Commit dfc68c42 by cwy

增加懂车帝 自媒体号获取

parent 132b70cb
......@@ -5,7 +5,6 @@ import java.util.List;
import java.util.Map;
import java.util.Objects;
import org.apache.commons.lang3.StringUtils;
import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.jsoup.Jsoup;
......
......@@ -4,7 +4,6 @@ import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
import com.zhiwei.proxy.config.SimpleConfig;
import com.zhiwei.source_forward.util.ProxyInit;
import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
......@@ -33,7 +32,7 @@ public class MediaSelfSource {
ProxyInit.initProxy();
List<String> urlList = new ArrayList<>();
urlList.add("https://www.tuicool.com/articles/nIfmu2B");
urlList.add("https://www.dcdapp.com/article/6819085953756299789");
List<MediaSelfSourceBean> u = MediaSelfSource.getMediaSelfSource(urlList);
for(MediaSelfSourceBean b : u) {
System.out.println(b.toString());
......
......@@ -6,14 +6,13 @@ import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
import com.zhiwei.source_forward.util.ProxyInit;
import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import com.zhiwei.crawler.core.HttpBoot;
import com.zhiwei.source_forward.bean.UrlLiveBean;
import com.zhiwei.source_forward.bean.UrlLiveBean.Attribution;
import com.zhiwei.source_forward.crawler.UrlLiveCrawler;
import com.zhiwei.source_forward.util.ProxyInit;
import com.zhiwei.source_forward.util.UrlLiveDataCallback;
/**
......@@ -26,8 +25,6 @@ public class URLLive {
private static Logger logger = LogManager.getLogger(URLLive.class);
private static HttpBoot httpBoot = new HttpBoot.Builder().build();
/**
* @Title: verificationURLLive
* @author hero
......@@ -42,7 +39,6 @@ public class URLLive {
for(Entry<String,Map<String,Object>> entry : dataMap.entrySet()){
urlList.add(entry.getKey());
}
System.out.println(urlList.size());
//验证数据是否已删除
List<UrlLiveBean> dataList = UrlLiveCrawlerThread.getUrlLiveCrawle(urlList);
for(UrlLiveBean ub : dataList){
......
......@@ -372,7 +372,13 @@ public class MatchSource {
source = "搜狐-" + source;
}
}
}else if(url.contains("tznew.58.com")){
}else if(url.contains("dcdapp.com/article")){
// 懂车帝
source = html.split("screen_name\":\"")[1].split("\"}")[0].trim();
if(source!=null && source.length()>1){
source = "懂车帝-" + source;
}
}else if(url.contains("tznew.58.com")){
source = JSONObject.parseObject(html).getJSONObject("result").getString("author");
if(source!=null && source.length()>1){
source = "58-" + source;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment