log4cpp-devel — technical discussions about the development of log4cpp

Re: [Log4cpp-devel] CategoryStream

[Paul D. <pa...@du...>]

On Mon, Oct 15, 2012 at 1:02 PM, Cantor, Scott <can...@os...> wrote:
> On 10/15/12 12:55 PM, "Paul Dugas" <pa...@du...> wrote:
>>
>>My search skills must be failing me.  Been looking into this for a
>>couple days with little luck.  Got a link or a search term I should be
>>using other than log4cpp and CategoryStream?
>
> Not that. Search for the compiler's generic error message info.

That helped.  Thanks much.

For the archive, my issue arose because by foo class was defined in a
namespace while my operator<<() routine was outside of it.  I moved
the routine into the same namespace as the class and it's working now.
 Got the idea from the explanation here:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7049#c6.

-- Paul

Pa...@Du... • +1.404.932.1355

522 Black Canyon Park, Canton GA 30114 USA

Re: [Log4cpp-devel] CategoryStream

[Cantor, S. <can...@os...>]

On 10/15/12 12:21 PM, "Paul Dugas" <pa...@du...> wrote:
>
>The compiler is balking about the "(*_buffer) << t;" statement in the
>operator<<() method template in CategoryStream saying that the call is
>ambiguous.  It's suggesting operator<<() routines for a whole mess of
>base types but doesn't see the one for my class.

That's a compiler strictness issue, it has to do with order of
declarations. You can find some info about it online with a bit of
searching (clang's web site mentions it).

-- Scott

Re: [Log4cpp-devel] CategoryStream

[Paul D. <pa...@du...>]

On Mon, Oct 15, 2012 at 12:46 PM, Cantor, Scott <can...@os...> wrote:
>
> On 10/15/12 12:21 PM, "Paul Dugas" <pa...@du...> wrote:
> >
> >The compiler is balking about the "(*_buffer) << t;" statement in the
> >operator<<() method template in CategoryStream saying that the call is
> >ambiguous.  It's suggesting operator<<() routines for a whole mess of
> >base types but doesn't see the one for my class.
>
> That's a compiler strictness issue, it has to do with order of
> declarations. You can find some info about it online with a bit of
> searching (clang's web site mentions it).

My search skills must be failing me.  Been looking into this for a
couple days with little luck.  Got a link or a search term I should be
using other than log4cpp and CategoryStream?

--Paul

[Log4cpp-devel] CategoryStream

[Paul D. <pa...@du...>]

I'm missing something.  I have a class like so:

    class foo
    {
      public:
        std::ostream& print(std::ostream& out) { out << m_value; return out; }

      private:
        int m_value;
    };

    inline std::ostream& operator<<(std::ostream& out, const foo& f)
    { return f.print(out); }

I am trying to stream an instance of foo to a CategoryStream.

    logger.infoStream() << "Foo: " << f;

The compiler is balking about the "(*_buffer) << t;" statement in the
operator<<() method template in CategoryStream saying that the call is
ambiguous.  It's suggesting operator<<() routines for a whole mess of
base types but doesn't see the one for my class.

This works:
    std::stringstream str;
    str << foo();

Why doesn't this?
    logger.infoStream() << foo();

Paul

[Log4cpp-devel] Fwd: Fixes suggested for 2940452 & 3109495

[Alexander <san...@gm...>]

Hello all,

Please find below suggested fixes for the memory leak bugs. If you feel
they are not quite right, please leave a feedback on that. Thanks.

https://sourceforge.net/tracker/?func=detail&aid=2940452&group_id=15190&atid=115190
https://sourceforge.net/tracker/?func=detail&aid=3109495&group_id=15190&atid=115190

Alexander.

Patch for 2940452 (releasing allocated static map if it has no elements
anymore):
*** ../../../log4cpp/src/Appender.cpp    2007-08-02 17:20:26.000000000 +0400
--- ./Appender.cpp    2012-08-01 20:29:09.784099993 +0400
***************
*** 38,43 ****
--- 38,46 ----
      void Appender::_removeAppender(Appender* appender) {
          threading::ScopedLock lock(_appenderMapMutex);
          _getAllAppenders().erase(appender->getName());
+     if(_getAllAppenders().size() == 0) {
+         delete _allAppenders; _allAppenders = 0;
+     }
      }

Patch for 3109495 (deleting allocated on the heap in the case exception is
thrown and caught):
*** ../../../log4cpp/src/PropertyConfiguratorImpl.cpp    2003-05-23
02:45:48.000000000 +0400
--- ./PropertyConfiguratorImpl.cpp    2012-08-01 20:39:04.489114799 +0400
***************
*** 268,273 ****
--- 268,274 ----

appender->setThreshold(Priority::getPriorityValue(thresholdName));
              }
          } catch(std::invalid_argument& e) {
+         delete appender;
              throw ConfigureFailure(std::string(e.what()) +
                  " for threshold of appender '" + appenderName + "'");
          }

[Log4cpp-devel] UNSUBSCRIBE

[Gareth F. <ear...@gm...>]

On Tue, 2012-04-17 at 12:06 +0200, Thomas Wabner wrote:
> Hi Alex,
> 
> thanks for your proposal!
> 
> Because I'am not longer involved in this project (see also the latest
> news), I have forewarded your request to the development maillinglist.
> There are some active developers.
> 
> I'am sure you can contribute to the project by fixing bugs or help to
> evolve the whole project.
> 
> Kind regards,
> 
> - waffel
> 
> Am 16.04.2012 16:05, schrieb Alex:
> > Hello Thomas,
> > 
> > Few years ago I ran into necessity to utilize well-designed
> > framework for logging purposes. After some search I chose
> > log4cpp from many others, and applied it successfully for
> > Windows apps. I faced some issues but got through them with
> > reasonable fixes. 
> > These days I got back to the question of logging framework
> > and after another investigation I again stopped at log4cpp,
> > but I still see quite large backlog on the tracker  page
> > (https://sourceforge.net/tracker/?group_id=15190&atid=115190).
> > The very first issue in backlog is which I faced immediately
> > on Ubuntu OS.
> > I see from the Changelog file that project has not moved
> > further since 2007, although there are still some bugs and
> > feature requests.
> > 
> > If you are planning to move this project forward, I would
> > like to contribute to the project as a developer.
> > I am software engineer with 10+ years of experience
> > primarily in C++ and Java plus other less frequently used
> > programming languages.
> > I could begin with fixing some bug requests, plus the isuues
> > which I have faced during personal usage of the framework.
> > If you are interested in this proposal, please let me know
> > your answer via email san...@gm... or sourceforge
> > contact page
> > https://sourceforge.net/sendmessage.php?touser=2383191
> > 
> > Best regards, 
> > Alexander Perepelkin.
> > 
> > 
> > --
> > This message was sent to your SourceForge.net email alias via the web mail form.  You may reply to this message directly, or via https://sourceforge.net/sendmessage.php?touser=2383191
> > To update your email alias preferences, please visit https://sourceforge.net/account
> 
> 
> ------------------------------------------------------------------------------
> Better than sec? Nothing is better than sec when it comes to
> monitoring Big Data applications. Try Boundary one-second 
> resolution app monitoring today. Free.
> http://p.sf.net/sfu/Boundary-dev2dev
> _______________________________________________
> Log4cpp-devel mailing list
> Log...@li...
> https://lists.sourceforge.net/lists/listinfo/log4cpp-devel

Re: [Log4cpp-devel] Contributing to the log4cpp project

[Thomas W. <tho...@ho...>]

Hi Alex,

thanks for your proposal!

Because I'am not longer involved in this project (see also the latest
news), I have forewarded your request to the development maillinglist.
There are some active developers.

I'am sure you can contribute to the project by fixing bugs or help to
evolve the whole project.

Kind regards,

- waffel

Am 16.04.2012 16:05, schrieb Alex:
> Hello Thomas,
> 
> Few years ago I ran into necessity to utilize well-designed
> framework for logging purposes. After some search I chose
> log4cpp from many others, and applied it successfully for
> Windows apps. I faced some issues but got through them with
> reasonable fixes. 
> These days I got back to the question of logging framework
> and after another investigation I again stopped at log4cpp,
> but I still see quite large backlog on the tracker  page
> (https://sourceforge.net/tracker/?group_id=15190&atid=115190).
> The very first issue in backlog is which I faced immediately
> on Ubuntu OS.
> I see from the Changelog file that project has not moved
> further since 2007, although there are still some bugs and
> feature requests.
> 
> If you are planning to move this project forward, I would
> like to contribute to the project as a developer.
> I am software engineer with 10+ years of experience
> primarily in C++ and Java plus other less frequently used
> programming languages.
> I could begin with fixing some bug requests, plus the isuues
> which I have faced during personal usage of the framework.
> If you are interested in this proposal, please let me know
> your answer via email san...@gm... or sourceforge
> contact page
> https://sourceforge.net/sendmessage.php?touser=2383191
> 
> Best regards, 
> Alexander Perepelkin.
> 
> 
> --
> This message was sent to your SourceForge.net email alias via the web mail form.  You may reply to this message directly, or via https://sourceforge.net/sendmessage.php?touser=2383191
> To update your email alias preferences, please visit https://sourceforge.net/account

Re: [Log4cpp-devel] Hi, log4cpp gives SEG11 at run time

[Konstantin L. <to....@gm...>]

On 04/09/2012 12:28 PM, islam wrote:
> hi again thank you for your kindly answer. What do you mean "broken
> build" ? if you mean compiling and building log4cpp properly, i used two
> ways :
> firstly i used direct FreeBSD ports,
> secondly i compiled log4cpp according to instructions, those are in website.

Broken build is situation when log4cpp complied with one set of flags 
and your application built with another. And these set are incompatible. 
That may cause very broad range of strange crashes like yours.
If you can write simple app that crush is the way you describe - I will 
look what might go wrong.

[Log4cpp-devel] Hi, log4cpp gives SEG11 at run time

[islam <isl...@su...>]

hi again thank you for your kindly answer. What do you mean "broken 
build" ? if you mean compiling and building log4cpp properly, i used two 
ways :
firstly i used direct FreeBSD ports,
secondly i compiled log4cpp according to instructions, those are in website.

-- 
Islam Yasar
Yazılım Geliştirme Mühendisi / Software Development Engineer
AR-GE Bölümü / R&D Division
SurGATE: West Coast Labs Premium Anti-Spam Certificated
Surgate Labs - http://www.surgate.com

Re: [Log4cpp-devel] Hi, log4cpp gives SEG11 at run time

[Konstantin L. <to....@gm...>]

On 04/06/2012 02:42 PM, islam wrote:

> (gdb) bt
> #0 0x00000008016cb40f in log4cpp::NDC::get () from
> /usr/local/lib/liblog4cpp.so.4
> #1 0x00000008016bcb60 in log4cpp::Category::_logUnconditionally2 () from
> /usr/local/lib/liblog4cpp.so.4
> #2 0x00000008016bc152 in log4cpp::Category::_logUnconditionally () from
> /usr/local/lib/liblog4cpp.so.4
> #3 0x00000008016bafbe in log4cpp::Category::error () from
> /usr/local/lib/liblog4cpp.so.4
> #4 0x0000000000420dcd in non-virtual thunk to
> boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::gregorian::bad_day_of_month>
>  >::rethrow() const ()
> #5 0x0000000000416d02 in ?? ()
> #6 0x000000000040c113 in ?? ()
> #7 0x000000000040d470 in ?? ()
>
> Something goes wrong in log4cpp::NDC::get (). Any help would be
> appreciated, thanks..

I don't think the problem caused by log4cpp. More likely - broken build. 
Can you provide minimal reproducable example?

[Log4cpp-devel] Hi, log4cpp gives SEG11 at run time

[islam <isl...@su...>]

Hi ,

I use log4cpp in my application. But at run time that gives segmentation 
fault. The System is FreeBSD 8.2 AMD64. The image of memory is below:

Loaded symbols for /lib/libcrypto.so.6
Reading symbols from /usr/local/lib/libintl.so.9...done.
Loaded symbols for /usr/local/lib/libintl.so.9
Reading symbols from /lib/libthr.so.3...done.
Loaded symbols for /lib/libthr.so.3
Reading symbols from /usr/local/lib/liblber-2.4.so.8...done.
Loaded symbols for /usr/local/lib/liblber-2.4.so.8
Reading symbols from /usr/local/lib/libiconv.so.3...done.
Loaded symbols for /usr/local/lib/libiconv.so.3
Reading symbols from /libexec/ld-elf.so.1...done.
Loaded symbols for /libexec/ld-elf.so.1
#0 0x00000008016cb40f in log4cpp::NDC::get () from 
/usr/local/lib/liblog4cpp.so.4
[New Thread 8029031c0 (LWP 100211)]
(gdb)
(gdb) bt
#0 0x00000008016cb40f in log4cpp::NDC::get () from 
/usr/local/lib/liblog4cpp.so.4
#1 0x00000008016bcb60 in log4cpp::Category::_logUnconditionally2 () from 
/usr/local/lib/liblog4cpp.so.4
#2 0x00000008016bc152 in log4cpp::Category::_logUnconditionally () from 
/usr/local/lib/liblog4cpp.so.4
#3 0x00000008016bafbe in log4cpp::Category::error () from 
/usr/local/lib/liblog4cpp.so.4
#4 0x0000000000420dcd in non-virtual thunk to 
boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::gregorian::bad_day_of_month> 
 >::rethrow() const ()
#5 0x0000000000416d02 in ?? ()
#6 0x000000000040c113 in ?? ()
#7 0x000000000040d470 in ?? ()

Something goes wrong in log4cpp::NDC::get (). Any help would be 
appreciated, thanks..

-- 
Islam Yasar
Yaz?l?m Gelis,tirme Mühendisi / Software Development Engineer
AR-GE Bölümü / R&D Division
SurGATE: West Coast Labs Premium Anti-Spam Certificated
Surgate Labs - http://www.surgate.com

Re: [Log4cpp-devel] Joining/merging the different log4c** implementations?

[S R. <kiw...@ma...>]

On Feb 5, 2012, at 02:42 , Konstantin Litvinenko wrote:

> On 02/02/2012 09:29 PM, Palic, Darko wrote:
>> Hello list,
>> 
>> I love log4j and log4net. And I would like to love log4c**. BUT, there
>> are at least 3 implementations I found for log4c**. Right now I am using
>> log4cxx from apache.org. I would like to be able to use the same logging
>> configurations for the my different application types. Isn't is perfect
>> to configure the logging in every major language in the same way? Isn't
>> it a really nice option to change the log level by changing a simple
>> config file? And isn't it nice to do it in the same way in java, c# or
>> c/c++?
> 
> Nothing preventing you from doing that. Just write for each package same 
> configuration code and have it. log4cpp van be configured in log4j way, 
> I think...
> 
>> 
>> But what I am trying to address with this mail?
>> log4cxx seems to be dead (no responses of developers, patches of user
>> aren't integrated for a long time, no commits since a very long time). I
>> decided to check also some alternative. I was looking into the log4cpp
>> repository and this also seems to be dead.
>> So I was wondering if there are more projects implementing log4c**. I've
>> found log4cplus, this seems to be a "sort of active".
> 
> These project seems to be dead because they riches a level of maturity 
> where 90% of user get what they need.
> 
>> Now I am wondering why it is necessary to maintain 3 projects which have
>> the same idea. Wouldn't it be more efficient for all projects, if the
>> available contributors and developers would maintain one and the same
>> project? So I am wondering if it wouldn't be possible to merge the
>> projects into one sourcetree? Lets say a new log4cpp which has all
>> features and enhancements of all log4c** projects.
> 
> Because that is what open source is :) Different people from different 
> reasons started these projects. Yes, they share same idea, but differs 
> in implementation. I think it common to have more then one project doing 
> the same thing in open source land.
> I don't think merging is good idea because all these project already 
> stops they evolution, have static user base. I don't think I'll move to 
> super-log4cpp just because it here.
> When I choose to stick log4cpp I look at functionality, which is the 
> same for all logging projects, AND simplicity of build. The major 
> problem is the BUILD. When you do only *nix development you just pick 
> some lib and go. BUT. When you do *nix/Windows development you are in 
> trouble. There is NO clearly defined way to packaging and distributing 
> libs on Windows. So you must choose some lib that have less dependencies 
> and can be build in a simple way.
> Unified packaging and distributing system for C/C++ is a major problem 
> for C/C++ developers and until it will be resolved there is no way to go…

We chose log4cpp over log4cxx and log4cplus as we needed something simple, understandable, and with the fewest reasonable dependancies, as we use it to log in hard real-time. Merging these 3 implementations together would most likely result in something unusable to us.

My 2c
S

Re: [Log4cpp-devel] Joining/merging the different log4c** implementations?

[Konstantin L. <to....@gm...>]

On 02/02/2012 09:29 PM, Palic, Darko wrote:
> Hello list,
>
> I love log4j and log4net. And I would like to love log4c**. BUT, there
> are at least 3 implementations I found for log4c**. Right now I am using
> log4cxx from apache.org. I would like to be able to use the same logging
> configurations for the my different application types. Isn't is perfect
> to configure the logging in every major language in the same way? Isn't
> it a really nice option to change the log level by changing a simple
> config file? And isn't it nice to do it in the same way in java, c# or
> c/c++?

Nothing preventing you from doing that. Just write for each package same 
configuration code and have it. log4cpp van be configured in log4j way, 
I think...

>
> But what I am trying to address with this mail?
> log4cxx seems to be dead (no responses of developers, patches of user
> aren't integrated for a long time, no commits since a very long time). I
> decided to check also some alternative. I was looking into the log4cpp
> repository and this also seems to be dead.
> So I was wondering if there are more projects implementing log4c**. I've
> found log4cplus, this seems to be a "sort of active".

These project seems to be dead because they riches a level of maturity 
where 90% of user get what they need.

>
> Now I am wondering why it is necessary to maintain 3 projects which have
> the same idea. Wouldn't it be more efficient for all projects, if the
> available contributors and developers would maintain one and the same
> project? So I am wondering if it wouldn't be possible to merge the
> projects into one sourcetree? Lets say a new log4cpp which has all
> features and enhancements of all log4c** projects.

Because that is what open source is :) Different people from different 
reasons started these projects. Yes, they share same idea, but differs 
in implementation. I think it common to have more then one project doing 
the same thing in open source land.
I don't think merging is good idea because all these project already 
stops they evolution, have static user base. I don't think I'll move to 
super-log4cpp just because it here.
When I choose to stick log4cpp I look at functionality, which is the 
same for all logging projects, AND simplicity of build. The major 
problem is the BUILD. When you do only *nix development you just pick 
some lib and go. BUT. When you do *nix/Windows development you are in 
trouble. There is NO clearly defined way to packaging and distributing 
libs on Windows. So you must choose some lib that have less dependencies 
and can be build in a simple way.
Unified packaging and distributing system for C/C++ is a major problem 
for C/C++ developers and until it will be resolved there is no way to go...

[Log4cpp-devel] Joining/merging the different log4c** implementations?

[Palic, D. <dar...@gm...>]

Hello list,

I love log4j and log4net. And I would like to love log4c**. BUT, there 
are at least 3 implementations I found for log4c**. Right now I am using 
log4cxx from apache.org. I would like to be able to use the same logging 
configurations for the my different application types. Isn't is perfect 
to configure the logging in every major language in the same way? Isn't 
it a really nice option to change the log level by changing a simple 
config file? And isn't it nice to do it in the same way in java, c# or 
c/c++?

But what I am trying to address with this mail?
log4cxx seems to be dead (no responses of developers, patches of user 
aren't integrated for a long time, no commits since a very long time). I 
decided to check also some alternative. I was looking into the log4cpp 
repository and this also seems to be dead.
So I was wondering if there are more projects implementing log4c**. I've 
found log4cplus, this seems to be a "sort of active".

Now I am wondering why it is necessary to maintain 3 projects which have 
the same idea. Wouldn't it be more efficient for all projects, if the 
available contributors and developers would maintain one and the same 
project? So I am wondering if it wouldn't be possible to merge the 
projects into one sourcetree? Lets say a new log4cpp which has all 
features and enhancements of all log4c** projects.

What can I provide?
I could provide Build-Servers and a Continuous Build environment for 
building and testing the project sources. I can also provide some 
patches against log4cxx. And I can provide my effort to for merging the 
projects into one log4cpp project.

Are here some developers and contributors which are willing to join the 
"merge-project" of the current log4c**?

bye,
Darko

Re: [Log4cpp-devel] Request for releasing Log4cpp

[Thomas W. <tho...@ho...>]

Hi Sameer,

as you have possible noted, I'am no longer maintining log4cpp ... I have
forewarded your request to the log...@li...
maillinglist ... hopefully someone else can help you.

Kind regards,

- waffel

Am 16.06.2011 08:57, schrieb sam...@us...:
> Hi Waffel,
> The version available for download is last updated in 2007-09-03.
> But there are lots of changed done till now. So i would like to have latest 
> stable version of log4cpp.
> 
> Can you please release a updated version of log4cpp ..?
> 
> Thanks 
> -Sameer 
> 
> 
> --
> This message has been sent to you, a registered SourceForge.net user,
> by another site user, through the SourceForge.net site.  This message
> has been delivered to your SourceForge.net mail alias.  You may reply
> to this message using the "Reply" feature of your email client, or
> using the messaging facility of SourceForge.net at:
> https://sourceforge.net/sendmessage.php?touser=3432890
> 

Re: [Log4cpp-devel] join

[Thomas W. <tho...@ho...>]

Hi,

Thanks for your mail.

As you may have read in the project news, I'am no longer active for 
log4cpp ... please ask on the official maillinglist.

I have cc this mail also to the maillinglist. Hope there is someone who 
can help you.

Kind regards,

- waffel

Am 31.03.2011 15:49, schrieb joa...@us...:
> hi. I am Joan in My work use a log4cpp. I want join you for
> work in log4cpp.
> I have to a lot of  examples and may be useful
>
> --
> This message has been sent to you, a registered SourceForge.net user,
> by another site user, through the SourceForge.net site.  This message
> has been delivered to your SourceForge.net mail alias.  You may reply
> to this message using the "Reply" feature of your email client, or
> using the messaging facility of SourceForge.net at:
> https://sourceforge.net/sendmessage.php?touser=3318315
>

Re: [Log4cpp-devel] Contributing to log4cpp

[Thomas W. <tho...@ho...>]

Hi,

Thanks for your mail.

As you may have read in the project news, I'am no longer active for 
log4cpp ... please ask on the official maillinglist (for example you can 
attach a patch to your mail, so a developer can apply the patch to the 
sources).

I have cc this mail also to the maillinglist. Hope there is someone who 
can help you.

Kind regards,

- waffel

Am 02.05.2011 10:36, schrieb yu...@us...:
> Hello,
> We are using log4cpp in a project at HP, and have been making
> several changes to the code. I want to check those changes into
> sourceforge so that they are available to the community, as well as to
> comply with the LGPL. I am new to open source development, so
> please tell me if there if I need to send the changes we made to your
> review and approval.
> Thanks,
> Yuval.
>
>
> --
> This message has been sent to you, a registered SourceForge.net user,
> by another site user, through the SourceForge.net site.  This message
> has been delivered to your SourceForge.net mail alias.  You may reply
> to this message using the "Reply" feature of your email client, or
> using the messaging facility of SourceForge.net at:
> https://sourceforge.net/sendmessage.php?touser=3360861
>

[Log4cpp-devel] Maintainer?

[S R. <kiw...@ma...>]

Does this project currently have a maintainer? 

We have a series of patches to log4cpp up to date with recent Linux and Mac distributions, support non-standard category factories, support microsecond level timestamp formatting, and support CMake on Linux/OSX/Win32 without needing any of the autotools-related files (eg config.h).

Is anyone interested in patches?
Stephen

Re: [Log4cpp-devel] Log file naming

[Konstantin L. <to....@gm...>]

On 19.04.2010 11:59, Gareth Foster wrote:
> So, for example, can do something like this?
>
> "
> log4j.appender.rootAppender.fileName=${programName}.log
> "
>
> Where ${programName} is the name of the program that reads the properties file?
>
    log4cpp can't configure himself use syntax you described. But if you generate config file from 
some template file where line

log4j.appender.rootAppender.fileName=${programName}.log

will be modified to look like

log4j.appender.rootAppender.fileName=MyApp.log

you will get what you want. Hope this helps.

Re: [Log4cpp-devel] Log file naming

[Gareth F. <ear...@go...>]

So, for example, can do something like this?

"
log4j.appender.rootAppender.fileName=${programName}.log
"

Where ${programName} is the name of the program that reads the properties file?

On 19 April 2010 09:54, Konstantin Litvinenko <to....@gm...> wrote:
> Gareth Foster пишет:
>> Hi,
>>
>> Is it possible to have a log file named after the program you are
>> running, so if the program is myApp, can I have it so that it will log
>> to myApp.log, by making changes in a properties file?
>>
>
>     Sure you can. Before you invoke configuring routine you make appropriate config file, with what
> you really need, and than run configuring routine to setup loggers/appendes/layouts. That all.
>     I am not sure I understand you problem clearly :).
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Log4cpp-devel mailing list
> Log...@li...
> https://lists.sourceforge.net/lists/listinfo/log4cpp-devel
>

Re: [Log4cpp-devel] Log file naming

[Konstantin L. <to....@gm...>]

On 19.04.2010 11:59, Gareth Foster wrote:
> So, for example, can do something like this?
> "
> log4j.appender.rootAppender.fileName=${programName}.log
> "
> Where ${programName} is the name of the program that reads the properties file?

   Now I understand what you want. No, there is no such functionality in 
PropertyConfiguratorImpl.cpp. Why do you need that? Have exactly same log config for multiple 
applications?

Re: [Log4cpp-devel] Log file naming

[Konstantin L. <to....@gm...>]

Gareth Foster пишет:
> Hi,
>
> Is it possible to have a log file named after the program you are
> running, so if the program is myApp, can I have it so that it will log
> to myApp.log, by making changes in a properties file?
>

     Sure you can. Before you invoke configuring routine you make appropriate config file, with what 
you really need, and than run configuring routine to setup loggers/appendes/layouts. That all.
     I am not sure I understand you problem clearly :).

Re: [Log4cpp-devel] confirm f82d5ea0f48556bc3b6ca21e67997fbb85258b7d

[Gareth F. <ear...@go...>]

Hi,

Is it possible to have a log file named after the program you are
running, so if the program is myApp, can I have it so that it will log
to myApp.log, by making changes in a properties file?

Many thanks,

Gaz

[Log4cpp-devel] looking for new maintainer

[waffel <wa...@us...>]

Hello log4cpp users and developers,

after some years of active maintainance of the log4cpp project with many 
fun it's now time to give back the maintainer position.

As my last activity I have moved the (hosted by me) wiki site to 
sourceforge which should be a better place.

Why? Live changes and I got a new position in my firm. Together with my 
family I didn't have enough time to do this job.

I hope there will be others to maintain the project.

It was a interesting time to develop and maintaining this project. Thanks!

- waffel

Re: [Log4cpp-devel] found memory leak in log4cpp

[waffel <wa...@us...>]

Hi,

ok ... I do not have anymore enough time to maintain log4cpp on a 
regular base .. but I can push out a security fix release the next days 
with your fix.

Can you please send me the patch as attachement or better would be to 
add an bug entry on our sourceforge page with your patch? I will also 
send your mail to our maillinglist so anybody covers the memory leak too.

Kind regards,

- waffel


Am 24.01.2010 22:54, schrieb cquike:
>
> Message body follows:
>
> Dear Thomas,
>   first of all sorry for not addressing this email to the
> right person. I  just
> saw it as the maintainer of the log4cpp library.
>   I have been using the library and have found a memory leak.
> Using the code snippet from
> http://developers.sun.com/solaris/articles/logging.html and
> running valgrind I get a memory leak report :
> ==31205== 24 bytes in 1 blocks are still reachable in loss
> record 1 of 1
> ==31205==    at 0x4025390: operator new(unsigned int)
> (vg_replace_malloc.c:214)
> ==31205==    by 0x4054E8B:
> log4cpp::Appender::_getAllAppenders() (in
> /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x405542F:
> log4cpp::Appender::_addAppender(log4cpp::Appender*) (in
> /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x40556B1:
> log4cpp::Appender::Appender(std::string const&) (in
> /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x40562C8:
> log4cpp::AppenderSkeleton::AppenderSkeleton(std::string
> const&) (in /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x405B4CB:
> log4cpp::LayoutAppender::LayoutAppender(std::string const&)
> (in /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x405B7FF:
> log4cpp::FileAppender::FileAppender(std::string const&,
> std::string const&, bool, unsigned int) (in
> /usr/lib/liblog4cpp.so.5.0.5)
> ==31205==    by 0x8048F6A: main (testlog.cpp:13)
>   The memory leak is related to the Appender class, namely
> the _allAppenders variable. I solved the problem using a
> std::auto_ptr.
> The patch is simple:
> --- log4cpp-1.0/src/Appender.cpp	2007-08-02
> 15:20:26.000000000 +0200
> +++ log4cpp-1.0.modified/src/Appender.cpp	2010-01-23
> 23:44:41.000000000 +0100
> @@ -11,13 +11,13 @@
>   #include<log4cpp/Appender.hh>
>
>   namespace log4cpp {
> -    Appender::AppenderMap* Appender::_allAppenders = 0;
> +    std::auto_ptr<Appender::AppenderMap>
> Appender::_allAppenders(0);
>       threading::Mutex Appender::_appenderMapMutex;
>
>       /* assume _appenderMapMutex locked */
>       Appender::AppenderMap&  Appender::_getAllAppenders() {
> -        if (!_allAppenders)
> -            _allAppenders = new Appender::AppenderMap();
> +        if (_allAppenders.get() == 0)
> +            _allAppenders.reset( new Appender::AppenderMap());
>
>           return *_allAppenders;
>       }
>
> You may find the patch in attachment memory_leak_fixed.patch.
>   I would like to know if there are plans to have a new
> release of the library (hopefully with this problem fixed).
> If you are not longer the maintainer I would appreciate if
> you point me who should I address to.
>
>   Thank you very much in advance,
>   Enrique
>
> --
> This message has been sent to you, a registered SourceForge.net user,
> by another site user, through the SourceForge.net site.  This message
> has been delivered to your SourceForge.net mail alias.  You may reply
> to this message using the "Reply" feature of your email client, or
> using the messaging facility of SourceForge.net at:
> https://sourceforge.net/sendmessage.php?touser=1178969
>