[
https://issues.apache.org/jira/browse/MJAVADOC-658?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jin Xu updated MJAVADOC-658:
----------------------------
Description:
original function:
{code:java}
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix", we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
* @param <T> a T object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix" another time, we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix" the third time, we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
* @param <T> a T object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
again.
and if we run fix for another time, it will continue the loop.
so do we really need {code:java}@param <T> a T object.{code}?
If not, then it is wrongly added.
If so, then it is wrongly deleted.
Be a bug either way.
was:
original function:
{code:java}
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix", we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
* @param <T> a T object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix" another time, we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
after running "fix" the third time, we get:
{code:java}
/**
* <p>testGeneric.</p>
*
* @param tClass a {@link java.lang.Class} object.
* @param o a {@link java.lang.Object} object.
* @param <T> a T object.
*/
public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
return;
}
{code}
again.
and if we run fix for another time, it will continue the loop.
so do we really need {code:java}@param <T> a T object.{code}?
> found a bug about wrongly handling generic functions when using fix operation
> -----------------------------------------------------------------------------
>
> Key: MJAVADOC-658
> URL: https://issues.apache.org/jira/browse/MJAVADOC-658
> Project: Maven Javadoc Plugin
> Issue Type: Bug
> Components: fix
> Reporter: Jin Xu
> Priority: Major
>
> original function:
> {code:java}
> public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
> return;
> }
> {code}
> after running "fix", we get:
> {code:java}
> /**
> * <p>testGeneric.</p>
> *
> * @param tClass a {@link java.lang.Class} object.
> * @param o a {@link java.lang.Object} object.
> * @param <T> a T object.
> */
> public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
> return;
> }
> {code}
> after running "fix" another time, we get:
> {code:java}
> /**
> * <p>testGeneric.</p>
> *
> * @param tClass a {@link java.lang.Class} object.
> * @param o a {@link java.lang.Object} object.
> */
> public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
> return;
> }
> {code}
> after running "fix" the third time, we get:
> {code:java}
> /**
> * <p>testGeneric.</p>
> *
> * @param tClass a {@link java.lang.Class} object.
> * @param o a {@link java.lang.Object} object.
> * @param <T> a T object.
> */
> public <T extends Object> void testGeneric(Class<T> tClass, Object o) {
> return;
> }
> {code}
> again.
> and if we run fix for another time, it will continue the loop.
> so do we really need {code:java}@param <T> a T object.{code}?
> If not, then it is wrongly added.
> If so, then it is wrongly deleted.
> Be a bug either way.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
