update afterextract api

webmagic-core/src/main/java/us/codecraft/webmagic/oo/AfterExtractor.java

@@ -9,7 +9,7 @@ import us.codecraft.webmagic.Page;
  * @date: 13-8-3 <br>
  * Time: 上午9:42 <br>
  */
-public interface AfterExtractor<T> {
+public interface AfterExtractor {
 
-    public void afterProcess(Page page, T t);
+    public void afterProcess(Page page);
 }

webmagic-core/src/main/java/us/codecraft/webmagic/oo/ObjectPageProcessor.java

@@ -4,11 +4,13 @@ import us.codecraft.webmagic.Page;
 import us.codecraft.webmagic.Request;
 import us.codecraft.webmagic.Site;
 import us.codecraft.webmagic.processor.PageProcessor;
+import us.codecraft.webmagic.selector.Selector;
 
 import java.util.ArrayList;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Set;
+import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
 /**
@@ -33,7 +35,7 @@ public class ObjectPageProcessor implements PageProcessor {
     }
 
 
-    public ObjectPageProcessor addPageModel(Class clazz){
+    public ObjectPageProcessor addPageModel(Class clazz) {
         PageModelExtractor pageModelExtractor = PageModelExtractor.create(clazz);
         targetUrlPatterns.addAll(pageModelExtractor.getTargetUrlPatterns());
         targetUrlPatterns.addAll(pageModelExtractor.getHelpUrlPatterns());
@@ -49,22 +51,34 @@ public class ObjectPageProcessor implements PageProcessor {
     public void process(Page page) {
         for (PageModelExtractor pageModelExtractor : pageModelExtractorList) {
             Object process = pageModelExtractor.process(page);
-            if (process==null){
+            if (process == null) {
                 page.getResultItems().setSkip(true);
             }
             postProcessPageModel(pageModelExtractor.getClazz(), process);
             page.putField(pageModelExtractor.getClazz().getCanonicalName(), process);
+            extractLinks(page, pageModelExtractor.getHelpUrlRegionSelector(), pageModelExtractor.getHelpUrlPatterns());
+            extractLinks(page, pageModelExtractor.getTargetUrlRegionSelector(), pageModelExtractor.getTargetUrlPatterns());
         }
-        for (String link : page.getHtml().links().all()) {
-            for (Pattern targetUrlPattern : targetUrlPatterns) {
-                if (targetUrlPattern.matcher(link).matches()){
-                    page.addTargetRequest(new Request(link));
+    }
+
+    private void extractLinks(Page page, Selector urlRegionSelector, List<Pattern> urlPatterns) {
+        List<String> links;
+        if (urlRegionSelector == null) {
+            links = page.getHtml().links().all();
+        } else {
+            links = urlRegionSelector.selectList(page.getHtml().toString());
+        }
+        for (String link : links) {
+            for (Pattern targetUrlPattern : urlPatterns) {
+                Matcher matcher = targetUrlPattern.matcher(link);
+                if (matcher.find()) {
+                    page.addTargetRequest(new Request(matcher.group(1)));
                 }
             }
         }
     }
 
-    protected void postProcessPageModel(Class clazz, Object object){
+    protected void postProcessPageModel(Class clazz, Object object) {
     }
 
     @Override

webmagic-core/src/main/java/us/codecraft/webmagic/oo/PageModelExtractor.java

@@ -21,14 +21,16 @@ class PageModelExtractor {
 
     private List<Pattern> targetUrlPatterns = new ArrayList<Pattern>();
 
+    private Selector targetUrlRegionSelector;
+
     private List<Pattern> helpUrlPatterns = new ArrayList<Pattern>();
 
+    private Selector helpUrlRegionSelector;
+
     private Class clazz;
 
     private List<FieldExtractor> fieldExtractors;
 
-    private AfterExtractor afterExtractor;
-
     public static PageModelExtractor create(Class clazz) {
         PageModelExtractor pageModelExtractor = new PageModelExtractor();
         pageModelExtractor.init(clazz);
@@ -39,13 +41,6 @@ class PageModelExtractor {
         this.clazz = clazz;
         initTargetUrlPatterns();
         fieldExtractors = new ArrayList<FieldExtractor>();
-        if (AfterExtractor.class.isAssignableFrom(clazz)) {
-            try {
-                afterExtractor = (AfterExtractor) clazz.newInstance();
-            } catch (Exception e) {
-                throw new IllegalArgumentException(e);
-            }
-        }
         for (Field field : clazz.getDeclaredFields()) {
             field.setAccessible(true);
             ExtractBy extractBy = field.getAnnotation(ExtractBy.class);
@@ -117,16 +112,24 @@ class PageModelExtractor {
         if (annotation == null) {
             targetUrlPatterns.add(Pattern.compile(".*"));
         } else {
-            String[] value = ((TargetUrl) annotation).value();
+            TargetUrl targetUrl = (TargetUrl) annotation;
+            String[] value = targetUrl.value();
             for (String s : value) {
-                targetUrlPatterns.add(Pattern.compile(s.replace(".", "\\.").replace("*", "[^\"'#]*")));
+                targetUrlPatterns.add(Pattern.compile("("+s.replace(".", "\\.").replace("*", "[^\"'#]*")+")"));
+            }
+            if (!targetUrl.sourceRegion().equals("")){
+                targetUrlRegionSelector = new Xpath2Selector(targetUrl.sourceRegion());
             }
         }
         annotation = clazz.getAnnotation(HelpUrl.class);
         if (annotation != null) {
-            String[] value = ((HelpUrl) annotation).value();
+            HelpUrl helpUrl = (HelpUrl) annotation;
+            String[] value = helpUrl.value();
             for (String s : value) {
-                helpUrlPatterns.add(Pattern.compile(s.replace(".", "\\.").replace("*", "[^\"'#]*")));
+                helpUrlPatterns.add(Pattern.compile("("+s.replace(".", "\\.").replace("*", "[^\"'#]*")+")"));
+            }
+            if (!helpUrl.sourceRegion().equals("")){
+                helpUrlRegionSelector = new Xpath2Selector(helpUrl.sourceRegion());
             }
         }
     }
@@ -179,8 +182,8 @@ class PageModelExtractor {
                     setField(o, fieldExtractor, value);
                 }
             }
-            if (afterExtractor != null) {
-                afterExtractor.afterProcess(page, o);
+            if (AfterExtractor.class.isAssignableFrom(clazz)) {
+                ((AfterExtractor)o).afterProcess(page);
             }
         } catch (InstantiationException e) {
             e.printStackTrace();
@@ -210,4 +213,12 @@ class PageModelExtractor {
     List<Pattern> getHelpUrlPatterns() {
         return helpUrlPatterns;
     }
+
+    Selector getTargetUrlRegionSelector() {
+        return targetUrlRegionSelector;
+    }
+
+    Selector getHelpUrlRegionSelector() {
+        return helpUrlRegionSelector;
+    }
 }

webmagic-core/src/test/java/us/codecraft/webmagic/oo/OschinaBlog.java

@@ -9,8 +9,8 @@ import java.util.List;
  * @date: 13-8-1 <br>
  * Time: 下午10:18 <br>
  */
-@TargetUrl("http://my.oschina.net/flashsword/blog/*")
-public class OschinaBlog implements AfterExtractor<OschinaBlog> {
+@TargetUrl(value="http://my.oschina.net/flashsword/blog/*",sourceRegion = "//div[@class='BlogLinks']")
+public class OschinaBlog implements AfterExtractor {
 
     @ExtractBy("//title")
     private String title;
@@ -22,7 +22,7 @@ public class OschinaBlog implements AfterExtractor<OschinaBlog> {
     private List<String> tags;
 
     @Override
-    public void afterProcess(Page page, OschinaBlog oschinaBlog) {
+    public void afterProcess(Page page) {
         content = null;
     }
 }