• "Katayama Hirofumi MZ" <katayama.hirofumi.mz gmail.com> May 29 2012
• Jonathan M Davis <jmdavisProg gmx.com> May 29 2012
• =?UTF-8?B?QWxleCBSw7hubmUgUGV0ZXJzZW4=?= <alex lycus.org> May 30 2012

"Katayama Hirofumi MZ" <katayama.hirofumi.mz gmail.com> writes:

Look this source:
http://codepad.org/QXAeG2aJ

The content of RbArray won't be released at the moment of 
deleting an RbArray.

Why?  Is it a bug?

Jonathan M Davis <jmdavisProg gmx.com> writes:

On Wednesday, May 30, 2012 07:47:22 Katayama Hirofumi MZ wrote:
 Look this source:
 http://codepad.org/QXAeG2aJ
 
 The content of RbArray won't be released at the moment of
 deleting an RbArray.
 
 Why?


It's probably only releasing the memory for RbArray and not what it refers to. 
m_objects is a dynamic array, and so it's memory is elsewhere on the heap 
rather than being part of RbArray itself. So, it probably doesn't get released 
when the RbArray is deleted.

  Is it a bug?


Even if it were, it wouldn't get fixed. delete is being removed from the 
language. You really shouldn't be using it.

- Jonathan M Davis

=?UTF-8?B?QWxleCBSw7hubmUgUGV0ZXJzZW4=?= <alex lycus.org> writes:

On 30-05-2012 07:47, Katayama Hirofumi MZ wrote:
 Look this source:
 http://codepad.org/QXAeG2aJ

 The content of RbArray won't be released at the moment of deleting an
 RbArray.

 Why? Is it a bug?


Just use GC.collect() from core.memory when cleaning up internals of 
extremely large objects is too troublesome.

-- 
Alex Rønne Petersen
alex lycus.org
http://lycus.org