[Rife-users] 答复: Is this a bug ?

wang wei Tue, 04 Apr 2006 08:12:50 -0700

More information: I use the latest checkout from svn.

And if I comment “mRequestState.setRequest (null)”, it works well. But I guess this line should be useful.

There is another strange issue in my use of rife. When I use ${bv something/}, it sometimes is kept in web page shown to user, while if I change it to ${bv something}${/bv}, it will be erased as it intends.

Of course, I always use ${bv}${/bv} even empty default value now. Just wonder if this is improper.

发件人: wei wang [mailto:[EMAIL PROTECTED]
发送时间: Tuesday, April 04, 2006 10:57 PM
收件人: RIFE users list : questions, bug reports and suggestions.
主题: Is this a bug ?

Hi,

When I use continuation in rife, I enable trace. After "pause", the program runs into processContext of com.uwyn.rife.engine.ElementContext, at line 539, here request is set to null by "mRequestState.setRequest (null)". then at line 629 mElementInfo.outputTrace(start, mRequestState), which calls:
output.append(state.getServerName()); and
output.append(state.getServerPort());
from ElementInfo line 897,899. As state.mRequest is set to null, a null-pointer exception will be thrown.

As I am totally new to rife, I dont know if it is ok to just turn off the trace for servername/port in this case, or there is really a bug in the logic ?

Best Regards
wangwei

_______________________________________________
Rife-users mailing list
[email protected]
http://lists.uwyn.com/mailman/listinfo/rife-users

[Rife-users] 答复: Is this a bug ?

Reply via email to