add jsonpathselector

webmagic-extension/pom.xml

@@ -31,6 +31,11 @@
             <groupId>junit</groupId>
             <artifactId>junit</artifactId>
         </dependency>
+        <dependency>
+            <groupId>com.jayway.jsonpath</groupId>
+            <artifactId>json-path</artifactId>
+            <version>0.8.1</version>
+        </dependency>
     </dependencies>
 
 </project>
\ No newline at end of file

webmagic-extension/src/main/java/us/codecraft/webmagic/selector/JsonPathSelector.java

+package us.codecraft.webmagic.selector;
+
+import com.jayway.jsonpath.JsonPath;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * @author code4crafter@gmail.com <br>
+ *         Date: 13-8-12 <br>
+ *         Time: 下午12:54 <br>
+ */
+public class JsonPathSelector implements Selector {
+
+    private String jsonPathStr;
+
+    private JsonPath jsonPath;
+
+    public JsonPathSelector(String jsonPathStr) {
+        this.jsonPathStr = jsonPathStr;
+        this.jsonPath = JsonPath.compile(jsonPathStr);
+    }
+
+    @Override
+    public String select(String text) {
+        Object object = jsonPath.read(text);
+        if (object == null) {
+            return null;
+        }
+        if (object instanceof List) {
+            List list = (List) object;
+            if (list != null && list.size() > 0) {
+                return list.iterator().next().toString();
+            }
+        }
+        return object.toString();
+    }
+
+    @Override
+    public List<String> selectList(String text) {
+        List<String> list = new ArrayList<String>();
+        Object object = jsonPath.read(text);
+        if (object == null) {
+            return list;
+        }
+        if (object instanceof List) {
+            return (List<String>)object;
+        } else {
+            list.add(object.toString());
+        }
+        return list;
+    }
+}

webmagic-extension/src/test/java/us/codecraft/webmagic/selector/JsonPathSelectorTest.java

+package us.codecraft.webmagic.selector;
+
+import junit.framework.Assert;
+import org.junit.Test;
+
+import java.util.List;
+
+/**
+ * @author code4crafter@gmai.com <br>
+ *         Date: 13-8-12 <br>
+ *         Time: 下午1:12 <br>
+ */
+public class JsonPathSelectorTest {
+
+    private String text = "{ \"store\": {\n" +
+            "    \"book\": [ \n" +
+            "      { \"category\": \"reference\",\n" +
+            "        \"author\": \"Nigel Rees\",\n" +
+            "        \"title\": \"Sayings of the Century\",\n" +
+            "        \"price\": 8.95\n" +
+            "      },\n" +
+            "      { \"category\": \"fiction\",\n" +
+            "        \"author\": \"Evelyn Waugh\",\n" +
+            "        \"title\": \"Sword of Honour\",\n" +
+            "        \"price\": 12.99,\n" +
+            "        \"isbn\": \"0-553-21311-3\"\n" +
+            "      }\n" +
+            "    ],\n" +
+            "    \"bicycle\": {\n" +
+            "      \"color\": \"red\",\n" +
+            "      \"price\": 19.95\n" +
+            "    }\n" +
+            "  }\n" +
+            "}";
+
+    @Test
+    public void test() {
+        JsonPathSelector jsonPathSelector = new JsonPathSelector("$.store.book[*].author");
+        String select = jsonPathSelector.select(text);
+        List<String> list = jsonPathSelector.selectList(text);
+        Assert.assertNotNull(select);
+        Assert.assertNotNull(list);
+        jsonPathSelector = new JsonPathSelector("$.store.book[?(@.category == 'reference')]");
+        list = jsonPathSelector.selectList(text);
+        select = jsonPathSelector.select(text);
+        Assert.assertNotNull(list);
+        Assert.assertNotNull(select);
+    }
+}