RE: 0.9.8 version (poll)

Given the interest in when 0.9.7 can be retired, I tried to get the CVS
tree today, and build it on Windows.  It seems rather broken, looks like
a few files are missing in the current CVS tree and I'm not sure whether
or not they should be there (they are there in the 0.9.7 version).  For
instance:

boundedfifo.cpp, semaphore.cpp, event.cpp(!?),
dailyrollingfileappender.cpp, criticalsection.cpp (and their associated
.h files.

I did a "clean" retrieval using WinCVS from the logging-log4cxx branch -
I also checked the web site version and these files are not part of the
CVS as displayed there.

So I guess I'd answer A) below as -1 -> I can't build it to even test it
yet, sorry.

But I am interested in testing it, so if this is fixed soon, I'll be
looking for it.  Thanks.

Myron Drapal
drapal@xxxxxxxxx

-----Original Message-----
From: Curt Arnold [mailto:carnold@xxxxxxxxxx] 
Sent: Tuesday, January 11, 2005 9:45 AM
To: Log4CXX User
Subject: Re: 0.9.8 version (poll)


On Jan 11, 2005, at 1:54 AM, on log4cxx-user maryline royer wrote:

> Hello,
>
> I wait for the 0.9.8 version. When is it  planned ?
>

Don't we all.  I would love to retire 0.9.7.

As an Apache incubated project, we can't actually have a release, but  
need approval from the incubator project management committee  
(http://incubator.apache.org/incubation/ 
Incubation_Policy.html#Releases%0D) to have a code snapshot.   
Currently, the PMC is voting on exiting log4cxx from the Incubator but  
it has taken a long time to get any type of quorum.  If we knew that we

would be out of incubation shortly, we could do plan for a real release

after leaving the incubator.  If we knew that we weren't getting out of

incubation, we could petition the Incubator PMC to sanction a code  
snapshot.

In addition, the current CVS is a regression from 0.9.8 in that some  
features that were in 0.9.7 have been disabled in the CVS.  The  
SocketAppender and SocketHubAppender are the first that come to mind.   
In a perfect world, we'd be able to get SocketAppender to talk with the

log4j implementation (http://issues.apache.org/jira/browse/LOGCXX-7).   
However, that needs to synchronize with changes in log4j and won't  
likely happen fast enough to go into 0.9.8.

The other thing that I would like to have done in time for 0.9.8, but  
might not make it is http://issues.apache.org/jira/browse/LOGCXX-35  
which would basically migrate log4cxx away from MSXML and libxml2 to  
the expat parser provided in APR.  This would likely requiring porting  
the new "Joran" configurator from log4cxx.

My take is to continue working off as many issues as possible and  
prepare a release as soon as we exit the incubator.

We are in a bit of a chicken and egg syndrome, people won't check out  
the new changes in log4cxx until there is a release and we shouldn't be

releasing software that hasn't been tested.  I would encourage as many  
people as possible to try the CVS version and log any issues that you  
have on the JIRA and mention whether you consider them to be  
showstoppers for a 0.9.8 release.

So, let me throw out a few questions for a poll.

A) Is the current CVS ready for a release (+1 - yes, 0 - haven't tried  
it/maybe, -1 no).  If 0 or -1 explain.

B) Should log4cxx encourage the Incubator PMC to finish up their vote?  
(+1/0/-1)

C) Should log4cxx petition the Incubator PMC for a sanction for a code  
snapshot?

D) What platforms/compilers need to be tested before release/code  
snapshot: (list all that apply)

1.    x86 Linux/gcc
2.     Windows/Visual C++ 5
3.     Windows/Visual C++ 6
4.     Windows/Visual C++ 2001
5.     Windows/Visual C++ 2003
6.     Windows/Visual C++ 2005
7.     Windows/Borland (5.5 or CBuilderX)
8.      Cygwin
9.     MinGW
10.    FreeBSD
11.    NetBSD
12.    Mac OS/X
13.  x86_64 Linux/gcc
14.  x86_64 Windows/appropriate MS compiler
15. Others (plus describe)

E) What platforms are you able to test: (list all that apply)

F) What platforms have you tested (list all that apply)

Drapal, Myron E (Myron).vcf
Description: Drapal, Myron E (Myron).vcf