axis-c-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adrian Dick <adrian.d...@uk.ibm.com>
Subject Re: Axis C++ Solaris port
Date Wed, 08 Sep 2004 14:19:33 GMT




Hi,

It would be advantageous to include the Solaris compile and link flags into
the ant build script.  We found this got us round a number of difficulties
with libtool, etc when porting to AIX.

I'm willing to assist with this, but I don't have access to a solaris
machine.

Regards,
Adrian
_______________________________________
Adrian Dick (adrian.dick@uk.ibm.com)
WebSphere MQ and ESB Development
Tel: +44-(0)-1962-819212
Notes: Adrian Dick/UK/IBM@IBMGB


                                                                           
             Valentine                                                     
             Kouznetsov                                                    
             <vkuznet@yahoo.co                                          To 
             m>                        Apache AXIS C Developers List       
                                       <axis-c-dev@ws.apache.org>          
             08-09-04 15:14                                             cc 
                                                                           
                                                                   Subject 
             Please respond to         Re: Axis C++ Solaris port           
              "Apache AXIS C                                               
             Developers List"                                              
                                                                           
                                                                           
                                                                           
                                                                           




Hi Damitha,
I'm not sure if option --with-solaris would be
valuable. May be as a temporary solution. I had so
many problems with libtool/aclocal/autoheader/automake
tools. By default the didn't produce desired
Makefiles. Configure fail on the tests and I was
forced to tweake it as well.
My next task is to fix this problem.
Regarding SolarisDefs.hpp, yes it's only happens if
xerces is enable. But I think we need it since users
may be willing to compile Axis/Xerces on Solaris. Even
I didn't use it I compile it.
Valentin.

--- damitha kumarage <damitha@opensource.lk> wrote:

> Hi Valentine
>
> > Dear Axis developers,
> > I successfully ported Axis to Solaris and verified
> > that my client code is working properly similar to
> > Linux. I've spent some time to figure out
> appropriate
> > options for Solaris platform which will allow Axis
> to
> > compile. I used Solaris 2.8, with native CC/cc
> C++/c
> > compilers. The compiler options was choosen to
> allow
> > global instantiation of templates.
> >
> > Here they are:
> > CXXFLAGS=" -KPIC -xildoff -pta -instances=global "
> > CFLAGS="-DENABLE_AXIS_EXCEPTION
> -DENABLE_AXISTRACE"
> > CPPFLAGS="-DENABLE_AXIS_EXCEPTION
> -DENABLE_AXISTRACE"
> > LDFLAGS = "-Qoption ld -zmuldefs"
> > and need to be setup in configure.ac
>
> Is it OK if i add a comiler option to support
> different platforms.
> eg: configure --with-solaris etc
>
> >
> > I used libtool-1.5.8, autoconf-2.59,
> automake-1.9.1
> >
> > All patches done wrt axis-c-src-1-2-linux (sorry I
> > just don't have time to test development release).
> > I also created SolarisDefs.hpp file similar to
> > LinuxDefs.hpp. I just copied content of endian.h
> and
> > LinuxDefs.hpp into SolarisDefs.hpp and it works
> (on
> > Solaris there is no endian.h).
> > Finally, I was need to rebuild libaxis_expat.so
> and
> > add there StartElement.o, Event.o, Element.o.
>
> LinuxDefs.hpp is related to xercesc. so I guess
> SolarisDefs.hpp is
> not needed if you use expat
>
>  It was
> > necessary to plug Axis into our 700+ package
> framework
> > (which is based on shared libraries). I did it
> > manually and need to verify if changes would be
> > necessary for Axis Makefile.am.
> >
> > I hope you will accept my changes and include them
> in
> > CVS tree as well as in forthcoming release. I'll
> be
> > happy to test Axis in a future on Solaris when new
> > release will come.
> >
> > Valentin.
> > P.S. The way we use Axis is to access Web Service
> > through our framework and legacy C++ application.
> No
> > tests done on the server side.
> >
> >
> >
> > __________________________________
> > Do you Yahoo!?
> > Read only the mail you want - Yahoo! Mail
> SpamGuard.
> > http://promotions.yahoo.com/new_mail
>
>
>
----------------------------------------------------------------------------

> ----
>
>
> > ---
>
original/axis-c-src-1-2-linux/src/common/AxisTime.cpp
> 2004-07-09
> 03:22:04.000000000 -0400
> > +++
> solaris/axis-c-src-1-2-linux/src/common/AxisTime.cpp
> 2004-08-20
> 13:03:08.000000000 -0400
> > @@ -18,7 +18,7 @@
> >   * @author Damitha Kumarage
> (damitha@opensource.lk,
> damitha@jkcsworld.com)
> >   *
> >   */
> > -
> > +#include <math.h>
> >  #include <stdio.h>
> >  #include <axis/server/AxisTime.h>
> >  #include "AxisUtils.h"
> > @@ -250,18 +250,18 @@
> >              m_TM.tm_year -= 1900;
> >              m_TM.tm_mon--;
> >              m_TM.tm_isdst = -1;
> > -#if !defined(WIN32) && !defined(AIX)
> > +#if !defined(WIN32) && !defined(AIX) &&
> !defined(__sun)
> >              m_TM.tm_zone = NULL;
> >              m_TM.tm_gmtoff = -1;
> >  #endif
> > -            cTemp2 = strpbrk (m_sValue.c_str (),
> "T");
> > +            cTemp2 = const_cast<char*>(strpbrk
> (m_sValue.c_str (),
> "T"));
> >              cTemp3 = strrchr (cTemp2, ':');
> >              cTemp3[0] = '\0';
> >              intLen = strlen (cTemp2);
> >              cTemp3[0] = ':';
> >
> >              /*if the timezone is represented
> adding 'Z' at the end */
> > -            if ((cTemp = strpbrk (m_sValue.c_str
> (), "Z")) != NULL)
> > +            if ((cTemp =
> const_cast<char*>(strpbrk (m_sValue.c_str
> (),
> "Z"))) != NULL)
> >              {
> >                  time_t temp = mktime (&m_TM);
> >                  if (temp == -1)
> > @@ -296,7 +296,7 @@
> >                  time_t t = mktime (&m_TM);
> >                  if (t == -1)
> >                      return AXIS_FAIL;
> > -                t = abs (t - d);
> > +                t = fabs (t - d);
> >                  pTm = gmtime (&t);
> >                  memcpy (&m_TMUTC, pTm, sizeof
> (tm));
> >              }
> > @@ -329,14 +329,14 @@
> >              m_TM.tm_min = 0;
> >              m_TM.tm_sec = 0;
> >              m_TM.tm_isdst = -1;
> > -#if !defined(WIN32) && !defined(AIX)
> > +#if !defined(WIN32) && !defined(AIX) &&
> !defined(__sun)
> >              m_TM.tm_zone = NULL;
> >              m_TM.tm_gmtoff = -1;
> >  #endif
> > -            cTemp3 = strpbrk (m_sValue.c_str (),
> ":");
> > +            cTemp3 = const_cast<char*>(strpbrk
> (m_sValue.c_str (),
> ":"));
> >
> >              /* if the timezone is represented
> adding 'Z' at the end
> */
> > -            if ((cTemp = strpbrk (m_sValue.c_str
> (), "Z")) != NULL)
> > +            if ((cTemp =
> const_cast<char*>(strpbrk (m_sValue.c_str
> (),
> "Z"))) != NULL)
> >              {
> >                  time_t timeInSecs = mktime
> (&m_TM);
> >                  if (timeInSecs == -1)
> > @@ -346,9 +346,9 @@
> >              }
> >              else if (cTemp3 != NULL)
> >              {
> > -                cUtc = strrchr (m_sValue.c_str
> (), '+');
> > +                cUtc = const_cast<char*>(strrchr
> (m_sValue.c_str (),
> '+'));
> >                  if (cUtc == NULL)
> > -                    cUtc = strrchr
> (m_sValue.c_str (), '-');
> > +                    cUtc =
> const_cast<char*>(strrchr (m_sValue.c_str
> (),
> '-'));
> >                  time_t timeInSecs = mktime
> (&m_TM);
> >                  if (timeInSecs == -1)
> >                      return AXIS_FAIL;
> > @@ -368,7 +368,7 @@
> >                  time_t t = mktime (&m_TM);
> >                  if (t == -1)
> >                      return AXIS_FAIL;
> > -                t = abs (t - d);
> > +                t = fabs (t - d);
> >                  pTm = gmtime (&t);
> >                  memcpy (&m_TMUTC, pTm, sizeof
> (tm));
> >              }
> > @@ -399,7 +399,7 @@
>
=== message truncated ===




__________________________________
Do you Yahoo!?
Y! Messenger - Communicate in real time. Download now.
http://messenger.yahoo.com



Mime
View raw message