You are viewing a plain text version of this content. The canonical link for it is here.
Posted to log4cxx-dev@logging.apache.org by Andreas Fester <af...@apache.org> on 2006/01/30 16:12:53 UTC
Intended commits for L7dTestCase (LOGCXX-125)
This patch fixes the localized logging which was still #if 0'ed out.
To read the resource files, a FileInputStream class was added.
This aligns the class structure better with the java world,
instead of having read/write methods directly on the File class.
Best Regards,
Andreas
Re: Intended commits for L7dTestCase (LOGCXX-125)
Posted by Curt Arnold <ca...@apache.org>.
On Mar 8, 2006, at 1:34 PM, Andreas Fester wrote:
> Ok, I would like to retry this one ;-)
>
> I added an InputStreamReader to split the reader and stream
> hierarchy. The classes should now be similar to java.
> InputStreamReader converts from char to LogString.
>
> Thanks,
>
> Andreas
Go ahead and commit what you have and will review after commit.
Re: Intended commits for L7dTestCase (LOGCXX-125)
Posted by Andreas Fester <af...@apache.org>.
Ok, I would like to retry this one ;-)
I added an InputStreamReader to split the reader and stream
hierarchy. The classes should now be similar to java.
InputStreamReader converts from char to LogString.
Thanks,
Andreas
On Jan 30, 2006, at 9:12 AM, Andreas Fester wrote:
> This patch fixes the localized logging which was still #if 0'ed out.
>
> To read the resource files, a FileInputStream class was added.
> This aligns the class structure better with the java world,
> instead of having read/write methods directly on the File class.
>
log4cxx::InputStream and log4cxx::FileInputStream in the patch work
on characters and are therefore really aligned with java.io.Reader
and java.io.FileReader, not their java.io namesakes. If there is to
be a log4cxx::InputStream its methods should be using
std::vector<unsigned char> instead of LogString.
I'd support either simply renaming the InputStream and
FileInputStream classes to Reader and FileReader and accepting the
patch as a basis for further work or let you try to split out the raw
byte-level IO classes from the character based classes.
log4cxx on HP UX with aCC compiler (was Re: Intended commits for L7dTestCase (LOGCXX-125))
Posted by Curt Arnold <ca...@apache.org>.
On Feb 4, 2006, at 2:22 PM, Narasimha Rao wrote:
> Dear Arnold,
> DO you know anybody using Log4CXX on HP UX with aCC compiler.
> I need to compile Log4CXX with aCC compiler on HP UX 11.11.
> I need a make file to compile the code and do we need any specific
> libraries lile xml2 etc.. before compile?
> Can you pleae help me?
> While compiling i got one error like error before or after ' ( '
> in socket.h line 248 etc...
> Thank you,
> Nara
>
I am not aware of anyone using log4cxx on HP UX with a aCC
compiler. If anyone else on the list knows, please speak up.
Do you have a Java VM on the machine? Have you tried using the Ant
based build?
1. Download Ant 1.6.5 from http://ant.apache.org
2. Checkout CVS HEAD of cpptasks (see http://sf.net/projects/ant-
contrib) and build
cvs -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/ant-contrib login
cvs -z3 -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/ant-contrib
co -P cpptasks
export PATH /home/foo/apache-ant-1.6.5/bin;$PATH$
cd cpptasks
ant
3. Put cpptasks.jar on classpath
export CLASSPATH /home/foo/cpptasks/build/lib/cpptasks.jar
4. Build log4cxx
I believe that the default mirror used for cppunit is no longer
responding. You can override it, but finding a mirror name from
http://sf.net/projects/cppunit files page and then specifying
cd log4cxx
ant -Dcompiler=aCC -Dcppunit.mirror=http://example.com/example/
sourceforge -l build.log
If you do get a compiler error on aCC, file a JIRA issue and attach
the full build session log (specified with the -l option switch)
Re: Intended commits for L7dTestCase (LOGCXX-125)
Posted by Narasimha Rao <t_...@yahoo.com>.
Dear Arnold,
DO you know anybody using Log4CXX on HP UX with aCC compiler.
I need to compile Log4CXX with aCC compiler on HP UX 11.11.
I need a make file to compile the code and do we need any specific libraries lile xml2 etc.. before compile?
Can you pleae help me?
While compiling i got one error like error before or after ' ( ' in socket.h line 248 etc...
Thank you,
Nara
Curt Arnold <ca...@apache.org> wrote:
On Jan 30, 2006, at 9:12 AM, Andreas Fester wrote:
> This patch fixes the localized logging which was still #if 0'ed out.
>
> To read the resource files, a FileInputStream class was added.
> This aligns the class structure better with the java world,
> instead of having read/write methods directly on the File class.
>
log4cxx::InputStream and log4cxx::FileInputStream in the patch work
on characters and are therefore really aligned with java.io.Reader
and java.io.FileReader, not their java.io namesakes. If there is to
be a log4cxx::InputStream its methods should be using
std::vector instead of LogString.
I'd support either simply renaming the InputStream and
FileInputStream classes to Reader and FileReader and accepting the
patch as a basis for further work or let you try to split out the raw
byte-level IO classes from the character based classes.
---------------------------------
Brings words and photos together (easily) with
PhotoMail - it's free and works with Yahoo! Mail.
Re: Intended commits for L7dTestCase (LOGCXX-125)
Posted by Curt Arnold <ca...@apache.org>.
On Jan 30, 2006, at 9:12 AM, Andreas Fester wrote:
> This patch fixes the localized logging which was still #if 0'ed out.
>
> To read the resource files, a FileInputStream class was added.
> This aligns the class structure better with the java world,
> instead of having read/write methods directly on the File class.
>
log4cxx::InputStream and log4cxx::FileInputStream in the patch work
on characters and are therefore really aligned with java.io.Reader
and java.io.FileReader, not their java.io namesakes. If there is to
be a log4cxx::InputStream its methods should be using
std::vector<unsigned char> instead of LogString.
I'd support either simply renaming the InputStream and
FileInputStream classes to Reader and FileReader and accepting the
patch as a basis for further work or let you try to split out the raw
byte-level IO classes from the character based classes.