DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7845>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE.
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7845 org.apache.avalon.excalibur.collections.VariableSizeBuffer loses data on resize ------- Additional Comments From [EMAIL PROTECTED] 2002-04-08 16:30 ------- This is how the add() method looks like now: public final void add( final Object o ) { if( size() + 1 >= m_buffer.length ) { Object[] tmp = new Object[ ( (m_buffer.length - 1) * 2) + 1 ]; int copyPos = m_head; for (int i = 0; i < m_buffer.length; i++, copyPos = (copyPos + 1) % m_buffer.length) { tmp[ i ] = m_buffer[copyPos]; m_buffer[ copyPos ] = null; } m_head = 0; m_tail = m_buffer.length - 1; m_buffer = tmp; } m_buffer[ m_tail ] = o; m_tail++; if (m_tail >= m_buffer.length) { m_tail = 0; } } Basically what I changed is the way the buffer is resized. It rewinds and so keeping the semantics of a circular buffer. About tests: it is easy to write one. I have one for using the db pooling part which uses this class then. How you see this is to do let's say inserts of 2 elements followed by a remove of an element. Do this step in a loop until the preallocated size (default is 32 now) gets full. You should be able to go one more such a step over the full buffer (in order to trigger the resize) and this is basically the test that will show this bug. The code that I inserted before fixes this issue. -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
