quickfix-developers — QuickFIX developers forum

Re: [Quickfix-developers] fragmentation?

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

Hi Alvin,

> How does quickfix handle fragmentation? Do I have to do it myself? Or 
> quickfix will decide and handle for me? Thanks!

What do you mean? If a socket read returns an incomplete FIX message, and the next socket 
read returns the next part and whether QF handles this?

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] fragmentation?

[Alvin W. <AW...@FF...>]

Hi,

How does quickfix handle fragmentation? Do I have to do it myself? Or 
quickfix will decide and handle for me? Thanks!

Alvin
**********************************************************************
This e-mail message is intended solely for the use of the addressee. The message
 may contain information that is privileged and confidential.  Disclosure to
anyone other than the intended recipient is prohibited. If you are not the
intended recipient, please do not disseminate, distribute or copy this
communication, by e-mail or otherwise. Instead, please notify us immediately by
return e-mail (including the original message with your reply) and then delete
and discard all copies of the message.
We have taken precautions to minimize the risk of transmitting software viruses
 but nevertheless advise you to carry out your own virus checks on any
attachment to  this  message.  We  accept  no  liability  for any loss or damage
 caused by software viruses.
**********************************************************************

Re: [Quickfix-developers] Quickfix as Eclipse plugin

[Marco M. <ma...@ma...>]

Joerg,

yes the complete version with the RuntimeConfig works fine under=20
Windows, I haven't had the time to recompile and test under Linux.


On 14/02/2005 17.19 Joerg Thoennes wrote:

> Scott Harrington wrote:
>> Oops, I forgot the most important file.  I've attached it here, and al=
so=20
>> uploaded a fixed ZIP file to the bugtracker.
>=20
> Thanks, Scott for this patch. I will have a look at it and check it in =
as soon as I got=20
> some time. Marco, could you please confirm that the now complete versio=
n (including=20
> RuntimeConfig) works fine? Just update the bugtracker.
>=20
> Thanks, J=F6rg
>=20

--=20
Marco Maccaferri
Consulting, Design and Software Development
http://www.maccasoft.com

Re: [Quickfix-developers] Quickfix as Eclipse plugin

[Clive M. <cl...@va...>]

On Monday 14 Feb 2005 16:19, Joerg Thoennes wrote:

> Scott Harrington wrote:
> > Oops, I forgot the most important file.  I've attached it here, and also
> > uploaded a fixed ZIP file to the bugtracker.
>
> Thanks, Scott for this patch. I will have a look at it and check it in as
> soon as I got some time. Marco, could you please confirm that the now
> complete version (including RuntimeConfig) works fine? Just update the
> bugtracker.

Yes, good patch Scott. I've been testing with a JBoss MBean deployment, minus 
the kludge of a ClassLoader hack I was using.

Regards

Clive
-- 
Clive Messer <cl...@va...>

Re: [Quickfix-developers] Quickfix as Eclipse plugin

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

Scott Harrington wrote:
> Oops, I forgot the most important file.  I've attached it here, and also 
> uploaded a fixed ZIP file to the bugtracker.

Thanks, Scott for this patch. I will have a look at it and check it in as soon as I got 
some time. Marco, could you please confirm that the now complete version (including 
RuntimeConfig) works fine? Just update the bugtracker.

Thanks, 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] Quickfix as Eclipse plugin

[Scott H. <sco...@fo...>]

Oops, I forgot the most important file.  I've attached it here, and also 
uploaded a fixed ZIP file to the bugtracker.

On Mon, 14 Feb 2005, Marco Maccaferri wrote:

> Scott,
> 
> your patch seems to work, I'll do more tests as the plugin development 
> proceed.
> 
> One thing: the RuntimeConfig.java file seems to be missing from the 
> archive attached to the bug report.
> 
> Many thanks for your help.
> 
> 
> On 13/02/2005 19.05 Scott Harrington wrote:
> 
> > Marco's problem sounds a lot like what happened to me when I needed to run
> > under WebStart, which employs a different ClassLoader that JNI's usual
> > FindClass method is not aware of.  Coupled with QF's current exception
> > handling under JNI, the resulting failures were difficult to debug even
> > with CallStack support compiled in.
> > 
> > I have just now added Bug #58 to the bugtracker, and uploaded a ZIP file
> > attachment with my fix.  These fixes improve JNI support generally, not
> > just for WebStart or other weird ClassLoader situtations.  Here's an
> > overview of what I did:
> > 
> > - new quickfix.RuntimeConfig java class
> > - modified src/java/JVM.cpp to optionally use alternate ClassLoader
> > - modified src/C++/CallStack.* and Exception.h to be usable from src/java
> > - added fatalError() function which shuts down the JVM cleanly on coredump
> > - added CallStack wrappers to JNI code
> > 
> > I'll be happy to discuss these changes and would love to see them merged
> > into CVS.
> > 
> > Scott
> > 
> > P.S. for some reason I cannot read from CVS today.  Is it not still
> > :pserver:ano...@cv...:/cvsroot/quickfix ?
> > 
> > 
> > On Sun, 13 Feb 2005, Marco Maccaferri wrote:
> > 
> >> Hi.
> >> 
> >> No, it pops-up a dialog 'Microsoft Visual C++ Runtime Error' with 
> >> '<path_to>\javaw.exe This application has requested the Runtime to 
> >> terminate it in an unusual way. Please contact...'. I can't find a more 
> >> detailed log.
> >> 
> >> The jni library seems to load correctly. If the client doesn't connect 
> >> to the server it keeps retrying the connection without problems, 
> >> whenever the client connects the error occurs.
> >> 
> >> On 13/02/2005 15.39 Steve Bate wrote:
> >> 
> >> > Hi Marco,
> >> > 
> >> > Can you be more specific about the crash? Is it complaining
> >> > about not finding the native library or ???.
> >> > 
> >> > Steve
> >> > 
> >> >> -----Original Message-----
> >> >> From: Marco Maccaferri [mailto:ma...@ma...]
> >> >> Sent: Sunday, February 13, 2005 5:28 AM
> >> >> To: qui...@li...
> >> >> Subject: [Quickfix-developers] Quickfix as Eclipse plugin
> >> >> 
> >> >> Hello All.
> >> >> 
> >> >> I'm trying to integrate Quickfix in an Eclipse plugin but without any 
> >> >> luck so far, it keeps crashing whenever a connection is established. The 
> >> >> JNI library seems to be loaded without problems, my guess is that it has 
> >> >> something to do with the Eclipse own class loader.
> >> >> Do someone have any hints on how to get Quickfix work in an 
> >> >> Eclipse plugin ?
> >> >> 
> >> >> -- 
> >> >> Marco Maccaferri
> >> >> Consulting, Design and Software Development
> >> >> http://www.maccasoft.com

Re: [Quickfix-developers] Quickfix as Eclipse plugin

[Marco M. <ma...@ma...>]

Scott,

your patch seems to work, I'll do more tests as the plugin development 
proceed.

One thing: the RuntimeConfig.java file seems to be missing from the 
archive attached to the bug report.

Many thanks for your help.


On 13/02/2005 19.05 Scott Harrington wrote:

> Marco's problem sounds a lot like what happened to me when I needed to run
> under WebStart, which employs a different ClassLoader that JNI's usual
> FindClass method is not aware of.  Coupled with QF's current exception
> handling under JNI, the resulting failures were difficult to debug even
> with CallStack support compiled in.
> 
> I have just now added Bug #58 to the bugtracker, and uploaded a ZIP file
> attachment with my fix.  These fixes improve JNI support generally, not
> just for WebStart or other weird ClassLoader situtations.  Here's an
> overview of what I did:
> 
> - new quickfix.RuntimeConfig java class
> - modified src/java/JVM.cpp to optionally use alternate ClassLoader
> - modified src/C++/CallStack.* and Exception.h to be usable from src/java
> - added fatalError() function which shuts down the JVM cleanly on coredump
> - added CallStack wrappers to JNI code
> 
> I'll be happy to discuss these changes and would love to see them merged
> into CVS.
> 
> Scott
> 
> P.S. for some reason I cannot read from CVS today.  Is it not still
> :pserver:ano...@cv...:/cvsroot/quickfix ?
> 
> 
> On Sun, 13 Feb 2005, Marco Maccaferri wrote:
> 
>> Hi.
>> 
>> No, it pops-up a dialog 'Microsoft Visual C++ Runtime Error' with 
>> '<path_to>\javaw.exe This application has requested the Runtime to 
>> terminate it in an unusual way. Please contact...'. I can't find a more 
>> detailed log.
>> 
>> The jni library seems to load correctly. If the client doesn't connect 
>> to the server it keeps retrying the connection without problems, 
>> whenever the client connects the error occurs.
>> 
>> On 13/02/2005 15.39 Steve Bate wrote:
>> 
>> > Hi Marco,
>> > 
>> > Can you be more specific about the crash? Is it complaining
>> > about not finding the native library or ???.
>> > 
>> > Steve
>> > 
>> >> -----Original Message-----
>> >> From: Marco Maccaferri [mailto:ma...@ma...]
>> >> Sent: Sunday, February 13, 2005 5:28 AM
>> >> To: qui...@li...
>> >> Subject: [Quickfix-developers] Quickfix as Eclipse plugin
>> >> 
>> >> Hello All.
>> >> 
>> >> I'm trying to integrate Quickfix in an Eclipse plugin but without any 
>> >> luck so far, it keeps crashing whenever a connection is established. The 
>> >> JNI library seems to be loaded without problems, my guess is that it has 
>> >> something to do with the Eclipse own class loader.
>> >> Do someone have any hints on how to get Quickfix work in an 
>> >> Eclipse plugin ?
>> >> 
>> >> -- 
>> >> Marco Maccaferri
>> >> Consulting, Design and Software Development
>> >> http://www.maccasoft.com
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
> 

-- 
Marco Maccaferri
Consulting, Design and Software Development
http://www.maccasoft.com

[Quickfix-developers] QuickFIX 1.9.4 RPMS for Fedora Core 3

[Clive M. <cl...@va...>]

I've built QuickFIX 1.9.4 RPMS for Fedora Core 3, which include support for 
MySQL, Python and Java. May be of use to someone else. Build notes and links 
to download .....

http://www.objectsoft-systems.ltd.uk/quickfix/

Regards

Clive
-- 
Clive Messer <cl...@va...>

Re: [Quickfix-developers] Quickfix as Eclipse plugin

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

Hi Marco,

> No, it pops-up a dialog 'Microsoft Visual C++ Runtime Error' with 
> '<path_to>\javaw.exe This application has requested the Runtime to 
> terminate it in an unusual way. Please contact...'. I can't find a more 
> detailed log.
> 
> The jni library seems to load correctly. If the client doesn't connect 
> to the server it keeps retrying the connection without problems, 
> whenever the client connects the error occurs.

Please tell us which QF version you are using. Some versions had the problem that if the 
configuration file was incomplete, an uncatched exception could pop up the stack and lead 
to a crash. Esp. check your host and port settings, e.g. whether port is already used on 
this machine.

In addition, the crash log of the JVM would be helpful.

Thanks, 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] Quickfix as Eclipse plugin

[Scott H. <sco...@fo...>]

Marco's problem sounds a lot like what happened to me when I needed to run
under WebStart, which employs a different ClassLoader that JNI's usual
FindClass method is not aware of.  Coupled with QF's current exception
handling under JNI, the resulting failures were difficult to debug even
with CallStack support compiled in.

I have just now added Bug #58 to the bugtracker, and uploaded a ZIP file
attachment with my fix.  These fixes improve JNI support generally, not
just for WebStart or other weird ClassLoader situtations.  Here's an
overview of what I did:

- new quickfix.RuntimeConfig java class
- modified src/java/JVM.cpp to optionally use alternate ClassLoader
- modified src/C++/CallStack.* and Exception.h to be usable from src/java
- added fatalError() function which shuts down the JVM cleanly on coredump
- added CallStack wrappers to JNI code

I'll be happy to discuss these changes and would love to see them merged
into CVS.

Scott

P.S. for some reason I cannot read from CVS today.  Is it not still
:pserver:ano...@cv...:/cvsroot/quickfix ?


On Sun, 13 Feb 2005, Marco Maccaferri wrote:

> Hi.
> 
> No, it pops-up a dialog 'Microsoft Visual C++ Runtime Error' with 
> '<path_to>\javaw.exe This application has requested the Runtime to 
> terminate it in an unusual way. Please contact...'. I can't find a more 
> detailed log.
> 
> The jni library seems to load correctly. If the client doesn't connect 
> to the server it keeps retrying the connection without problems, 
> whenever the client connects the error occurs.
> 
> On 13/02/2005 15.39 Steve Bate wrote:
> 
> > Hi Marco,
> > 
> > Can you be more specific about the crash? Is it complaining
> > about not finding the native library or ???.
> > 
> > Steve
> > 
> >> -----Original Message-----
> >> From: Marco Maccaferri [mailto:ma...@ma...]
> >> Sent: Sunday, February 13, 2005 5:28 AM
> >> To: qui...@li...
> >> Subject: [Quickfix-developers] Quickfix as Eclipse plugin
> >> 
> >> Hello All.
> >> 
> >> I'm trying to integrate Quickfix in an Eclipse plugin but without any 
> >> luck so far, it keeps crashing whenever a connection is established. The 
> >> JNI library seems to be loaded without problems, my guess is that it has 
> >> something to do with the Eclipse own class loader.
> >> Do someone have any hints on how to get Quickfix work in an 
> >> Eclipse plugin ?
> >> 
> >> -- 
> >> Marco Maccaferri
> >> Consulting, Design and Software Development
> >> http://www.maccasoft.com

[Quickfix-developers] FIX engine performance benchmarks?

[Steve B. <st...@xp...>]

Does anybody know of published FIX engine performance benchmarks 
(not the results, but the actual benchmark definitions)?

I've seen performance numbers from various FIX engine vendors
but they don't give much details about the actual tests,
the hardware used (processor type and speed, memory, ...)
or if they were logging messages to disk versus memory, etc.

Steve

Re: [Quickfix-developers] Quickfix as Eclipse plugin

[Marco M. <ma...@ma...>]

Hi.

No, it pops-up a dialog 'Microsoft Visual C++ Runtime Error' with 
'<path_to>\javaw.exe This application has requested the Runtime to 
terminate it in an unusual way. Please contact...'. I can't find a more 
detailed log.

The jni library seems to load correctly. If the client doesn't connect 
to the server it keeps retrying the connection without problems, 
whenever the client connects the error occurs.

On 13/02/2005 15.39 Steve Bate 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 Marco,
> 
> Can you be more specific about the crash? Is it complaining
> about not finding the native library or ???.
> 
> Steve
> 
>> -----Original Message-----
>> From: Marco Maccaferri [mailto:ma...@ma...]
>> Sent: Sunday, February 13, 2005 5:28 AM
>> To: qui...@li...
>> Subject: [Quickfix-developers] Quickfix as Eclipse plugin
>> 
>> 
>> 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
>> 
>> Hello All.
>> 
>> I'm trying to integrate Quickfix in an Eclipse plugin but without any 
>> luck so far, it keeps crashing whenever a connection is established. The 
>> JNI library seems to be loaded without problems, my guess is that it has 
>> something to do with the Eclipse own class loader.
>> Do someone have any hints on how to get Quickfix work in an 
>> Eclipse plugin ?
>> 
>> -- 
>> Marco Maccaferri
>> Consulting, Design and Software Development
>> http://www.maccasoft.com
>> 
>> 
>> 
>> -------------------------------------------------------
>> SF email is sponsored by - The IT Product Guide
>> Read honest & candid reviews on hundreds of IT Products from real users.
>> Discover which products truly live up to the hype. Start reading now.
>> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
>> _______________________________________________
>> Quickfix-developers mailing list
>> Qui...@li...
>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>> 
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
> 

-- 
Marco Maccaferri
Consulting, Design and Software Development
http://www.maccasoft.com

RE: [Quickfix-developers] Quickfix as Eclipse plugin

[Steve B. <st...@xp...>]

Hi Marco,

Can you be more specific about the crash? Is it complaining
about not finding the native library or ???.

Steve

> -----Original Message-----
> From: Marco Maccaferri [mailto:ma...@ma...]
> Sent: Sunday, February 13, 2005 5:28 AM
> To: qui...@li...
> Subject: [Quickfix-developers] Quickfix as Eclipse plugin
> 
> 
> 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
> 
> Hello All.
> 
> I'm trying to integrate Quickfix in an Eclipse plugin but without any 
> luck so far, it keeps crashing whenever a connection is established. The 
> JNI library seems to be loaded without problems, my guess is that it has 
> something to do with the Eclipse own class loader.
> Do someone have any hints on how to get Quickfix work in an 
> Eclipse plugin ?
> 
> -- 
> Marco Maccaferri
> Consulting, Design and Software Development
> http://www.maccasoft.com
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
> 

[Quickfix-developers] Quickfix as Eclipse plugin

[Marco M. <ma...@ma...>]

Hello All.

I'm trying to integrate Quickfix in an Eclipse plugin but without any 
luck so far, it keeps crashing whenever a connection is established. The 
JNI library seems to be loaded without problems, my guess is that it has 
something to do with the Eclipse own class loader.
Do someone have any hints on how to get Quickfix work in an Eclipse plugin ?

-- 
Marco Maccaferri
Consulting, Design and Software Development
http://www.maccasoft.com

RE: [Quickfix-developers] unhandled exception

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

<div>QF_STACK_CATCH shouldn´t prevent the application from crashing.&nbsp;
In fact it should crash on an unhandled exception so you can get the
stack trace.&nbsp; The solution is to make sure that the exception is
handled appropriately in the quickfix library.&nbsp; I believe this
particular problem has been fixed, but we will research it further.</div>
<div>&nbsp;</div>
<div>--oren</div>
<BLOCKQUOTE style="PADDING-LEFT: 8px; MARGIN-LEFT: 8px; BORDER-LEFT:
blue 2px solid"><BR>-------- Original Message --------<BR>Subject:
[Quickfix-developers] unhandled exception<BR>From: "Sean Kirkpatrick"
&lt;Sea...@Pi...&gt;<BR>Date: Thu, February 10,
2005 12:54 pm<BR>To: qui...@li...<BR><BR>
<META content="MS Exchange Server version 6.5.7226.0"
name=Generator><!-- Converted from text/rtf format -->
<div><FONT face=Arial size=2>Hello everyone,</FONT> </div>
<div><FONT face=Arial size=2>We experienced a crash within quickfix this
morning due to an unhandled std::exception.</FONT> <BR><FONT face=Arial
size=2>We're currently running 1.7.0 in our production system and will
be upgrading to 1.9.4 soon,</FONT> <BR><FONT face=Arial size=2>but
after going through the exception handling, it seems like the potential
for a crash is</FONT> <BR><FONT face=Arial size=2>still there.&nbsp; The
code that appears to be problematic is in CallStack.h:</FONT> </div>
<div><FONT face=Arial size=2>#define QF_STACK_CATCH \</FONT> <BR><FONT
face=Arial size=2>&nbsp; } catch( ... ) { FIX::CallStack::output();
throw; }</FONT> </div>
<div><FONT face=Arial size=2>The problem we had in 1.7.0 occurred when a
reject was attempted in response to a logon</FONT> <BR><FONT face=Arial
size=2>message after finding a tag out of order.&nbsp; The
std::exception that was thrown bubbled all the</FONT> <BR><FONT
face=Arial size=2>way up to Acceptor::startThread, where exceptions are
handled by QF_STACK_POP and</FONT> <BR><FONT face=Arial
size=2>QF_STACK_CATCH.&nbsp; I would have thought that because this is
the entry point for the</FONT> <BR><FONT face=Arial size=2>acceptor
thread, the buck should stop there.&nbsp; As you can see from the
definition of</FONT> <BR><FONT face=Arial size=2>QF_STACK_CATCH, the
exception was caught and re-thrown, resulting in the crash.&nbsp;
The</FONT> <BR><FONT face=Arial size=2>code for QF_STACK_CATCH the same
in 1.9.4, but if a reject is attempted when a session</FONT> <BR><FONT
face=Arial size=2>is not logged in, a std::logic_error is thrown
instead of a std::exception.</FONT> </div>
<div><FONT face=Arial size=2>Should the throw be removed from the
definition of QF_STACK_CATCH?</FONT> </div>
<div><FONT face=Arial size=2>Regards,</FONT> </div>
<div><FONT face=Arial size=2>Sean Kirkpatrick</FONT> </div></BLOCKQUOTE>

[Quickfix-developers] unhandled exception

[Sean K. <Sea...@Pi...>]

Hello everyone,

We experienced a crash within quickfix this morning due to an unhandled =
std::exception.
We're currently running 1.7.0 in our production system and will be =
upgrading to 1.9.4 soon,
but after going through the exception handling, it seems like the =
potential for a crash is
still there.  The code that appears to be problematic is in CallStack.h:

#define QF_STACK_CATCH \
  } catch( ... ) { FIX::CallStack::output(); throw; }

The problem we had in 1.7.0 occurred when a reject was attempted in =
response to a logon
message after finding a tag out of order.  The std::exception that was =
thrown bubbled all the
way up to Acceptor::startThread, where exceptions are handled by =
QF_STACK_POP and
QF_STACK_CATCH.  I would have thought that because this is the entry =
point for the
acceptor thread, the buck should stop there.  As you can see from the =
definition of
QF_STACK_CATCH, the exception was caught and re-thrown, resulting in the =
crash.  The
code for QF_STACK_CATCH the same in 1.9.4, but if a reject is attempted =
when a session
is not logged in, a std::logic_error is thrown instead of a =
std::exception.

Should the throw be removed from the definition of QF_STACK_CATCH?

Regards,

Sean Kirkpatrick

Re: [Quickfix-developers] TestRequest messages

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

Jonathan Duke wrote:

> Hmm, you're right. The default post-build step in VC++.NET is failing to
> copy TestRequest.h, QuoteStatusReport.h and QuoteStatusRequest.h to the
> include/quickfix/fix* directories I was looking at. The post-build step
> is:
> 
> echo test > EXCLUDE
> echo stdafx.h >> EXCLUDE
> xcopy src\C++\*.h /S/Y /EXCLUDE:EXCLUDE include\quickfix del /F EXCLUDE
> 
> I don't know exactly how xcopy works but I am guessing the "test >
> EXCLUDE" bit excludes from the copy all files with "test" in their name
> so as not to copy the src/C++/test/ files, but also covers the 3 files
> missing files.
> 
> I'll move the test directory somewhere else and remove that line from
> the post-build step!

Looks like a Deja vu to me. Found this in the mailing list:

http://sourceforge.net/mailarchive/message.php?msg_id=6286023

Please check whether "echo test\ > EXCLUDE" works and respond to bug tracker:

http://www.quickfixengine.org/bugtracker/bug.php?op=show&bugid=57

Thanks, 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] TestRequest messages

[Jonathan D. <j....@wi...>]

> -----Original Message-----
> From: Joerg Thoennes [mailto:Joe...@ma...]=20
> Sent: 10 February 2005 17:21
> To: Jonathan Duke
> Cc: qui...@li...
> Subject: Re: [Quickfix-developers] TestRequest messages
>=20
> Hi Duke,
>=20
> > I was wondering why there is no TestRequest class when there are=20
> > classes for just about every other type of message (even=20
> Heartbeat, Logon etc.)?
> > Not that it's hard to write one or to build up the message=20
> manually; I=20
> > was just wondering if there is a reason not to have a class=20
> as standard.
>=20
> Which part of the QF API you are talking about? For the C++=20
> part I found
>=20
> src/C++/fix4*/TestRequest.h
>=20
> Dito
>=20
> src/quickfix/fix4*/TestRequest.java
>=20
> What are you looking for?

Hmm, you're right. The default post-build step in VC++.NET is failing to
copy TestRequest.h, QuoteStatusReport.h and QuoteStatusRequest.h to the
include/quickfix/fix* directories I was looking at. The post-build step
is:

echo test > EXCLUDE
echo stdafx.h >> EXCLUDE
xcopy src\C++\*.h /S/Y /EXCLUDE:EXCLUDE include\quickfix del /F EXCLUDE

I don't know exactly how xcopy works but I am guessing the "test >
EXCLUDE" bit excludes from the copy all files with "test" in their name
so as not to copy the src/C++/test/ files, but also covers the 3 files
missing files.

I'll move the test directory somewhere else and remove that line from
the post-build step!

Thanks!

Jonathan Duke

Re: [Quickfix-developers] TestRequest messages

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

Hi Duke,

> I was wondering why there is no TestRequest class when there are classes
> for just about every other type of message (even Heartbeat, Logon etc.)?
> Not that it's hard to write one or to build up the message manually; I
> was just wondering if there is a reason not to have a class as standard.

Which part of the QF API you are talking about? For the C++ part I found

src/C++/fix4*/TestRequest.h

Dito

src/quickfix/fix4*/TestRequest.java

What are you looking for?

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] tag 11 format restrictions and new clordid in cxl and repl msg

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

Narayan, Arvind wrote:
> TransactTime I believe is type UTCTimeStamp, which upon instantiation, populates itself
> with the current time. Ie if an object has field transacttime, that value does not need
> to be set explicitly by the application. It is set when the object is created.

True, it did not check that case.

> I was not clear about the difference between setting fields and setting values to
> fields.

In our system, the TransactTime is pulled out of the propriatary message send by the 
exchange, so just setting the current time would be wrong.

>> Yes, at the moment this is the case. But it could be a good idea to have some more
>> support to ease application level programming.
> 
> What did you have in mind? This would indeed be powerful functionality if done
> correctly.

Nothing too specific yet. But for our own FIX application level interface I found that you 
need a couple of utility functions to ease programming. I wonder whether we could collect 
some application level utilities to provide a library. The additional benefit would be 
that this library could do things in a FIX compliant way (think of ClOrdID / OrigClOrdID 
chaining...) If you did some FIX programming, you probably have some functions in mind.

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] TestRequest messages

[Jonathan D. <j....@wi...>]

Hi,

I was wondering why there is no TestRequest class when there are classes
for just about every other type of message (even Heartbeat, Logon etc.)?
Not that it's hard to write one or to build up the message manually; I
was just wondering if there is a reason not to have a class as standard.

Thanks!

--=20
Jonathan Duke

RE: [Quickfix-developers] tag 11 format restrictions and new clordid in cxl and repl msg

[Narayan, A. <Arv...@le...>]

Thanks.

TransactTime I believe is type UTCTimeStamp, which upon instantiation, =
populates itself with the current time. Ie if an object has field =
transacttime, that value does not need to be set explicitly by the =
application. It is set when the object is created.

I was not clear about the difference between setting fields and setting =
values to fields.

-> Yes, at the moment this is the case. But it could be a good=20
-> idea to have some more support=20
-> to ease application level programming.

What did you have in mind=3F This would indeed be powerful functionality if=
 =
done correctly.

-> -----Original Message-----
-> From: Joerg Thoennes [mailto:Joe...@ma...]=20
-> Sent: Thursday, February 10, 2005 11:53 AM
-> To: Narayan, Arvind
-> Cc: Asim; qui...@li...
-> Subject: Re: [Quickfix-developers] tag 11 format=20
-> restrictions and new clordid in cxl and repl msg
->=20
->=20
-> Narayan, Arvind wrote:
-> > AFAIK, you have to populate all field data yourself. The only=20
-> > exceptions that I know of are:
-> > =20
-> > MsgSeqNum
-> > SendingTime
-> > TransactTime
->=20
-> Sorry to correct you, but TransactTime is part of the body=20
-> of some application level=20
-> messages, so you should set it yourself.
->=20
-> > SenderCompID
-> > TargetCompID
->=20
-> QuickFIX handles all required header and trailer fields,=20
-> which also include
->=20
-> MsgType, MsgSeqNo, (PossDupFlag, OrigSendingTime for resent=20
-> messages) and CheckSum for the trailer
->=20
-> You may add further header fields in the toApp() callback=20
-> for application level messages.
->=20
-> In addition, QuickFIX handles all 7 administrative messages for you:
->=20
-> 	Heartbeat, TestRequest, ResendRequest, Reject,=20
-> SequenceReset, Logon, Logout
->=20
-> > The sequence number is managed by the engine.
-> > SendingTime and TransactTime are set on creation
->=20
-> Only SendingTime is inserted just before QuickFIX puts the=20
-> message on the wire.
->=20
-> > SenderCompID and TargetCompID are in the session file
-> > =20
-> > Some other fields can have default values. (See the configuration=20
-> > section of the QuickFix manual)
-> > =20
-> > all orderid type stuff is managed by the application.
->=20
-> Yes, at the moment this is the case. But it could be a good=20
-> idea to have some more support=20
-> to ease application level programming.
->=20
-> > Ideally, you provide a FIX interface to your clients, and=20
-> substitute=20
-> > fields where necessary so the exchange is happy.
->=20
-> Cheers, J=F6rg
->=20
-> > 	-----Original Message-----
-> > 	From: qui...@li...
-> > [mailto:qui...@li...]=20
-> On Behalf Of=20
-> > Asim
-> > 	Sent: Thursday, February 10, 2005 9:27 AM
-> > 	To: qui...@li...
-> > 	Subject: [Quickfix-developers] tag 11 format=20
-> restrictions and new=20
-> > clordid in cxl and repl msg
-> > =09
-> > =09
-> > 	Hi all - my first post here, Just started looking into q.fix.
-> > 	=20
-> > 	While using quickfix API, all the orderids set has to=20
-> be maintained=20
-> > by the application=3F Or the api can do that as well =3F ya=20
-> know..clorder,=20
-> > orderid, origclorderid...in cxls and cxl replace messages,
-> > 	Also some firms have diff requirement in cxl messages....like
-> > they never want a new clorderid in cxl message and no new=20
-> clorderid in
-> > cxl/replace msg allowed while reducing quantity, Looks=20
-> like violating
-> > fix rules=3F Another prob is some firms restrict the client sides to
-> > format their clorder like yyyymmdd-999, whereas others=20
-> leave it on order
-> > sending firm's discretion.
-> > 	So I was wondering if quickfix has the capability to cope with
-> > these variant requirements=3F
-> > 	=20
-> > 	Please help!
-> > 	=20
-> > 	Thanks,
-> > 	Asim
-> > 	=20
-> >=20
-> >=20
-> -------------------------------------------------------------
-> ---------
-> > --------
-> > This message is intended only for the personal and=20
-> confidential use of the designated recipient(s) named above.=20
->  If you are not the intended recipient of this message you=20
-> are hereby notified that any review, dissemination,=20
-> distribution or copying of this message is strictly=20
-> prohibited.  This communication is for information purposes=20
-> only and should not be regarded as an offer to sell or as a=20
-> solicitation of an offer to buy any financial product, an=20
-> official confirmation of any transaction, or as an official=20
-> statement of Lehman Brothers.  Email transmission cannot be=20
-> guaranteed to be secure or error-free.  Therefore, we do not=20
-> represent that this information is complete or accurate and=20
-> it should not be relied upon as such.  All information is=20
-> subject to change without notice.
-> >=20
->=20
->=20
-> --=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
->=20
->=20
->=20

---------------------------------------------------------------------------=
---
This message is intended only for the personal and confidential use of the =
designated recipient(s) named above.  If you are not the intended recipient=
 =
of this message you are hereby notified that any review, dissemination, =
distribution or copying of this message is strictly prohibited.  This =
communication is for information purposes only and should not be regarded a=
s=
 an offer to sell or as a solicitation of an offer to buy any financial =
product, an official confirmation of any transaction, or as an official =
statement of Lehman Brothers.  Email transmission cannot be guaranteed to b=
e=
 secure or error-free.  Therefore, we do not represent that this informatio=
n=
 is complete or accurate and it should not be relied upon as such.  All =
information is subject to change without notice.

Re: [Quickfix-developers] tag 11 format restrictions and new clordid in cxl and repl msg

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

Asim wrote:
> [...]
> Also some firms have diff requirement in cxl messages..like they never
> want a new clorderid in cxl message and no new clorderid in cxl/replace
> msg allowed while reducing quantity, Looks like violating fix rules?

Yes, in this case they violate the rules. But there are many FIX implementations which 
violate the spec in one or another way. QuickFIX is fully compliant at the session level 
(except for bugs of course), but on the application level you are at your own.

> Another prob is some firms restrict the client sides to format their
> clorder like yyyymmdd-999, whereas others leave it on order sending
> firm's discretion.
> So I was wondering if quickfix has the capability to cope with these
> variant requirements?

Perhaps some extra functions to ease application level development would make sense, but 
not as a part of the core QF API.

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] tag 11 format restrictions and new clordid in cxl and repl msg

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

Narayan, Arvind wrote:
> AFAIK, you have to populate all field data yourself. The only exceptions
> that I know of are:
>  
> MsgSeqNum
> SendingTime
> TransactTime 

Sorry to correct you, but TransactTime is part of the body of some application level 
messages, so you should set it yourself.

> SenderCompID
> TargetCompID

QuickFIX handles all required header and trailer fields, which also include

MsgType, MsgSeqNo, (PossDupFlag, OrigSendingTime for resent messages) and
CheckSum for the trailer

You may add further header fields in the toApp() callback for application level messages.

In addition, QuickFIX handles all 7 administrative messages for you:

	Heartbeat, TestRequest, ResendRequest, Reject, SequenceReset, Logon, Logout

> The sequence number is managed by the engine.
> SendingTime and TransactTime are set on creation

Only SendingTime is inserted just before QuickFIX puts the message on the wire.

> SenderCompID and TargetCompID are in the session file
>  
> Some other fields can have default values. (See the configuration
> section of the QuickFix manual)
>  
> all orderid type stuff is managed by the application. 

Yes, at the moment this is the case. But it could be a good idea to have some more support 
to ease application level programming.

> Ideally, you provide a FIX interface to your clients, and substitute
> fields where necessary so the exchange is happy.

Cheers, Jörg

> 	-----Original Message-----
> 	From: qui...@li...
> [mailto:qui...@li...] On Behalf Of
> Asim
> 	Sent: Thursday, February 10, 2005 9:27 AM
> 	To: qui...@li...
> 	Subject: [Quickfix-developers] tag 11 format restrictions and
> new clordid in cxl and repl msg
> 	
> 	
> 	Hi all - my first post here, Just started looking into q.fix.
> 	 
> 	While using quickfix API, all the orderids set has to be
> maintained by the application? Or the api can do that as well ? ya
> know..clorder, orderid, origclorderid...in cxls and cxl replace
> messages,
> 	Also some firms have diff requirement in cxl messages....like
> they never want a new clorderid in cxl message and no new clorderid in
> cxl/replace msg allowed while reducing quantity, Looks like violating
> fix rules? Another prob is some firms restrict the client sides to
> format their clorder like yyyymmdd-999, whereas others leave it on order
> sending firm's discretion.
> 	So I was wondering if quickfix has the capability to cope with
> these variant requirements?
> 	 
> 	Please help!
> 	 
> 	Thanks,
> 	Asim
> 	 
> 
> ------------------------------------------------------------------------------
> This message is intended only for the personal and confidential use of the designated recipient(s) named above.  If you are not the intended recipient of this message you are hereby notified that any review, dissemination, distribution or copying of this message is strictly prohibited.  This communication is for information purposes only and should not be regarded as an offer to sell or as a solicitation of an offer to buy any financial product, an official confirmation of any transaction, or as an official statement of Lehman Brothers.  Email transmission cannot be guaranteed to be secure or error-free.  Therefore, we do not represent that this information is complete or accurate and it should not be relied upon as such.  All information is subject to change without notice.
> 


-- 
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] tag 11 format restrictions and new clordid in cxl and repl msg

[Narayan, A. <Arv...@le...>]

AFAIK, you have to populate all field data yourself. The only exceptions
that I know of are:
=20
MsgSeqNum
SendingTime
TransactTime=20
SenderCompID
TargetCompID
=20
The sequence number is managed by the engine.
SendingTime and TransactTime are set on creation
SenderCompID and TargetCompID are in the session file
=20
Some other fields can have default values. (See the configuration
section of the QuickFix manual)
=20
all orderid type stuff is managed by the application.=20
=20
Ideally, you provide a FIX interface to your clients, and substitute
=66ields where necessary so the exchange is happy.
=20
=20
	-----Original Message-----
	From: qui...@li...
[mailto:qui...@li...] On Behalf Of
Asim
	Sent: Thursday, February 10, 2005 9:27 AM
	To: qui...@li...
	Subject: [Quickfix-developers] tag 11 format restrictions and
new clordid in cxl and repl msg
=09
=09
	Hi all - my first post here, Just started looking into q.fix.
	=20
	While using quickfix API, all the orderids set has to be
maintained by the application=3F Or the api can do that as well =3F ya
know..clorder, orderid, origclorderid...in cxls and cxl replace
messages,
	Also some firms have diff requirement in cxl messages....like
they never want a new clorderid in cxl message and no new clorderid in
cxl/replace msg allowed while reducing quantity, Looks like violating
=66ix rules=3F Another prob is some firms restrict the client sides to
=66ormat their clorder like yyyymmdd-999, whereas others leave it on order
sending firm's discretion.
	So I was wondering if quickfix has the capability to cope with
these variant requirements=3F
	=20
	Please help!
	=20
	Thanks,
	Asim
	=20

---------------------------------------------------------------------------=
---
This message is intended only for the personal and confidential use of the =
designated recipient(s) named above.  If you are not the intended recipient=
 =
of this message you are hereby notified that any review, dissemination, =
distribution or copying of this message is strictly prohibited.  This =
communication is for information purposes only and should not be regarded a=
s=
 an offer to sell or as a solicitation of an offer to buy any financial =
product, an official confirmation of any transaction, or as an official =
statement of Lehman Brothers.  Email transmission cannot be guaranteed to b=
e=
 secure or error-free.  Therefore, we do not represent that this informatio=
n=
 is complete or accurate and it should not be relied upon as such.  All =
information is subject to change without notice.