java - 使用 jersey/apache REST 客户端解析多行、多列文本响应

Question

我正在尝试访问返回多行、多列响应的 REST 端点，例如:

A1   B1    C1
A2   B2    C2
A3   B3    C3
...
...

I'm currently using jersey-client to hit this endpoint and trying to look for the neatest way to parse this response. Here, each line would represent a bean, say MyBean and each column on that would represent a property in that bean. The order of values in the response is always fixed.

I can get the response back as a long String, split it at line-feeds and tabs to get individual values.

However, I would like to know if there is a way where I can get the results as a List<String>, where each element in the List would represent a line of the response. I can then split it on \t to get individual values.

Here's what I've tried:

    WebResource resource = client.resource(NETSPEAK_URL)
                 .type(MediaType.TEXT_PLAIN)
                 .get(new GenericType<List<String>>(){});

但这会导致以下异常:

    A message body reader for Java class java.util.List, 
    and Java type java.util.List<java.lang.String>, 
    and MIME media type text/plain; charset=UTF-8 was not found

如果我更加贪婪，我想知道是否可以获得映射到我的 bean MyBean 属性的各个列值。我考虑过围绕 MyBean 创建一个包装器，以获得一个 MyBean 列表，但是如何对其进行注释以协助解析呢？这对于 xml/json 响应来说是有意义的。但这是纯文本。

是否有可能以某种方式告诉 Jersey 客户端有关此文本/纯响应的解析？如果这可以通过 Apache HTTP 客户端实现，我就准备好了。

谢谢

Answer 1

您可能想要实现一个代表您的 bean 列表的类，例如 class BeanList extends List<Bean> ，并实现 MessageBodyReader<BeanList> (参见 http://jsr311.java.net/nonav/releases/1.1/javax/ws/rs/ext/MessageBodyReader.html )教 Jersey 如何将字符串读取为 BeanList .

然后你可以使用BeanList.class而不是List<String>作为 get 的参数称呼。