Gary,

I'll look at the test (but have to leave for work now).

About the name, yes, let's change it.
The main differences with the existing FileAppenders are
- the Fast versions are always buffered
- the Fast versions use a RandomAccessFile instead of a Stream, which was quite 
a bit faster in my testing
(If you're interested, I got there because of this article:
http://mechanical-sympathy.blogspot.jp/2011/12/java-sequential-io-performance.html
 )


Suggestions for better names:
- RandomAccessFileAppender
- RAFileAppender

Other suggestions welcome...

-Remko


________________________________
 From: Gary Gregory <[email protected]>
To: Log4J Developers List <[email protected]> 
Sent: Wednesday, July 10, 2013 7:10 AM
Subject: Fast file problem with XML layouts
 


Hi All:

I created and @Ignored this failing test for a FastFileAppender with a 
'complete' (aka well-formed) XML layout:

org.apache.logging.log4j.core.appender.FastXmlFileAppenderTest

which is a duplicate of the normal file test I also just created:

org.apache.logging.log4j.core.appender.XmlFileAppenderTest

Is this a bug or did I misconfigure the test?

Also, more Javadoc on FastFileAppender would help.

Can we change "Fast" to something more meaningful and descriptive? No Javadoc, 
no idea, but I did see a thread I think... is this an AsyncFileAppender? How is 
it different than wrapping a file appender with an AsyncApppender? Filtering? 
What else?

Thank you,
Gary


-- 

E-Mail: [email protected] | [email protected] 
Java Persistence with Hibernate, Second Edition
JUnit in Action, Second Edition
Spring Batch in Action
Blog: http://garygregory.wordpress.com 
Home: http://garygregory.com/
Tweet! http://twitter.com/GaryGregory

Reply via email to