Skip to content

W
webmagic
Commit 77e6ca29 authored by yihua.huang's avatar yihua.huang
Browse files
Options

update comments

parent 50732582
Hide whitespace changes
Inline Side-by-side
Showing with 17 additions and 19 deletions
webmagic-core/src/main/java/us/codecraft/webmagic/selector/PlainText.java
View file @ 77e6ca29
......@@ -6,10 +6,11 @@ import java.util.ArrayList;
import java.util.List;
/**
* 可抽取的纯文本，不包括xpath和css selector实现。<br>
* Selectable plain text.<br>
* Can not be selected by XPath or CSS Selector.
*
* @author code4crafter@gmail.com <br>
* Date: 13-4-21
* Time: 上午7:54
* @since 0.1.0
*/
public class PlainText implements Selectable {
......@@ -59,7 +60,7 @@ public class PlainText implements Selectable {
List<String> results = new ArrayList<String>();
for (String string : strings) {
String result = selector.select(string);
if (result!=null){
if (result != null) {
results.add(result);
}
}
......
webmagic-core/src/main/java/us/codecraft/webmagic/utils/Experimental.java
View file @ 77e6ca29
package us.codecraft.webmagic.utils;
/**
* Stands for features unstable.
* @author code4crafter@gmail.com <br>
* Stands for features not stable.
*/
public @interface Experimental {
}
webmagic-core/src/main/java/us/codecraft/webmagic/utils/FilePersistentBase.java
View file @ 77e6ca29
......@@ -3,11 +3,10 @@ package us.codecraft.webmagic.utils;
import java.io.File;
/**
* 文件持久化的基础类。<br>
* Base object of file persistence.
*
* @author code4crafter@gmail.com <br>
* Date: 13-8-11 <br>
* Time: 下午4:21 <br>
* @since 0.2.0
*/
public class FilePersistentBase {
......
webmagic-core/src/main/java/us/codecraft/webmagic/utils/ThreadUtils.java
View file @ 77e6ca29
......@@ -6,10 +6,8 @@ import java.util.concurrent.ThreadPoolExecutor;
import java.util.concurrent.TimeUnit;
/**
* 线程工具类。<br>
* @author code4crafer@gmail.com
* Date: 13-6-23
* Time: 下午7:11
* @since 0.1.0
*/
public class ThreadUtils {
......
webmagic-core/src/main/java/us/codecraft/webmagic/utils/UrlUtils.java
View file @ 77e6ca29
......@@ -6,20 +6,20 @@ import java.util.regex.Matcher;
import java.util.regex.Pattern;
/**
* url及html处理工具类。<br>
* url and html utils.
*
* @author code4crafter@gmail.com <br>
* Date: 13-4-21
* Time: 下午1:52
* @since 0.1.0
*/
public class UrlUtils {
private static Pattern relativePathPattern = Pattern.compile("^([\\.]+)/");
/**
* 将url想对地址转化为绝对地址
* @param url url地址
* @param refer url地址来自哪个页面
* @return url绝对地址
* canonicalizeUrl
* @param url
* @param refer
* @return canonicalizeUrl
*/
public static String canonicalizeUrl(String url, String refer) {
if (StringUtils.isBlank(url) || StringUtils.isBlank(refer)) {
......
webmagic-core/src/main/java/us/codecraft/webmagic/utils/package.html
View file @ 77e6ca29
<html>
<body>
提供一些处理链接的静态工具类。
Static utils of webmagic.
</body>
</html>
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