fix telnet trace times is always 1

[LiZhenNet]

What is the purpose of the change

old:

 if (service == null || service.length() == 0) {
            service = parts.length > 0 ? parts[0] : null;
            method = parts.length > 1 ? parts[1] : null;
        } else {
            method = parts.length > 0 ? parts[0] : null;
        }
        if (StringUtils.isInteger(method)) {
            times = method;
            method = null;
        } else {
            times = parts.length > 2 ? parts[2] : "1";
        }

Usually we use telnet in order，first，we invoke ”cd xxxService“ to change the path，then we invoke "trace xxxMethod 10" to trace a method 。

If we invoke like this "trace xxxMethod 10" and the service is not equals null（we have changed the path） , times will not change the value ，It‘s always 1。

Brief changelog

XXXXX

Verifying this change

XXXXX

Follow this checklist to help us incorporate your contribution quickly and easily:

• Make sure there is a GITHUB_issue field for the change (usually before you start working on it). Trivial changes like typos do not require a GITHUB issue. Your pull request should address just this issue, without pulling in other changes - one PR resolves one issue.
• Format the pull request title like [Dubbo-XXX] Fix UnknownException when host config not exist #XXX. Each commit in the pull request should have a meaningful subject line and body.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Write necessary unit-test to verify your logic correction, more mock a little better when cross module dependency exist. If the new feature or significant change is committed, please remember to add integration-test in test module.
• Run mvn clean install -DskipTests=false & mvn clean test-compile failsafe:integration-test to make sure unit-test and integration-test pass.
• If this contribution is large, please follow the Software Donation Guide.

[LiZhenNet]

@beiwei30 Can you help me review it？

[khanimteyaz]

@beiwei30 I would like to review this. would you help me to understand by functionality wise what @LiZhenNet to trying to achieve here?

[LiZhenNet]

@beiwei30 I would like to review this. would you help me to understand by functionality wise what @LiZhenNet to trying to achieve here?

I try to fix a bug about trace command in telnet。You can see “What is the purpose of the change”

[codecov-io]

Codecov Report

Merging #3038 into master will decrease coverage by 0.12%.
The diff coverage is 60%.

@@             Coverage Diff              @@
##             master    #3038      +/-   ##
============================================
- Coverage     63.69%   63.56%   -0.13%     
  Complexity       75       75              
============================================
  Files           652      652              
  Lines         28199    28206       +7     
  Branches       4782     4784       +2     
============================================
- Hits          17960    17928      -32     
- Misses         7988     8022      +34     
- Partials       2251     2256       +5

Impacted Files	Coverage Δ	Complexity Δ
.../rpc/protocol/dubbo/telnet/TraceTelnetHandler.java	0% <0%> (ø)	0 <0> (ø)	⬇️
...rpc/protocol/dubbo/telnet/InvokeTelnetHandler.java	73.8% <70.58%> (+1.3%)	0 <0> (ø)	⬇️
.../apache/dubbo/qos/protocol/QosProtocolWrapper.java	64.1% <0%> (-17.95%)	0% <0%> (ø)
...ache/dubbo/remoting/transport/AbstractChannel.java	75% <0%> (-12.5%)	0% <0%> (ø)
.../remoting/transport/netty4/NettyServerHandler.java	73.52% <0%> (-11.77%)	0% <0%> (ø)
...che/dubbo/remoting/transport/mina/MinaChannel.java	43.42% <0%> (-10.53%)	0% <0%> (ø)
...onfig/spring/extension/SpringExtensionFactory.java	70.73% <0%> (-9.76%)	0% <0%> (ø)
...in/java/org/apache/dubbo/common/utils/JVMUtil.java	73.58% <0%> (-7.55%)	0% <0%> (ø)
...ng/transport/dispatcher/all/AllChannelHandler.java	51.42% <0%> (-5.72%)	0% <0%> (ø)
.../org/apache/dubbo/remoting/ExecutionException.java	15.78% <0%> (-5.27%)	0% <0%> (ø)
... and 8 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ddb518d...cbcb13b. Read the comment docs.

[LiZhenNet]

I have add a commit c833771 , to fix #3105

[gudegg]

@Data
public class User1 implements Serializable {
    private static final long serialVersionUID = -4729079781904156102L;
    private String name;
}

@Data
public class User2 implements Serializable {
    private static final long serialVersionUID = -2112637973209566052L;
    private String name;
}

public interface XXXService{
String hello(User1 user);
String hello(User2 user);

}

May find not the correct overload method; I think must class param when the method is overloaded.
telnet: invoke xxxService.hello({"name":"xxxx"})

[LiZhenNet]

I have fix it

[LiZhenNet]

@beiwei30

PR 上没看到你的留言。我的意思是，如果是 JSONObject 的话，不需要反序列话做检查，直接让它过，最终调用的时候如果不对，肯定出错的。如果对的话，按照你的逻辑，会做两次反序列化，太重了。

当情况是这样的：

void methodA(ObjectA a);
void methodA(ObjectB b);

class ObjectA{
String key1;
}
class ObjectB{
String key2;
}

如果不进行校验，可能每次match到的都是 ObjectA 的方法。

还有另一个问题，invokeTelnetHandler 把 JsonObject 转换成对应参数类型时使用的 PojoUtils#realize，内部其实是遍历JsonObject 的 key 反射调用set 方法。 这也就造成了当我传递 ObjectB{ key2="2"} 时， 会已 ObjectA {key1:null} 进行调用。 这种情况很容易产生事故。