在 RIAK 上获取 MapReduce 结果(使用 Java 客户端)
我在 RIAK 上存储 Person POJO(4 个字符串字段 - id、name、lastUpdate、Data),然后尝试使用 MapReduce 获取这些对象.
I am storing Person POJOs (4 string fields - id, name, lastUpdate, Data) on RIAK, then trying to fetch these objects with MapReduce.
我的做法与 Basho 文档非常相似:
I am doing it very similar to Basho documentation:
BucketMapReduce m = riakClient.mapReduce("person");
m.addMapPhase(new NamedJSFunction("Riak.mapByFields"), true);
MapReduceResult result = m.execute();
Collection<Person> tmp = result.getResult(Person.class);
调用了 Person 的 String 构造函数:
the Person's String constructor is invoked:
public Person(String str){}
(我必须有这个构造函数,否则我会因为它丢失而出现异常)在那里,我将对象作为字符串 - 一个字符串中的对象字段带有一个奇怪的分隔符.
(I must have this constructor, otherwise I get an exception for it is missing) In there I get the object as a String - the Object's fields in one string with a strange delimiter.
为什么我没有将对象自动转换为我的 POJO?我真的需要遍历字符串并反序列化它吗?我做错了什么吗?s
why am I not getting the object automatically converted to my POJO? do I really need to go over the string and deserialize it? am i doing something wrong?s
推荐答案
您使用的 JS 函数并没有按照您的想法执行 :) 它根据您必须提供的特定值的字段来选择对象作为相位的参数.
The JS function you're using doesn't do what you think it does :) It selects objects based on a field with a specific value you have to supply as an argument to the phase.
我认为您正在寻找的是 mapValuesJson
它将执行您似乎想要做的事情.
I think what you're looking for is mapValuesJson
which will do what you seem to be wanting to do.
此外,您的 POJO 中根本不需要构造函数.
Also, you don't need a constructor at all in your POJO.
下面的代码应该为您指明正确的方向(显然这是超级简单的,POJO 中的所有公共字段都没有注释):
The code below should point you in the right direction (obviously this is super-simple with all public fields in the POJO and no annotations):
public class App {
public static void main( String[] args ) throws IOException, RiakException
{
IRiakClient client = RiakFactory.httpClient();
Bucket b = client.fetchBucket("test_mr").execute();
b.store("myobject", new Person()).execute();
IRiakObject o = b.fetch("myobject").execute();
System.out.println(o.getValueAsString());
BucketMapReduce m = client.mapReduce("test_mr");
m.addMapPhase(new NamedJSFunction("Riak.mapValuesJson"), true);
MapReduceResult result = m.execute();
System.out.println(result.getResultRaw());
Collection<Person> tmp = result.getResult(Person.class);
for (Person p : tmp)
{
System.out.println(p.data);
}
client.shutdown();
}
}
class Person
{
public String id = "12345";
public String name = "my name";
public String lastUpdate = "some time";
public String data = "some data";
}
相关文章