It definitely unnecessary code, so I think a bug-report is a good idea.

2013/1/10 Maarten Bosteels <[email protected]>

> Hello,
>
>
> https://github.com/qos-ch/slf4j/blob/master/jul-to-slf4j/src/main/java/org/slf4j/bridge/SLF4JBridgeHandler.java#L283
>
>
>
>   public void publish(LogRecord record) {
>     // Silently ignore null records.
>     if (record == null) {
>       return;
>     }
>
>
>     Logger slf4jLogger = getSLF4JLogger(record);
>     String message = record.getMessage(); // can be null!
>     // this is a check to avoid calling the underlying logging system
>     // with a null message. While it is legitimate to invoke j.u.l. with
>     // a null message, other logging frameworks do not support this.
>     // see also http://bugzilla.slf4j.org/show_bug.cgi?id=108
>     if (message == null) {
>       message = "";
>     }
>     if (slf4jLogger instanceof LocationAwareLogger) {
>       callLocationAwareLogger((LocationAwareLogger) slf4jLogger, record);
>     } else {
>       callPlainSLF4JLogger(slf4jLogger, record);
>     }
>   }
>
>
> I don't understand the point of changing the local variable message to the 
> empty string since that local variable is not used afterwards.
> Shouldn't this be like this :
>
>     if (message == null) {
>       record.setMessage("");
>     }
>
> Should a file a bug ?
>
>
> Best regards
> Maarten
>
>
> _______________________________________________
> slf4j-user mailing list
> [email protected]
> http://mailman.qos.ch/mailman/listinfo/slf4j-user
>
_______________________________________________
slf4j-user mailing list
[email protected]
http://mailman.qos.ch/mailman/listinfo/slf4j-user

Reply via email to