<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<br>
<br>
<div class="moz-cite-prefix">On 4/18/2016 6:12 PM, Alexander
Scherbatiy wrote:<br>
</div>
<blockquote cite="mid:5714F973.3070909@oracle.com" type="cite">On
4/1/2016 1:29 PM, Semyon Sadetsky wrote:
<br>
<blockquote type="cite">Please review the updated version :
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~ssadetsky/8147842/webrev.01/">http://cr.openjdk.java.net/~ssadetsky/8147842/webrev.01/</a>
<br>
<br>
fix was changed to avoid possible recursion in native IME evens
handling which the SendMessage() call might cause.
<br>
</blockquote>
<br>
The SendMessage call has been added by the fix JDK-8079595
Resizing dialog which is JWindow parent makes JVM crash
<br>
Could you give more details why removing this code back will not
cause new JVM crash?
<br>
</blockquote>
I did not remove the code back. Before JDK-8079595 the
DefSubclassProc function was called actually (see
<meta http-equiv="content-type" content="text/html; charset=utf-8">
ComCtl32Util::DefWindowProc). This function is not threadsafe and
that was the root cause of the JDK-8079595.<br>
<br>
--Semyon<br>
<blockquote cite="mid:5714F973.3070909@oracle.com" type="cite">
<br>
Thanks,
<br>
Alexandr.
<br>
<br>
<br>
<blockquote type="cite">
<br>
--Semyon
<br>
<br>
On 3/31/2016 7:38 PM, Semyon Sadetsky wrote:
<br>
<blockquote type="cite">
<br>
<br>
On 3/31/2016 7:11 PM, Semyon Sadetsky wrote:
<br>
<blockquote type="cite">The solution is to support positioning
for the IME candidate window.
<br>
</blockquote>
<br>
It should be : "the IME *composition* window" of cause...
<br>
<br>
--Semyon
<br>
<br>
</blockquote>
<br>
</blockquote>
<br>
</blockquote>
<br>
</body>
</html>