Commit 90fc68cf by zhiwei

1.添加渠道匹配规则

2.添加英文来源匹配规则
parent f67a402c
......@@ -25,15 +25,11 @@
<artifactId>WebCollector</artifactId>
<version>2.71</version>
</dependency>
<dependency>
<groupId>us.codecraft</groupId>
<artifactId>webmagic-core</artifactId>
<version>0.6.1</version>
</dependency>
<dependency>
<groupId>us.codecraft</groupId>
<artifactId>webmagic-extension</artifactId>
......@@ -45,7 +41,6 @@
</exclusion>
</exclusions>
</dependency>
<dependency>
<groupId>us.codecraft</groupId>
<artifactId>webmagic-saxon</artifactId>
......
......@@ -66,12 +66,16 @@ public class SourceForwardPageProcessor implements PageProcessor {
*/
private static String verifyChannel(String url){
String channel = null;
if(url.contains("news.")){
if(url.contains("news.") || url.contains("cj.sina.com.cn")
|| url.contains("wemedia.ifeng.com")){
channel = "新闻";
}else if(url.contains("finance.") || url.contains("business.")
|| url.contains("money.")){
|| url.contains("money.") || url.contains("stock.")
|| url.contains("10jqka.com.cn")){
channel = "财经";
}else if(url.contains("tech.") || url.contains("it.")){
}else if(url.contains("tech.") || url.contains("it.")
|| url.contains("pcedu.") || url.contains("mobile.")
|| url.contains("vr.")){
channel = "科技";
}else if(url.contains("sports.")){
channel = "体育";
......@@ -90,6 +94,14 @@ public class SourceForwardPageProcessor implements PageProcessor {
channel = "房产";
}else if(url.contains("games.")){
channel = "游戏";
}else if(url.contains("intl.")){
channel = "国际";
}else if(url.contains("science.")){
channel = "科学";
}else if(url.contains("city.")){
channel = "城市";
}else if(url.contains("sc.")){
channel = "市场";
}
return channel;
......
......@@ -49,7 +49,7 @@ public class SourceForward {
String isForward = "转发";
if(root_source == null){
isForward = "原创";
}else if(root_source.equals(source)){
}else if(root_source.toUpperCase().trim().equals(source.toUpperCase().trim())){
isForward = "原创";
}
data.put("是否转发", isForward);
......
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