request = HttpRequest.get(url)
                .header("accept-language", "zh-CN,zh-Hans;q=0.9")
                .header("cookie", "UIFID=73355a799e41c2edb6d004baa6cda0116425031dff9117e11075ec8bf266082874fe897f43e66be83a0501afe4a08cfc7e1066ab88423af122641493c7af9f0a745eb85c50fddb096de5cc77cd5ff05503312d84d36ab2681c6e6d930bbe68edaebf8fae03b04eb669359965e01c266b;"
                        + "__ac_nonce=0666fd1a00053bf535b9f;"
                        + "__ac_signature=_02B4Z6wo00f01u8PTiQAAIDBvfBuP-YjUQbvL0qAAN25bWfWXQrzRNCBKvFYKS5wAOYPXg5XV1Ck9JEroeWeWKijH2v3i4lxXM37JogiJJfEtYD.8sbXul2-4v.VRRta4xa07ignRnGj5Voh83;"
                        + "ttwid=1%7C9b5sTIuwZxZKt0wFsvE-2t5OoFxH_Q5VIpVNWEREbAo%7C1718605316%7C9dfc9322350e713e6109ed46a7047ed31c0ab5a724e84de0bb766c195043207c")
                .header("user-agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3")
                .header("referer", "https://www.ixigua.com/" + itemId);

        // 发送请求
        HttpResponse response = request.execute();

        // 解析响应
        String responseBody = response.body();
        System.out.println(responseBody);
        return extractInfoByRegex(responseBody);
    }

    public static JSONObject extractInfoByRegex(String text) {
        JSONObject result = extractVideoUrl(text);

        // 标题
        String titleContent = extractTitle(text);
        result.put("video_title", titleContent);

        return result;
    }

    private static String extractTitle(String text) {
        Pattern pattern = Pattern.compile("<title[^>]*>(.*?)</title>");
        Matcher matcher = pattern.matcher(text);
        if (matcher.find()) {
            String title = matcher.group(1);
            // 处理标题内容
            String[] parts = title.split(" - ");
            String titleContent = parts[0];

            // 尝试将标题内容转换为 UTF-8
            try {
                return new String(titleContent.getBytes(StandardCharsets.ISO_8859_1), StandardCharsets.UTF_8);
            } catch (Exception e) {
                e.printStackTrace();
            }
            return titleContent;
        }
        return "";
    }

    public static JSONObject extractVideoUrl(String text) {
        Document document = Jsoup.parse(text);
        Elements scriptElements = document.select("script#SSR_HYDRATED_DATA");
        String str2 = Objects.requireNonNull(scriptElements.first()).html();
        String json2 = str2.substring(str2.indexOf('{'), str2.lastIndexOf('}') + 1);
        String[] irregulars = {"null", "undefined", "=false", "=true", "false", "true"};

        // 处理不规则定义
        for (String i : irregulars) {
            if ("=false".equals(i) || "=true".equals(i)) {
                json2 = json2.replace(i, "=" + Character.toUpperCase(i.charAt(1)) + i.substring(2));
            } else {
                json2 = json2.replace(i, "12");
            }
        }

        try {
            ObjectMapper objectMapper = new ObjectMapper();
            JsonNode dict2 = objectMapper.readTree(json2).path("anyVideo").path("gidInformation").path("packerData").path("video");

            long duration = dict2.path("video_duration").asLong();
            long playCnt = dict2.path("video_watch_count").asLong();
            long publishTime = dict2.path("video_publish_time").asLong();
            long likeCnt = dict2.path("video_like_count").asLong();
            String videoTitle = dict2.path("title").asText();
            String videoId = dict2.path("vid").asText();
            JsonNode videoRes = dict2.path("videoResource");
            String coverUrl = dict2.path("poster_url").asText().replace("\\u002F", "/");

            String realVideoUrl;
            if (videoRes.path("dash").asInt() == 12) {
                JsonNode obj = videoRes.path("normal");
                String ptk = obj.path("ptk").asText();
                JsonNode videoList = obj.path("video_list");
                String mainUrl = videoList.fields().next().getValue().path("main_url").asText();
                realVideoUrl = aesDecrypt(mainUrl, ptk);
            } else {
                JsonNode obj = videoRes.path("dash");
                String ptk = obj.path("ptk").asText();
                String videoUrl = obj.path("dynamic_video").path("main_url").asText();
                realVideoUrl = aesDecrypt(videoUrl, ptk);
            }

            JSONObject result = new JSONObject();
            result.put("video_url", realVideoUrl);
            result.put("cover_url", coverUrl);
            result.put("video_id", videoId);
            result.put("video_title", videoTitle);
            result.put("like_cnt", likeCnt);
            result.put("play_cnt", playCnt);
            result.put("publish_time", publishTime);
            result.put("duration", duration);
            return result;

        } catch (Exception e) {
            e.printStackTrace();
            return new JSONObject();
        }
    }

    private static String aesDecrypt(String data, String key) {
        try {
            byte[] keyBytes = key.getBytes(StandardCharsets.UTF_8);
            SecretKeySpec secretKey = new SecretKeySpec(keyBytes, "AES");
            Cipher cipher = Cipher.getInstance("AES/CBC/PKCS5Padding");
            IvParameterSpec ivParams = new IvParameterSpec(keyBytes, 0, 16);
            cipher.init(Cipher.DECRYPT_MODE, secretKey, ivParams);
            byte[] decryptedBytes = cipher.doFinal(Base64.getDecoder().decode(data));
            return new String(decryptedBytes, StandardCharsets.UTF_8);
        } catch (Exception e) {
            e.printStackTrace();
            return null;
        }
    }


    // 假设 sensitiveFlag 方法的实现
    private static boolean sensitiveFlag(List<String> sensitiveWords, String title) {
        return true;
    }

    public static void main(String[] args) {
        String keyword = "测试";  // 替换为实际关键词
        List<String> sensitiveWords = List.of("敏感词1", "敏感词2");  // 替换为实际敏感词
        List<JSONObject> results = xiguaSearchV2(keyword, sensitiveWords);
        results.forEach(System.out::println);
    }
}