> It certainly looks like some code was copied; one file, for instance, is
about 70% the same. This is not an issue as it is under a compatible MIT
license, but that needs to be mentioned in the LICENSE file.
Could you please tell the file name and the file on OpenSumi that is
overlapped? Of course we'd like to add the license info if it's the case,
and actually I'd generally add a link to the original file. Now we cannot
find the origin and the author states that they write the code by
themselves.
You state that there is 70% the same, as what? It's quite confusing to give
such a conclusion without certain reference.
> A possible explanation for what has happened here is that the developer
used AI to generate the code, and it’s duplicated that code from somewhere.
Do you know if that is the case?
If you have a tool to check such similarity, could you share it or share
the report. When you state a file is copied from elsewhere, there must be
an origin. AFAIK we don't use AI to generate this code, if you take a look
at the code (please look), the content is:
public static class ArrayAsList extends AbstractList<Object> implements
RandomAccess, java.io.Serializable {
private static final Object[] EMPTY = new Object[0];
private Object[] array;
private int size;
public ArrayAsList(int size) { array = new Object[size]; }
@Override public int size() { return size; }
@Override public boolean add(Object e) { array[size++] = e; return
true; }
@Override public Object get(int index) { return array[index]; }
public void clearArray() { size = 0; array = EMPTY; }
public void setArray(Object[] a) { array = a; size = a.length; }
public Object[] getArray() { return array; }
@Override public Object set(int index, Object element) { throw new
UnsupportedOperationException(); }
@Override public int indexOf(Object o) { throw new
UnsupportedOperationException(); }
@Override public boolean contains(Object o) { throw new
UnsupportedOperationException(); }
@Override public Object[] toArray() { return array; }
@Override public Iterator<Object> iterator() { return new
Iterator<Object>() {
private int index;
@Override public boolean hasNext() { return index < array.length; }
@Override public Object next() { return array[index++]; }
};
}
}
Getters, trivial add / clear / set, three unsupported methods. The fields
and non-inherited methods are also written originally.
Best,
tison.
Justin Mclean <jus...@classsoftware.com> 于2024年4月25日周四 10:36写道：
> Hi,
>
> I’m currently travelling and may be slow in responding to emails.
>
> > The MemoryBufferWritableChannel[5] and MockWritableChannel[6] was
> written
> > by me before we open-sourced Fury and
> > I submitted it to Ray in PR[8] ,which was planned to optimize zero-copy
> > serialization in Ray. I think it's OK to include it
> > here since both are written by me.
>
> That should be fine, but having an ASF header on the files may be a little
> misleading? What do you think?
>
> > For files [1][2][3][4], could you give more details how those files
> include
> > code from OpenSumi. The OpenSumi core developer
> > bytemain[9] does commit some code to Fury, see his commits[10]. But I
> > talked to him offline, he said he didn't copy code from
> > OpenSumi.
>
> It certainly looks like some code was copied; one file, for instance, is
> about 70% the same. This is not an issue as it is under a compatible MIT
> license, but that needs to be mentioned in the LICENSE file.
>
> > For Fury ArrayAsList, could you give more details how this class is
> copied
> > from somewhere? It's just a simple wrapper for java
> > object arrays.
>
> A possible explanation for what has happened here is that the developer
> used AI to generate the code, and it’s duplicated that code from somewhere.
> Do you know if that is the case?
>
> Kind Regards,
> Justin
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
> For additional commands, e-mail: general-h...@incubator.apache.org
>
>
