quickfix-developers — QuickFIX developers forum

Re: [Quickfix-developers] Session Problem

[Oren M. <or...@qu...>]

Call Session::logout to fully disable the session, and Session::logon 
when you want to reactivate it.  In the future we are looking at was to 
specify multiple time ranges for a day from the configuration file.

--oren

On Sep 16, 2004, at 5:12 AM, Shamanth wrote:

> Hi
>
> We have a provider, whose Session starts at "22:30:00" GMT and stops 
> at "21:00:00"GMT, but the provider resets the sequence numbers only at 
> "00:00:00" GMT.
>
>  So we have set our StartTime and EndTime at 00:00:00, so that the 
> sequence numbers get reset at that time.
>
>  But the problem is that, our quickfix instance will keep sending 
> Logon requests to the Provider between "21:00:00GMT" to "22:30:00GMT", 
> i.e. when the provider is down. We want to avoid this.
>
> Can anyone suggest a solution for this.
>
> thanks
>  R Shamanth

Re: [Quickfix-developers] Unit test C++ quickfix 1.9.2 with MySQL on Linux

[Oren M. <or...@qu...>]

Hans,

Did you build the quickfix/src/sql/crate_mysql.sh script?

--oren

On Sep 16, 2004, at 3:56 AM, Hans Bausewein wrote:

> QuickFIX Documentation:  
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ:  
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> The first unit and acceptance tests were all (C++ and Java) without
> failures, but now I've configured with mysql:
>
> ./configure --with-mysql=/opt/mysql --enable-callstack
>
> and I got this output from the 'test/runut' unit test:
>
> <ut>
>   <output>
> ....................................................................... 
> .........
> ...............FFFF.............................................
>   </output>
>   <results total="144" failures="4">
>     <failure  line= "0" file= "unknown">
>       <test>
>         <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
>       </test>
>       <text>
>         <![CDATA[ assert(no futher information available)]]>
>       </text>
>     </failure>
>     <failure  line= "0" file= "unknown">
>       <test>
>         <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
>       </test>
>       <text>
>         <![CDATA[ assert(no futher information available)]]>
>       </text>
>     </failure>
>     <failure  line= "0" file= "unknown">
>       <test>
>         <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
>       </test>
>       <text>
>         <![CDATA[ assert(no futher information available)]]>
>       </text>
>     </failure>
>     <failure  line= "0" file= "unknown">
>       <test>
>         <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
>       </test>
>       <text>
>         <![CDATA[ assert(no futher information available)]]>
>       </text>
>     </failure>
>   </results>
> </ut>
>
> What failed here: the test or the application?
>
> 'test/runat' and both Java unit and acceptance tests had no failures.
>
>
> Hans Bausewein
>
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

[Quickfix-developers] Session Problem

[Shamanth <sha...@in...>]

Hi

We have a provider, whose Session starts at "22:30:00" GMT and stops at =
"21:00:00"GMT, but the provider resets the sequence numbers only at =
"00:00:00" GMT.=20
So we have set our StartTime and EndTime at 00:00:00, so that the =
sequence numbers get reset at that time.=20

But the problem is that, our quickfix instance will keep sending Logon =
requests to the Provider between "21:00:00GMT" to "22:30:00GMT", i.e. =
when the provider is down. We want to avoid this.

Can anyone suggest a solution for this.

thanks
R Shamanth

[Quickfix-developers] Unit test C++ quickfix 1.9.2 with MySQL on Linux

[Hans B. <han...@ma...>]

The first unit and acceptance tests were all (C++ and Java) without
failures, but now I've configured with mysql:

./configure --with-mysql=/opt/mysql --enable-callstack

and I got this output from the 'test/runut' unit test:

<ut>
  <output>
................................................................................
...............FFFF.............................................
  </output>
  <results total="144" failures="4">
    <failure  line= "0" file= "unknown">
      <test>
        <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
      </test>
      <text>
        <![CDATA[ assert(no futher information available)]]>
      </text>
    </failure>
    <failure  line= "0" file= "unknown">
      <test>
        <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
      </test>
      <text>
        <![CDATA[ assert(no futher information available)]]>
      </text>
    </failure>
    <failure  line= "0" file= "unknown">
      <test>
        <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
      </test>
      <text>
        <![CDATA[ assert(no futher information available)]]>
      </text>
    </failure>
    <failure  line= "0" file= "unknown">
      <test>
        <![CDATA[ PN7CPPTest4TestIN3FIX12MessageStoreEEE]]>
      </test>
      <text>
        <![CDATA[ assert(no futher information available)]]>
      </text>
    </failure>
  </results>
</ut>

What failed here: the test or the application?

'test/runat' and both Java unit and acceptance tests had no failures.


Hans Bausewein

[Quickfix-developers] FIX.FieldMap.=(FieldMap* , FieldMap* ) Error

[Ramprakash U. <ram...@ca...>]

Hi,

I get this error now and then (not all the times)

at FIX.FieldMap.=3D(FieldMap* , FieldMap* )
   at Application.toApp(Application* , Message* message, SessionID*
sessionID)
   at FIX.Session.sendToTarget(Message* , SenderCompID* , TargetCompID* =
)
   at QuickFix.Session.sendToTarget(Message message, String =
senderCompID,
String targetCompID)
   at CabQuickFixServices.QFApplication.newOrderSingle(mdlOrder =
oQFOrder) in
D:\PSO-Development\Projects\WinApps\QuickFixServices\QFApplication.vb:lin=
e
467|09:42|ERROR|Object reference not set to an instance of an object.

Usually this error appears on my log when I try to send NewOrderSingle =
with
heavy inflow of orders. This error seems to be appearing with versions =
1.8.0
and 1.9.1

Could not make it is a bug/ problem in my coding?! I attached the file
QFApplication.vb for reference.

I use VS.NET 2003/Windows2003/QF1.8.0/QF1.9.1

TIA,
Ramprakash Umapathy
*******************
Consultant
Banca Caboto s.p.a
RCM, Brokerage e Canali
Via Arrigo Boito, 7
20121 Milano (Italy)
ram...@ca...
Ph : +39 02 8021 5754

Re: [Quickfix-developers] Re: [Quickfix-users] QuickFIX 1.9.1 Available

[Oren M. <or...@qu...>]

I put up a 1.9.2 version for unix on the download page: 
http://www.quickfixengine.org/download.html

--oren

On Sep 15, 2004, at 6:23 AM, Hans Bausewein wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: 
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> On Wed, 2004-09-15 at 13:12, Joerg Thoennes wrote:
>> Hans Bausewein wrote:
>>> On Wed, 2004-09-15 at 12:30, Hans Bausewein wrote:
>>>>
>>>> I get a linker problem:
>>>>
>>>> Exception in thread "main" java.lang.UnsatisfiedLinkError:
>>>> /home/hans/src/quickfix/src/java/.libs/libquickfix_jni.so.7.1.0:
>>>> /home/hans/src/quickfix/src/C++/.libs/libquickfix.so.3: undefined
>>>> symbol: _ZNKSt11logic_error4whatEv
>>>>
>>>>
>>>> I also get many libraries with different version numbers:
>>>> libquickfix_jni.la        libquickfix_jni.so.8      libquickfix.so.3
>>>> libquickfix_jni.so        libquickfix_jni.so.8.0.0  
>>>> libquickfix.so.3.1.0
>>>> libquickfix_jni.so.7      libquickfix.la            libquickfix.so.4
>>>> libquickfix_jni.so.7.1.0  libquickfix.so            
>>>> libquickfix.so.4.0.0
>>>>
>>>> The 1.9.1 libraries are libquickfix.so.3.1.0 and
>>>> libquickfix_jni.so.7.1.0, while the 1.9.0 libraries had newer 
>>>> version
>>>> numbers: libquickfix.so.4.0.0 and libquickfix_jni.so.8.0.0 .
>>>>
>>>>
>>>> I'll try to delete the 1.9.0 libraries and recompile.
>>>>
>>>> OS = RedHat Linux 8.0 (kernel 2.4.20)
>>
>> Which GCC version do you use?
>>
>
> "gcc --version" returns:
> gcc (GCC) 3.2.3 20030502 (Red Hat Linux 3.2.3-24)
>
> "rpm -qv gcc" returns "gcc-3.2.3-24"
>
>
> 1.9.0 worked fine (both the Java banzai app and the unit and acceptance
> tests).
>
> Hans
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: thawte's Crypto Challenge Vl
> Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam
> Camcorder. More prizes in the weekly Lunch Hour Challenge.
> Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

[Quickfix-developers] Bug Tracker for QuickFIX

[Oren M. <or...@qu...>]

A bug tracker is now available at 
http://www.quickfixengine.org/bugtracker

--oren

[Quickfix-developers] Re: [Quickfix-users] QuickFIX 1.9.1 Available

[Oren M. <or...@qu...>]

Indeed.  The versions were incorrectly incremented in autotools causing 
it to believe it is backwards compatible which is why the versions were 
knocked back.  I'll fix this and repackage.

--oren

On Sep 15, 2004, at 6:23 AM, Hans Bausewein wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: 
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> On Wed, 2004-09-15 at 13:12, Joerg Thoennes wrote:
>> Hans Bausewein wrote:
>>> On Wed, 2004-09-15 at 12:30, Hans Bausewein wrote:
>>>>
>>>> I get a linker problem:
>>>>
>>>> Exception in thread "main" java.lang.UnsatisfiedLinkError:
>>>> /home/hans/src/quickfix/src/java/.libs/libquickfix_jni.so.7.1.0:
>>>> /home/hans/src/quickfix/src/C++/.libs/libquickfix.so.3: undefined
>>>> symbol: _ZNKSt11logic_error4whatEv
>>>>
>>>>
>>>> I also get many libraries with different version numbers:
>>>> libquickfix_jni.la        libquickfix_jni.so.8      libquickfix.so.3
>>>> libquickfix_jni.so        libquickfix_jni.so.8.0.0  
>>>> libquickfix.so.3.1.0
>>>> libquickfix_jni.so.7      libquickfix.la            libquickfix.so.4
>>>> libquickfix_jni.so.7.1.0  libquickfix.so            
>>>> libquickfix.so.4.0.0
>>>>
>>>> The 1.9.1 libraries are libquickfix.so.3.1.0 and
>>>> libquickfix_jni.so.7.1.0, while the 1.9.0 libraries had newer 
>>>> version
>>>> numbers: libquickfix.so.4.0.0 and libquickfix_jni.so.8.0.0 .
>>>>
>>>>
>>>> I'll try to delete the 1.9.0 libraries and recompile.
>>>>
>>>> OS = RedHat Linux 8.0 (kernel 2.4.20)
>>
>> Which GCC version do you use?
>>
>
> "gcc --version" returns:
> gcc (GCC) 3.2.3 20030502 (Red Hat Linux 3.2.3-24)
>
> "rpm -qv gcc" returns "gcc-3.2.3-24"
>
>
> 1.9.0 worked fine (both the Java banzai app and the unit and acceptance
> tests).
>
> Hans
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: thawte's Crypto Challenge Vl
> Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam
> Camcorder. More prizes in the weekly Lunch Hour Challenge.
> Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

Re: [Quickfix-developers] Re: [Quickfix-users] QuickFIX 1.9.1 Available

[Hans B. <han...@ma...>]

On Wed, 2004-09-15 at 13:12, Joerg Thoennes wrote:
> Hans Bausewein wrote:
> > On Wed, 2004-09-15 at 12:30, Hans Bausewein wrote:
> >>
> >>I get a linker problem:
> >>
> >>Exception in thread "main" java.lang.UnsatisfiedLinkError:
> >>/home/hans/src/quickfix/src/java/.libs/libquickfix_jni.so.7.1.0:
> >>/home/hans/src/quickfix/src/C++/.libs/libquickfix.so.3: undefined
> >>symbol: _ZNKSt11logic_error4whatEv
> >>
> >>
> >>I also get many libraries with different version numbers:
> >>libquickfix_jni.la        libquickfix_jni.so.8      libquickfix.so.3
> >>libquickfix_jni.so        libquickfix_jni.so.8.0.0  libquickfix.so.3.1.0
> >>libquickfix_jni.so.7      libquickfix.la            libquickfix.so.4
> >>libquickfix_jni.so.7.1.0  libquickfix.so            libquickfix.so.4.0.0
> >>
> >>The 1.9.1 libraries are libquickfix.so.3.1.0 and
> >>libquickfix_jni.so.7.1.0, while the 1.9.0 libraries had newer version
> >>numbers: libquickfix.so.4.0.0 and libquickfix_jni.so.8.0.0 .
> >>
> >>
> >>I'll try to delete the 1.9.0 libraries and recompile.
> >>
> >>OS = RedHat Linux 8.0 (kernel 2.4.20)
> 
> Which GCC version do you use?
> 

"gcc --version" returns:
gcc (GCC) 3.2.3 20030502 (Red Hat Linux 3.2.3-24)

"rpm -qv gcc" returns "gcc-3.2.3-24"


1.9.0 worked fine (both the Java banzai app and the unit and acceptance
tests). 

Hans

Re: [Quickfix-developers] Re: [Quickfix-users] QuickFIX 1.9.1 Available

[Joerg T. <Joe...@ma...>]

Hans Bausewein wrote:
> On Wed, 2004-09-15 at 12:30, Hans Bausewein wrote:
>>
>>I get a linker problem:
>>
>>Exception in thread "main" java.lang.UnsatisfiedLinkError:
>>/home/hans/src/quickfix/src/java/.libs/libquickfix_jni.so.7.1.0:
>>/home/hans/src/quickfix/src/C++/.libs/libquickfix.so.3: undefined
>>symbol: _ZNKSt11logic_error4whatEv
>>
>>
>>I also get many libraries with different version numbers:
>>libquickfix_jni.la        libquickfix_jni.so.8      libquickfix.so.3
>>libquickfix_jni.so        libquickfix_jni.so.8.0.0  libquickfix.so.3.1.0
>>libquickfix_jni.so.7      libquickfix.la            libquickfix.so.4
>>libquickfix_jni.so.7.1.0  libquickfix.so            libquickfix.so.4.0.0
>>
>>The 1.9.1 libraries are libquickfix.so.3.1.0 and
>>libquickfix_jni.so.7.1.0, while the 1.9.0 libraries had newer version
>>numbers: libquickfix.so.4.0.0 and libquickfix_jni.so.8.0.0 .
>>
>>
>>I'll try to delete the 1.9.0 libraries and recompile.
>>
>>OS = RedHat Linux 8.0 (kernel 2.4.20)

Which GCC version do you use?

> This happened when running the "test/runat_java 10005" command.
> 
> bin/run_banzai resulted in a similar error.

Cheers, Jörg

-- 
Joerg Thoennes
                                http://macd.com
Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen

[Quickfix-developers] Re: [Quickfix-users] QuickFIX 1.9.1 Available

[Hans B. <han...@ma...>]

On Wed, 2004-09-15 at 12:30, Hans Bausewein wrote:
> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
> 
> On Tue, 2004-09-14 at 21:55, Oren Miller wrote:
> > QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
> > QuickFIX FAQ: http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> > QuickFIX Support: http://www.quickfixengine.org/services.html
> > 
> > You know where, but in case you don't:  http://www.quickfixengine.org/
> > 
> > This fixes some problems reported with 1.9.0.  You should get this if 
> > you are using MySQL, STLPort, or currencies.  
> 
> I get a linker problem:
> 
> Exception in thread "main" java.lang.UnsatisfiedLinkError:
> /home/hans/src/quickfix/src/java/.libs/libquickfix_jni.so.7.1.0:
> /home/hans/src/quickfix/src/C++/.libs/libquickfix.so.3: undefined
> symbol: _ZNKSt11logic_error4whatEv
> 
> 
> I also get many libraries with different version numbers:
> libquickfix_jni.la        libquickfix_jni.so.8      libquickfix.so.3
> libquickfix_jni.so        libquickfix_jni.so.8.0.0  libquickfix.so.3.1.0
> libquickfix_jni.so.7      libquickfix.la            libquickfix.so.4
> libquickfix_jni.so.7.1.0  libquickfix.so            libquickfix.so.4.0.0
> 
> The 1.9.1 libraries are libquickfix.so.3.1.0 and
> libquickfix_jni.so.7.1.0, while the 1.9.0 libraries had newer version
> numbers: libquickfix.so.4.0.0 and libquickfix_jni.so.8.0.0 .
> 
> 
> I'll try to delete the 1.9.0 libraries and recompile.
> 
> OS = RedHat Linux 8.0 (kernel 2.4.20)
> 
> Hans

This happened when running the "test/runat_java 10005" command.

bin/run_banzai resulted in a similar error.

Hans

[Quickfix-developers] QuickFIX 1.9.1 Available

[Oren M. <or...@qu...>]

You know where, but in case you don't:  http://www.quickfixengine.org/

This fixes some problems reported with 1.9.0.  You should get this if 
you are using MySQL, STLPort, or currencies.  This should also fix 
compilation problems people were seeing with SUNPro and some versions 
of Visual Studio.  Documentation is now included in the installation 
docs for SUNPro, and the pretty collaboration diagrams are back.  
Release notes here: 
http://sourceforge.net/project/shownotes.php?release_id=267968

One important change to take note is the XML that comes out of 
Message::toXML is in a slightly different format.  A field used to come 
out like this:

<field name="Text" value="Whatever value that is in this field"/>

Now, it comes out like this:

<field name="Text"><![CDATA[Whatever value that is in this 
field]]></field>

This is so we can handle characters in field values that could 
otherwise make the produces XML invalid.

--oren

[Quickfix-developers] QF raw throughput measurements

[Caleb E. <cal...@gm...>]

I decided to do a small test of optimal QuickFIX throughput, just to
try and gauge the theoretical upper bound that one can expect to get
out of QF as an engine.

My test involved a modified version of the "executor" example C++
application which accepts both market and limit orders, and responds
to each NewOrderSingle with two ExecutionReports (ack and fill), and a
simple sender application which just reads orders from a flat file
(format should be lines like "B 100 XYZ" or "S 1000 IBM 25") and sends
them to the executor.

Both the executor and sender are configured to use the
MemoryStoreFactory for message persistence and I have eliminated all
output other than handling of exceptions and printing of admin msgs.

Here are my results, using an input file of 1972 orders with most of
the Russell 2000 components:

Hardware: 2 blades in an IBM BladeCenter, each w/two Xeon 3.2 GHz CPUs
Operating Sysrtem: RedHat AS3.0_U1
Compiler: GNU g++ 3.3.2 -O2 -g -DHAVE_FTIME
Network: 1 Gbit switched Ethernet (I think)

Sender and Executor on Different Blades:
Sent 1972 messages from 20040914-15:51:31.468 to 20040914-15:51:31.694
= 0.226s (8725.66 msgs/sec)
Rcvd 3944 messages from 20040914-15:51:31.469 to 20040914-15:51:32.575
= 1.106s (3566 msgs/sec)

Sender and Executor on Same Blade:
Sent 1972 messages from 20040914-15:54:14.736 to 20040914-15:54:14.924
= 0.188s (10489.4 msgs/sec)
Rcvd 3944 messages from 20040914-15:54:14.737 to 20040914-15:54:15.709
= 0.972s (4057.61 msgs/sec)

Attached is the source code for the sending application.

-- 
Caleb Epstein
cal...@gm...

Re: [Quickfix-developers] Small issue with UtcTimeStamp and milliseconds

[Oren M. <or...@qu...>]

Perhaps we should copy config.h into the include directory and have 
that file #include <quickfix/config.h>

--oren

On Sep 14, 2004, at 10:34 AM, Caleb Epstein wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: 
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> When compiling an application with QuickFIX on a platform where ftime
> is available, the UtcTimeStamp class will not set the m_ms member
> unless the HAVE_FTIME preprocessor macro is defined.
>
> This macro should be used when compiling QuickFIX (and is taken
> automatically from config.h), but users of the compiled library
> shouldn't need to know about it.  Perhaps UtcTimeStamp::setCurrent
> should be made non-inline?  This could impact performance though.
>
> The simple workaround is just to define HAVE_FTIME when building ones
> own code, but there should probably be a cleaner way to do this.
>
> -- 
> Caleb Epstein
> cal...@gm...
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: thawte's Crypto Challenge Vl
> Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam
> Camcorder. More prizes in the weekly Lunch Hour Challenge.
> Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

[Quickfix-developers] Small issue with UtcTimeStamp and milliseconds

[Caleb E. <cal...@gm...>]

When compiling an application with QuickFIX on a platform where ftime
is available, the UtcTimeStamp class will not set the m_ms member
unless the HAVE_FTIME preprocessor macro is defined.

This macro should be used when compiling QuickFIX (and is taken
automatically from config.h), but users of the compiled library
shouldn't need to know about it.  Perhaps UtcTimeStamp::setCurrent
should be made non-inline?  This could impact performance though.

The simple workaround is just to define HAVE_FTIME when building ones
own code, but there should probably be a cleaner way to do this.

-- 
Caleb Epstein
cal...@gm...

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Joerg T. <Joe...@ma...>]

Oren Miller wrote:
>> On Sep 13, 2004, at 7:56 AM, Joerg Thoennes wrote:
>> 
>>> Caleb Epstein wrote:
>>>
>>>> On Fri, 10 Sep 2004 12:21:22 -0500, Oren Miller 
>>>> <or...@qu...> wrote:
>>>>
>>>>> As for the AR_FLAGS, I would expect that autoconf would be doing that
>>>>> work in the AC_PROG_CXX macro.  What version are you using?  Maybe a
>>>>
>>>> I'm not using autoconf at all, just the supplied configure script that
>>>> comes with QuickFIX.
>>>
>>> I feel there is some ambiguity in the QF setup: On the one site, there 
>>> are autoconf files, but one the other side, pre-build conf files are 
>>> supplied (and even checked-in into the repository).
>>>
>>> IMHO, the most stable solution would be to drop the autoconf-generated 
>>> files altogether and let the users generate them in any case using the 
>>> bootstrap script. They have to do that step anyway if there is any 
>>> problem with the pre-built config files.
>>>
>>> Of course, this adds another step to the installation of QF and 
>>> requires  some assistance in the installation of autoconf tools (e.g. 
>>> on Solaris), but in the end this is the way autoconf is meant to work.

> I don't agree with this.  Generated files are distributed with the 
> releases, but I don't know of any that are checked into the repository. 
>  If you know of some, make a list and we'll have them removed.

Sorry about that. My memory fooled me. Of course, there are no generated 
files in the repository.

> Autotools generate a portable configure script so everyone knows they 
> can do a simple './configure && make' and get a reasonable build.  This 
> is pretty standard and expected behavior for open source distributions. 
>  I don't know what the value of adding additional dependencies would be 
> for people who have no desire to modify the build scripts.

Agreed. I was a bit unsure about *how* portable the configure scripts 
are. Probably the most generic approach is not always the best one.

Cheers, Jörg

-- 
Joerg Thoennes
                                http://macd.com
Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Caleb E. <cal...@gm...>]

On Mon, 13 Sep 2004 14:56:29 +0200, Joerg Thoennes
<joe...@ma...> wrote:

> Of course, this adds another step to the installation of QF and requires
>   some assistance in the installation of autoconf tools (e.g. on
> Solaris), but in the end this is the way autoconf is meant to work.

I'd highly recommend AGAINST doing this.  Its another barrier to entry
for people to get started with QuickFIX and goes against the grain of
just about any other open source project you're liable to run across. 
They all ship a generated "configure".

> Caleb, could you give autoconf a try?

Done.  It didn't help.  Autoconf  2.58 does not detect that "CC" needs
to be used instead of "ar" when generating the libtool script.

I think this stems from the fact that these tools (autoconf etc) are
used largely with C programs and the C++ support is less mature and
more folks are using them on Linux or other systems where gcc is used
than on Solaris with SunPRO.

Anyhow, seeing as one needs to specify CC and CXX, specifying AR isn't
too much to ask.

-- 
Caleb Epstein
cal...@gm...

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Caleb E. <cal...@gm...>]

On Mon, 13 Sep 2004 09:40:34 -0500, Oren Miller <or...@qu...> wrote:
>   I don't know what the value of adding additional dependencies would be
> for people who have no desire to modify the build scripts.

It would be a huge step backward IMHO.  Anything that makes it harder
for people to use the code means there will be fewer of them doing so.

-- 
Caleb Epstein
cal...@gm...

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Oren M. <or...@qu...>]

I don't agree with this.  Generated files are distributed with the=20
releases, but I don't know of any that are checked into the repository.=20=

  If you know of some, make a list and we'll have them removed.

Autotools generate a portable configure script so everyone knows they=20
can do a simple './configure && make' and get a reasonable build.  This=20=

is pretty standard and expected behavior for open source distributions.=20=

  I don't know what the value of adding additional dependencies would be=20=

for people who have no desire to modify the build scripts.

--oren

On Sep 13, 2004, at 7:56 AM, Joerg Thoennes wrote:

> Caleb Epstein wrote:
>> On Fri, 10 Sep 2004 12:21:22 -0500, Oren Miller=20
>> <or...@qu...> wrote:
>>> As for the AR_FLAGS, I would expect that autoconf would be doing =
that
>>> work in the AC_PROG_CXX macro.  What version are you using?  Maybe a
>> I'm not using autoconf at all, just the supplied configure script =
that
>> comes with QuickFIX.
>
> I feel there is some ambiguity in the QF setup: On the one site, there=20=

> are autoconf files, but one the other side, pre-build conf files are=20=

> supplied (and even checked-in into the repository).
>
> IMHO, the most stable solution would be to drop the autoconf-generated=20=

> files altogether and let the users generate them in any case using the=20=

> bootstrap script. They have to do that step anyway if there is any=20
> problem with the pre-built config files.
>
> Of course, this adds another step to the installation of QF and=20
> requires  some assistance in the installation of autoconf tools (e.g.=20=

> on Solaris), but in the end this is the way autoconf is meant to work.
>
> Caleb, could you give autoconf a try?
>
> Cheers, J=F6rg
>
> --=20
> Joerg Thoennes
>                                http://macd.com
> Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
> Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen
>

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Joerg T. <Joe...@ma...>]

Caleb Epstein wrote:
> On Fri, 10 Sep 2004 12:21:22 -0500, Oren Miller <or...@qu...> wrote:
> 
>>As for the AR_FLAGS, I would expect that autoconf would be doing that
>>work in the AC_PROG_CXX macro.  What version are you using?  Maybe a
> 
> I'm not using autoconf at all, just the supplied configure script that
> comes with QuickFIX.

I feel there is some ambiguity in the QF setup: On the one site, there 
are autoconf files, but one the other side, pre-build conf files are 
supplied (and even checked-in into the repository).

IMHO, the most stable solution would be to drop the autoconf-generated 
files altogether and let the users generate them in any case using the 
bootstrap script. They have to do that step anyway if there is any 
problem with the pre-built config files.

Of course, this adds another step to the installation of QF and requires 
  some assistance in the installation of autoconf tools (e.g. on 
Solaris), but in the end this is the way autoconf is meant to work.

Caleb, could you give autoconf a try?

Cheers, Jörg

-- 
Joerg Thoennes
                                http://macd.com
Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen

Re: [Quickfix-developers] Problem with 1.9.0

[Caleb E. <cal...@gm...>]

On Sat, 11 Sep 2004 11:17:04 -0500, Oren Miller <or...@qu...> wrote:

> I think the problem may be that integer_to_string and
> integer_to_string_padded are declared with the static modifier.  Try
> removing that and see if it helps.

Indeed.  They are "static inline", which makes almost no sense at all.

-- 
Caleb Epstein
cal...@gm...

Re: [Quickfix-developers] Build Quickfix with Sun SPRO compiler 6.2 or 8.0

[Caleb E. <cal...@gm...>]

On Fri, 10 Sep 2004 12:21:22 -0500, Oren Miller <or...@qu...> wrote:
> As for the AR_FLAGS, I would expect that autoconf would be doing that
> work in the AC_PROG_CXX macro.  What version are you using?  Maybe a

I'm not using autoconf at all, just the supplied configure script that
comes with QuickFIX.

> newer version will handle this.  Can you put together a little blurb
> about what you did to get these setting in so I can add it to the
> documentation?

Basically with the SunPRO compilers, you can't use "ar" or "ld" to
build static or shared libraries of C++ code, because these don't pull
in all of the template instantiations.  You need to use the C++
compiler (CC) with the flag -xar (static lib) or -G (shared).

The -xs flag allows for debugging without having the .o files still
lying around, and is optional.  If you choose to not use the -g flag,
you shouldn't specify -xs either.

Kristofer's original email to the list that included a patch against
1.7.1 has this all spelled out (the paths to various things are
system-specific; my C++ compiler lives in /opt not /usr/local):

---8<---
BUILD INSTRUCTIONS

- untar quickfix-1.7.1.tar.gz
- convert the eight files above to Unix test format (strip CRs with dos2unix)
- apply the unified patch below (you'll need GNU patch)
- export the following environment variables, editing for compiler and
  java location where appropriate)
      CC=/usr/local/SUNWspro/WS6U2/bin/CC
  CFLAGS=-g -xs
     CXX=/usr/local/SUNWspro/WS6U2/bin/CC
CXXFLAGS=-g -xs
      LD=/usr/local/SUNWspro/WS6U2/bin/CC
 LDFLAGS=-g -xs -lCstd
      AR=/usr/local/SUNWspro/WS6U2/bin/CC
AR_FLAGS=-g -xs -xar -o
JAVA_HOME=/usr/local/java1.4.2
    PATH=$JAVA_HOME:$PATH
- run configure
- make
---8<---

-- 
Caleb Epstein
cal...@gm...

[Quickfix-developers] Errors in XML Spec documents?

[James W. <wi...@wi...>]

 Folks,

   Just to notify you of a few apparent goofs in the FIX4?.xml
files on the website:

1) FIX40.xml, line 88:
<field name="IOIOthSvc" type="CHAR"/>
should be:
<field name="IOIOthSvc" required="N"/>

2) FIX41.xml, line 110:
<field name="IOIOthSvc" type="CHAR"/>
should be:
<field name="IOIOthSvc" required="N"/>

regards,
Jim

Re: [Quickfix-developers] toXML function for 'special XML characters'

[Oren M. <or...@qu...>]

Hmm.  Yeah we should be handling this.  Although I wonder if maybe we 
should just be putting the values into a CDATA section?

--oren

On Sep 10, 2004, at 7:53 AM, Van Gelder Eddy (KATO 2) wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: 
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> Hi Oren,
>
> after receiving a message in the Fixengine I use the toXML function to 
> obtain the XML version of the message and then perform XSLT to 
> transform to our internal format.
>
> for some special XML characters I had to add a bit of code in the 
> Message.cpp to transform
>
> & in &
> < in &lt;
> " in &quot;
>
> otherwise the generated XML is not OK.
>
> <field name="Text" number="58" value="PRICE < 100"/>
>
> should be
>
> <field name="Text" number="58" value="PRICE &lt; 100"/>
>
>
> is this somthing that you would want to put in the base quickfix 
> release ?
>
> Rgds,
> Eddy
>
>
>
> std::string Message::toXMLFields(const FieldMap& fields, int space) 
> const
> { QF_STACK_PUSH(Message::toXMLFields)
>
>   std::stringstream stream;
>   FieldMap::iterator i;
>   std::string name;
>   for(i = fields.begin(); i != fields.end(); ++i)
>   {
>     int field = i->first;
>     std::string value = i->second.getString();
>
> *********************
>     size_t lookHere = 0;
>     size_t foundHere;
>     while ((foundHere = value.find("&",lookHere))
>     != std::string::npos) {
>     value.replace(foundHere, 1, "&amp;");
>      lookHere = foundHere + 5;
>     }
>     lookHere = 0;
>     while ((foundHere = value.find("<",lookHere))
>     != std::string::npos) {
>     value.replace(foundHere, 1, "&lt;");
>     lookHere = foundHere + 4;
>     }
>     lookHere = 0;
>     while ((foundHere = value.find('"',lookHere))
>     != std::string::npos) {
>     value.replace(foundHere, 1, "&quot;");
>     lookHere = foundHere + 6;
>     }
> ******************
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

Re: [Quickfix-developers] Problem with 1.9.0

[Oren M. <or...@qu...>]

I think the problem may be that integer_to_string and 
integer_to_string_padded are declared with the static modifier.  Try 
removing that and see if it helps.

--oren

On Sep 10, 2004, at 6:00 AM, Kiran Khatra wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: 
> http://www.quickfixengine.org/wikifix/index.php?QuickFixFAQ
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
>
> Hi
> I'm trying to build quickfix 1.9.0 using quickfix.sln under visual
> studio.net and am getting the following error during compilation.
>
> quickfix_net error LNK2005: "char * __cdecl
> FIX::integer_to_string_padded<int>(char *,unsigned int,int,unsigned
> int,char)" (??$integer_to_string_padded@H@FIX@@$$FYAPADPADIHID@Z)
> already defined in Group.obj
> quickfix_net error LNK2005: "char * __cdecl
> FIX::integer_to_string_padded<int>(char *,unsigned int,int,unsigned
> int,char)" (??$integer_to_string_padded@H@FIX@@$$FYAPADPADIHID@Z)
> already defined in Group.obj
> quickfix_net error LNK2005: "char * __cdecl
> FIX::integer_to_string<int>(char *,unsigned int,int)"
> (??$integer_to_string@H@FIX@@$$FYAPADPADIH@Z) already defined in
> Group.obj
> quickfix_net error LNK2005: "char * __cdecl
> FIX::integer_to_string<int>(char *,unsigned int,int)"
> (??$integer_to_string@H@FIX@@$$FYAPADPADIH@Z) already defined in
> Group.obj
> quickfix_net error LNK2005: "char * __cdecl
> FIX::integer_to_string<long>(char *,unsigned int,long)"
> (??$integer_to_string@J@FIX@@$$FYAPADPADIJ@Z) already defined in
> Group.obj
> quickfix_net fatal error LNK1169: one or more multiply defined symbols
> found
> Metadata file 'C:\QuickFixSln\quickfix\lib\quickfix_net.dll' could not
> be found
>
> Has anyone seen this or does anyone know how to resolve this?
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>