quickfix-developers — QuickFIX developers forum

Re: [Quickfix-developers] Privately defined messages

[<OM...@th...>]

First, one point.  Your messages can be in whatever namespace you like,
however you will likely want to inherit from the FIX42::Message class.
Basically if you wanted to put that message in another namespace, like FIX
or something else, it would look like this:

namespace FIX
{
   class UsrMessage1 : public FIX42::Message
  {
  ...
  }
}

This does a couple things for you.  First it will set the BeginString in
the header to FIX.4.2 for you, and second it will supply your message with
a typesafe FIX4.2 header.  You don't strictly *need* this (you can set the
beginstring in the constructor or manually in your code if you like),
however this is the best way.

What you have here looks reasonable.  In fact very similar to what QF would
be using for standard messages (except the if/else statement is
encapsulated in the MessageCrackers).   You can perhaps even use the QF
code generation (see spec\generate_c++.bat) to do much of this work for
you.  In the future I hope to set up the code generation so it is more
friendly for creating user defined standards.  That way you just create an
XML file and the rest is taken care of for you.

--oren



|---------+----------------------------------------------->
|         |           "Hugo Leote"                        |
|         |           <Hug...@in...>         |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           02/04/2003 08:31 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       qui...@li...                                    |
  |       cc:                                                                                    |
  |       Subject:  [Quickfix-developers] Privately defined messages                             |
  >----------------------------------------------------------------------------------------------|





Hi,

I had to define some private (user defined) messages that have been
agreeed by both firms. After a bit of work, I got something that works.
However, I wonder if there is a cleaner way to do this.
The code follows.

Somewhere in "myfields.h":

(...)
             namespace FIX
             {
             USER_DEFINE_STRING(UsrField1, 5000);
             USER_DEFINE_STRING(UsrFiled2, 5001);
             }
(...)


Somewhere in "mymessages.h":

(...)
             namespace FIX
             {
             const char MsgType_UsrMessage1[] = "U1";
             }


             namespace FIX42
             {
                         // Had to do this in FIX42 namespace.
                         // User messages defined inside the FIX namespace
                         // always cause an automatic "reject" on
receiving,
                         // thereby never reaching crack().

             class UsrMessage1 : public Message
             {
             public:
                         UsrMessage1() : Message( MsgType() ) {}
                         UsrMessage1 ( const Message& m ) : Message( m ) {}
                         static FIX::MsgType MsgType() { return
FIX::MsgType(
FIX::MsgType_UsrMessage1 ); }

                         FIELD_SET( *this, FIX::UsrField1 );
                         FIELD_SET( *this, FIX::UsrField2 );
             };

             }
(...)


Somewhere in "myfixapplication.cpp" :

(...)
             // Receive user message

             void CFixApplication::onMessage (const FIX42::Message&
message,
                                                             const
FIX::SessionID& sessionID)
             {


                                     // Ugly. Had to create this "catch
all" to handle
the user defined messages
                                     // Maybe there's a
cleaner/prettier/better way to
do this...

                         FIX::MsgType msgType;
                         message.getHeader().getField( msgType );

                         if (msgType == FIX::MsgType_UsrMessage1)
                                     onMessage((FIX42::UsrMessage1)message,
sessionID);
                         else
                                     throw FIX::UnsupportedMessageType();
             }

             void CFixApplication::onMessage (const FIX42::UsrMessage1&
message,
                                                             const
FIX::SessionID& sessionID)
             {
                         // TODO: Proper handling of this message...
                         AfxMessageBox ("Got proprietary message
UsrMessage1 !");
             }
(...)



Thanks in advance,
Hugo Leote



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] Privately defined messages

[Hugo L. <Hug...@in...>]

Hi,

I had to define some private (user defined) messages that have been 
agreeed by both firms. After a bit of work, I got something that works. 
However, I wonder if there is a cleaner way to do this.
The code follows.

Somewhere in "myfields.h":

(...)
	namespace FIX
	{
	USER_DEFINE_STRING(UsrField1, 5000);
	USER_DEFINE_STRING(UsrFiled2, 5001);
	}
(...)
	

Somewhere in "mymessages.h":

(...)
	namespace FIX
	{
	const char MsgType_UsrMessage1[] = "U1";
	}
	
	
	namespace FIX42
	{
		// Had to do this in FIX42 namespace. 
		// User messages defined inside the FIX namespace 
		// always cause an automatic "reject" on receiving, 
		// thereby never reaching crack().
		
	class UsrMessage1 : public Message
	{
	public:
		UsrMessage1() : Message( MsgType() ) {}
		UsrMessage1 ( const Message& m ) : Message( m ) {}
		static FIX::MsgType MsgType() { return FIX::MsgType( 
FIX::MsgType_UsrMessage1 ); }
	
		FIELD_SET( *this, FIX::UsrField1 );
		FIELD_SET( *this, FIX::UsrField2 );
	};
	
	}
(...)


Somewhere in "myfixapplication.cpp" :

(...)
	// Receive user message
	
	void CFixApplication::onMessage (const FIX42::Message& message, 
					const FIX::SessionID& sessionID)
	{								
	
			// Ugly. Had to create this "catch all" to handle 
the user defined messages
			// Maybe there's a cleaner/prettier/better way to 
do this...
	
		FIX::MsgType msgType;
		message.getHeader().getField( msgType );
	
		if (msgType == FIX::MsgType_UsrMessage1)
			onMessage((FIX42::UsrMessage1)message, sessionID);
		else
			throw FIX::UnsupportedMessageType();
	}
	
	void CFixApplication::onMessage (const FIX42::UsrMessage1& 
message, 
					const FIX::SessionID& sessionID)
	{
		// TODO: Proper handling of this message...
		AfxMessageBox ("Got proprietary message UsrMessage1 !");
	}
(...)



Thanks in advance,
Hugo Leote

[Quickfix-developers] ASIAN FIX Conference 2003

[Asian F. <asi...@eu...>]

<html>
<head>
	<title>Asian Fix Conference 2003</title>
</head>

<body bgcolor="White">
<table border="0" width="600" cellpadding="0" cellspacing="0">
	<tr>
		<td colspan="2"><img
src="http://www.asianfix2003.com/images/invitetop.gif" border="0"></td>
	</tr>
	<tr>
		<td><a href="http://www.asianfix2003.com/"><img
src="http://www.asianfix2003.com/images/asianfixlogo.gif"
border="0"></a></td>
		<td align="center"><b><font face="Arial" size="4"><font color="Navy">The
Life Cycle of a Trade</font><br>
			<font color="Crimson"> - achieving a seamless front to back process
utilising FIX and STP</font>
			</font></b><br>
		<a href="http://www.asianfix2003.com/registration.asp"><img
src="http://www.asianfix2003.com/images/button.gif" border="0"></a></td>
	</tr>
	<tr>
		<td colspan="2">
			<p><b><font face="Arial" size="2">The Asian Pacific FIX* (Financial
Information eXchange) 
			Committee will host the FIRST Asian FIX conference this March in Hong
Kong. This event 
			seeks to leverage the phenomenal success of the European FIX
conferences. Where buy and sell 
			side participants and vendor specialists, have explored the dynamic
world of electronic trading 
			and connectivity, to formulate solutions to the automated trading
challenge.</font></b></p>
			<p><b><font face="Arial" size="2">The one-day event features a series
of presentations, 
			panel discussions, working group and interactive voting covering the
general theme of the 
			past, present and future of trading technologies and technological
advancements.</font></b></p>
			<p><b><font face="Arial" size="2">The day will be brought to a close
with a cocktail evening, 
			giving you the opportunity to talk to everyone involved on the day.    
PLEASE REQUEST FOR AN INVITATION NOW!!!!</font></b></p>
			<p><b><font face="Arial" size="2"><i>More information is available at 
			<a href="http://www.asianfix2003.com">www.asianfix2003.com</a>. For
enquiries, please 
			contact Barbara Law via <a
href="mailto:bar...@eu...">bar...@eu...</a> 
			or at (852) 2810 6201.</i></font></b></p>
			<p align="right"><a href="http://www.fixprotocol.org/"><img
src="http://www.asianfix2003.com/images/fixprotocol.gif" border="0"></a>
		</td>
	</tr>
</table>


</body>
</html>

Re: [Quickfix-developers] Using QuickFix to analyze logged messages.

[<OM...@th...>]

m.getField(subID);

should be:

m.getHeader().getField(subID);

--oren



|---------+----------------------------------------------->
|         |           "Rob Kulseth" <ro...@ri...>   |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           02/03/2003 09:57 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       <qui...@li...>                                  |
  |       cc:                                                                                    |
  |       Subject:  [Quickfix-developers] Using QuickFix to analyze logged messages.             |
  >----------------------------------------------------------------------------------------------|




I have a FIX message log of all orders that I have sent on a particular
day.  I need to parse those messages to get a report of all traffic for a
particular user.  I tried the following strategy using C#

Fix.Message m = new Fix.Message(mes);
Fix.SenderSubID subID = new SenderSubID();
m.getField(subID);

mes is the raw fix message read from a database. I always get an Exception
"FixFieldNotFound" even though inspection of the string reveals that tag 50
is present.  I realize I am using QuickFix slightly different than it was
designed for, but I think I should be able to do this.  any suggestions on
how to parse Fix messages this way?



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] Using QuickFix to analyze logged messages.

[Rob K. <ro...@ri...>]

I have a FIX message log of all orders that I have sent on a particular =
day.  I need to parse those messages to get a report of all traffic for =
a particular user.  I tried the following strategy using C#

Fix.Message m =3D new Fix.Message(mes);
Fix.SenderSubID subID =3D new SenderSubID();
m.getField(subID);

mes is the raw fix message read from a database. I always get an =
Exception "FixFieldNotFound" even though inspection of the string =
reveals that tag 50 is present.  I realize I am using QuickFix slightly =
different than it was designed for, but I think I should be able to do =
this.  any suggestions on how to parse Fix messages this way?

Re: [Quickfix-developers] How to implement our own authorization scheme

[<OM...@th...>]

I think the only thing holding you back from using the cracker to do th=
is
is the exceptions that it throws.  In the fromAdmin callback you can th=
row
a RejectLogon exception, which will do exactly that with a logon messag=
e.
The cracker currently doesn't allow you to throw this exception so you =
can
either:

A) Modify the onMessage( Logon, SessionID ) method to allow it to throw=

RejectLogon
B) In your fromAdmin, cast Logon messages yourself and do your check
directly in that method.

Then you can look inside the message and decide whether you want to thr=
ow
the RejectLogon exception.

--oren



|---------+----------------------------------------------->
|         |           Joerg Thoennes                      |
|         |           <Joe...@ma...>           |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/31/2003 10:07 AM                 |
|         |           Please respond to Joerg.Thoennes    |
|         |                                               |
|---------+----------------------------------------------->
  >--------------------------------------------------------------------=
--------------------------|
  |                                                                    =
                          |
  |       To:       qui...@li...          =
                          |
  |       cc:                                                          =
                          |
  |       Subject:  [Quickfix-developers] How to implement our own auth=
orization scheme          |
  >--------------------------------------------------------------------=
--------------------------|




Hello,

currently we develop a FIX gateway to some other system. We use Java 1.=
4
and the provided JNI
library. I think the preferred way to use the message cracker for
processing is very usable.
But we need to configure session level processing in that way, that we
check the login names
against an application server and then tell the FIX engine whether the
Logon is accepted.
Is there any hook to do this?

Thanks, J=F6rg

--
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




-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld =3D Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers



=

[Quickfix-developers] How to implement our own authorization scheme

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

Hello,

currently we develop a FIX gateway to some other system. We use Java 1.4 and the provided JNI
library. I think the preferred way to use the message cracker for processing is very usable.
But we need to configure session level processing in that way, that we check the login names
against an application server and then tell the FIX engine whether the Logon is accepted.
Is there any hook to do this?

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

[Quickfix-developers] Quickfix examples: gcc 3.0 syntax errors (std::type_info)

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

Hello,

after compiling and installing the Quickfix 1.3.2 distribution, I started to compile the
examples. The first problem was easily fixed: The include files were not found and so
I made a link from the examples directory:

/opt2/QuickFIX/quickfix-1.3.2/quickfix/examples $ ll quickfix
lrwxrwxrwx    1 joerg    joerg          14 Jan 29 18:09 quickfix -> ../../quickfix

Perhaps my configuration steps were wrong:

	./bootstrap
	./configure

But then the GCC 3.0 complained about "SO C++ forbids declaration of `type_info' with no
  type". (For the complete error log see below.)
It was always the same kind of line in Test.h:

               const type_info& type = typeid(this);

All other errors origin in this one. Searching the web I found that this line should be

               const std::type_info& type = typeid(this);
                     ^^^^^
but there is a "using std::type_info" statement in the header section of this file:

include <typeinfo>
#if TYPEINFO_IN_STD
using std::type_info;
#endif

I checked config.h to see that TYPEINFO_IN_STD is defined and therefore the "using" line
is active.

Now I put "std::" before all occurences of type_info to get form suggested in the web.
But this resulted in lots of error messages...

Please, could anybody with more C++ knowledge point out the right way to fix this problem?

Cheers, Jörg

---- make log -----

make[3]: Entering directory `/net/fileserver/export/opt2/QuickFIX/quickfix-1.3.2/quickfix/examples/ordermatch/test'
g++-3.0 -DHAVE_CONFIG_H -I. -I. -I../.. -I.. -I../../..    -Wall -fexceptions -finline-functions -I/usr/include/libxml2/libxml -I/usr/include/libxml2  -c ordermatch_ut.cpp
In file included from ../../../CPPTest/TestCase.h:4,
                  from ../../../CPPTest/TestSuite.h:4,
                  from TestSuite.h:51,
                  from ordermatch_ut.cpp:55:
../../../CPPTest/Test.h: In member function `bool
    CPPTest::Test<TClass>::setup(CPPTest::TestDisplay&)':
../../../CPPTest/Test.h:46: ISO C++ forbids declaration of `type_info' with no
    type
../../../CPPTest/Test.h:46: parse error before `;' token
../../../CPPTest/Test.h: In member function `void
    CPPTest::Test<TClass>::teardown(CPPTest::TestDisplay&)':
../../../CPPTest/Test.h:62: ISO C++ forbids declaration of `type_info' with no
    type
../../../CPPTest/Test.h:62: parse error before `;' token
../../../CPPTest/Test.h: In member function `bool
    CPPTest::Test<TClass>::run(CPPTest::TestDisplay&)':
../../../CPPTest/Test.h:72: ISO C++ forbids declaration of `type_info' with no
    type
../../../CPPTest/Test.h:72: parse error before `;' token
../../../CPPTest/Test.h:88: ISO C++ forbids declaration of `type_info' with no
    type
../../../CPPTest/Test.h:88: parse error before `;' token
../../../CPPTest/Test.h: In member function `bool
    CPPTest::Test<TClass>::setup(CPPTest::TestDisplay&) [with TClass =
    OrderMatcher]':
/usr/include/g++-v3/bits/locale_facets.tcc:75:   instantiated from here
../../../CPPTest/Test.h:46: uninitialized const `type_info'
../../../CPPTest/Test.h:47: `type' undeclared (first use this function)
../../../CPPTest/Test.h:47: (Each undeclared identifier is reported only once
    for each function it appears in.)
../../../CPPTest/Test.h:46: warning: unused variable `const int type_info'
../../../CPPTest/Test.h: In member function `void
    CPPTest::Test<TClass>::teardown(CPPTest::TestDisplay&) [with TClass =
    OrderMatcher]':
/usr/include/g++-v3/bits/locale_facets.tcc:75:   instantiated from here
../../../CPPTest/Test.h:62: uninitialized const `type_info'
../../../CPPTest/Test.h:62: warning: unused variable `const int type_info'
../../../CPPTest/Test.h: In member function `bool
    CPPTest::Test<TClass>::run(CPPTest::TestDisplay&) [with TClass =
    OrderMatcher]':
/usr/include/g++-v3/bits/locale_facets.tcc:75:   instantiated from here
../../../CPPTest/Test.h:72: uninitialized const `type_info'
/usr/include/g++-v3/bits/locale_facets.tcc:75:   instantiated from here
../../../CPPTest/Test.h:72: warning: unused variable `const int type_info'
../../../CPPTest/Test.h:88: uninitialized const `type_info'
../../../CPPTest/Test.h:88: warning: unused variable `const int type_info'

-------

-- 
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 & J2EE

[Alex H. <al...@an...>]

On Wed, 2003-01-29 at 18:02, CW Johnson wrote:
> I use QuickFIX + JBoss. I built a fix engine using QuickFIX, which would
> listen for and send messages using FIX. The fix engine also interfaces
> with JBoss via JMS. So far, things have been pretty smooth.
> 
> For examples, I guess I have something sitting around in my repository.
> I made the examples provided with QuickFIX work with JMS, as a
> proof-of-concept.. I'll dig it out for you.
> 
> 
> HTH,
> cedrick

Hi Cedrick,

That would be most helpful.

Thanks,
Alex.

RE: [Quickfix-developers] QuickFIX & J2EE

[CW J. <jav...@ce...>]

I use QuickFIX + JBoss. I built a fix engine using QuickFIX, which would
listen for and send messages using FIX. The fix engine also interfaces
with JBoss via JMS. So far, things have been pretty smooth.

For examples, I guess I have something sitting around in my repository.
I made the examples provided with QuickFIX work with JMS, as a
proof-of-concept.. I'll dig it out for you.


HTH,
cedrick


-----Original Message-----
From: qui...@li...
[mailto:qui...@li...] On Behalf Of
Alex Hornby
Sent: Wednesday, January 29, 2003 5:00 AM
To: quickfix-developers
Subject: [Quickfix-developers] QuickFIX & J2EE


Hi,

Has anyone here used the QuickFIX Java bindings to accept FIX messages
inside a J2EE app server? (In particular JBoss)

The reason behind this is that I'd like to have both incoming FIX and
web page requests affect Java "business objects".

Is anyone doing this - any sample code?

Cheers,
Alex.



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

RE: [Quickfix-developers] QuickFIX & J2EE

[Day, J. B. W. <Je...@ba...>]

Alex,

  I should probably start by saying that i/we are not using
  QuickFIX (i'm a lurker!) so the following is pure conjecture..

  This looks like the sort of problem that the J2EE connector
  architecture was designed to address

            http://java.sun.com/j2ee/connector/

  
  It's also possible you might be able to deploy an MBean into
  JBoss that then would call-out to your Session Bean business
  logic.

  I would be interesting to hear how other people have achieved
  this (in a J2EE compliant manner).

  One thing to consider is how you would address scalability
  issues, one potential solution (although much more work) would
  be to have your J2EE logic send/recieve JMS messages and have
  some sort of external gateway process acting as a bridge between
  your JMS Messages and QuckFIX. By separating the "FIX Gateway"
  you have opened the door to running multiple gateways and a
  fully clustered J2EE environment.

  You might also take a look at STPml (http://www.stpml.org/)
  which is an XML abstraction of the trading life-cycle and
  (should) map to FIX, SWIFT, etc.

Jem..

-----Original Message-----
From: Alex Hornby [mailto:al...@an...]
Sent: Wednesday, January 29, 2003 3:00 AM
To: quickfix-developers
Subject: [Quickfix-developers] QuickFIX & J2EE


Hi,

Has anyone here used the QuickFIX Java bindings to accept FIX messages
inside a J2EE app server? (In particular JBoss)

The reason behind this is that I'd like to have both incoming FIX and
web page requests affect Java "business objects".

Is anyone doing this - any sample code?

Cheers,
Alex.



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

Re: [Quickfix-developers] QuickFIX & J2EE

[<gar...@su...>]

We're using QuickFIX to talk to our J2EE server (weblogic) but don't have
it directly within it.  We've found that it's generally too risky to run
anything that uses native interfaces within the app server.  Not that
QuickFIX has really caused any problems, but we've had 3rd party software
sometimes cause core dumps and if that happens, there's not much you can do
to save your app server from going down.

Instead, I've created a separate FIX Engine (using QuickFIX) that acts as
an RMI server that our app server can talk to and listen from for sending
and receiving FIX messages.

Gary Mui
Prescient Markets, Inc         914-989-3118  (W)
445 Hamilton Avenue          914-422-3693  (F)
White Plains, NY 10601

Please visit us at http://www.cpmarket.com


                                                                                                                                       
                    Alex Hornby <al...@an...>                                                                                       
                    Sent by:                                     To:     quickfix-developers                                           
                    qui...@li...urc        <qui...@li...>                           
                    eforge.net                                   cc:                                                                   
                                                                 Subject:     [Quickfix-developers] QuickFIX & J2EE                    
                                                                                                                                       
                    01/29/03 05:59 AM                                                                                                  
                                                                                                                                       
                                                                                                                                       



Hi,

Has anyone here used the QuickFIX Java bindings to accept FIX messages
inside a J2EE app server? (In particular JBoss)

The reason behind this is that I'd like to have both incoming FIX and
web page requests affect Java "business objects".

Is anyone doing this - any sample code?

Cheers,
Alex.



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] QuickFIX & J2EE

[Alex H. <al...@an...>]

Hi,

Has anyone here used the QuickFIX Java bindings to accept FIX messages
inside a J2EE app server? (In particular JBoss)

The reason behind this is that I'd like to have both incoming FIX and
web page requests affect Java "business objects".

Is anyone doing this - any sample code?

Cheers,
Alex.

Re: [Quickfix-developers] Acceptor cleanup problem when port is busy (QuickFIX 1.3.2)

[<OM...@th...>]

Actually we can probably handle the onRun situation.  Right now QF first
starts the onRun thread, then does the initialization.  I think this is
backwards.  It should first initialize (create all the sessions, files,
sockets, blah blah), and only then start the onRun thread if everything is
succesful.  That way in this situation we can throw an exception like we do
for everything else.  In this case the developer doesn't need to clean up
onRun since it never actually started.

--oren



|---------+----------------------------------------------->
|         |           Gene Gorokhovsky                    |
|         |           <mus...@ya...>                |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/28/2003 12:03 PM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       qui...@li...                                    |
  |       cc:                                                                                    |
  |       Subject:  [Quickfix-developers] Acceptor cleanup problem when port is busy (QuickFIX   |
  |        1.3.2)                                                                                |
  >----------------------------------------------------------------------------------------------|




Quickfix 1.3.2 has a following problem:
when the port specified for acceptor is already taken
(for example FIX server is started erroneously twice),
acceptor fails, but gives no indication to of that to
the host program.  Acceptor (both ST and MT) "start"
method still returns true as if application has shut
down normally, and no Application method is called to
indicate that something is wrong.

Since thread that executes Application->onRun does not
exit on user action (as is normally the case),  and no
indication is given to program that something is
wrong, this behavior usually results in an unpleasant
situtaion such as deadlock (when thread_join is
attempted with never-exiting onRun), or a crash (on my
Linux due to a libpthread bug thread_join returns
fine, but onRun is still executing and crashes when
application class is cleaned up)

Although there is nothing that FIX engine can do to
stop onRun thread by itself, it should somehow
indicate to the host program that the acceptor failed
to initiate the accepting socket. Returning "false"
from the start method would allow the developers to
clean up onRun themselves.

Fixing return value of the start method will require
small modifications to SocketAccepor::onStart, and
ThreadedSocketAcceptor::onStart; if the blocking call
(SocketServer->block or socket_accept) fails, the
whole method should return false.

Gene

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] Acceptor cleanup problem when port is busy (QuickFIX 1.3.2)

[Gene G. <mus...@ya...>]

Quickfix 1.3.2 has a following problem:
when the port specified for acceptor is already taken
(for example FIX server is started erroneously twice),
acceptor fails, but gives no indication to of that to
the host program.  Acceptor (both ST and MT) "start"
method still returns true as if application has shut
down normally, and no Application method is called to
indicate that something is wrong. 

Since thread that executes Application->onRun does not
exit on user action (as is normally the case),  and no
indication is given to program that something is
wrong, this behavior usually results in an unpleasant
situtaion such as deadlock (when thread_join is
attempted with never-exiting onRun), or a crash (on my
Linux due to a libpthread bug thread_join returns
fine, but onRun is still executing and crashes when
application class is cleaned up)

Although there is nothing that FIX engine can do to
stop onRun thread by itself, it should somehow
indicate to the host program that the acceptor failed
to initiate the accepting socket. Returning "false"
from the start method would allow the developers to
clean up onRun themselves. 

Fixing return value of the start method will require
small modifications to SocketAccepor::onStart, and
ThreadedSocketAcceptor::onStart; if the blocking call
(SocketServer->block or socket_accept) fails, the
whole method should return false.

Gene

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

[Quickfix-developers] 3rd party apps for viewing fix message in log files

[Han T. <han...@am...>]

I was reading through the mailing list archives and
saw a reference to 3rd party apps for viewing fix
messages in the quickfix log - can someone post a url
where one can download such an app?

Thanks.

[Quickfix-developers] help

[Kannan R <kan...@ss...>]

Hi All,

    I am having problem in starting the QuickFIX Engine..Can any one tell me
how to start the Engine.

I installed Quickfix and generated Quickfix.jar also for Java..After that I
don't know how to start the engine and

how to proceed further.Please help.

Regards

Kannan.R
****************************************************************************
This communication contains information, which is confidential and may also
be privileged. It is for the exclusive use of the intended recipient(s). If
you are not the intended recipient(s), please note that any distribution,
printing, copying or use of this communication or the information in it is
strictly prohibited. If you have received this communication in error,
please notify the sender immediately and then destroy any copies of it.
****************************************************************************

[Quickfix-developers] Defining users at runtime

[Bill A. <bad...@ae...>]

 Is there a way to add a user to the SessionSettings at runtime after I
start the acceptor instead of having to predefine them.

Billy




-----Original Message-----
From: Bill Adelman [mailto:bad...@ae...]
Sent: Thursday, January 09, 2003 8:53 AM
To: 'qui...@li...'
Subject: ThreadedSocketAcceptor vrs SocketAcceptor


What are the differences between ThreadedSocketAcceptor.cpp and
SocketAcceptor.cpp and between ThreadedSocketInitiator.cpp and
SocketInitiator.cpp

thanks
Billy

[Quickfix-developers] deploying quickfix.NET applications

[<OM...@th...>]

If you are deplying an application that uses quickfix_net.dll, make sure
that msvcp70.dll is in your path.  We wasted a day on this issue at a
client site.  It turns out this dll was not on any of the client machines
even though they were running windows 2000 and had the latest versions of
the .NET framework installed.  This is the MS C++ library that the
QuickFIX.NET library needs to run its unmanaged code.

The error message that the CLR gives you if this DLL is not present is NOT
at all helpful.  So you can recognize it, this is how it looks.

-----
Application.exe - Common Language Runtime Debugging Services
-----
Application has generated an exception that could not be handled

Process id=0x538(1336), Thread id=0x600(1536).

Click OK to terminate the application.
Click CANCEL to debug the application.
-----

This is pretty much what happens whenever an unmanaged DLL dependency is
not found.  Your applications will just croak.  This will be added to the
FAQ.

--oren

RE: [Quickfix-developers] VB.NET

[Vitor C. <vc...@hi...>]

I've had some other problems with .Net but not exclusively with VB. If
you create any managed components in .Net and try to expose them as COM
objects you will have some problems. This has to do, to the extent of my
knowledge, with a bug in the VC++ compiler in the current version of
Visual Studio regarding the calling of unmanaged code from managed code.
Nonetheless I've compiled both QuickFix and some managed components done
with C# and afterwards exposed them as COM components with the Beta 2 of
Visual Studio 2003 (codenamed Everett) and all worked fine. So the bug
gets fixed in the next Visual Studio version.

Regards
Vitor Castro

-----Original Message-----
From: qui...@li...
[mailto:qui...@li...] On Behalf Of
Rob Kulseth
Sent: sexta-feira, 10 de Janeiro de 2003 22:34
To: qui...@li...
Subject: RE: [Quickfix-developers] VB.NET


I have made the changes using the namespace "FixNet" in my copy of quick
fix.  The change was easy to make.  All I did was search for the text
"Fix::" and replace it with "FixNet::" and change all the namespace
statements to read "namespace FixNet".  This appears to be working fine.
If you guys decide to make a standard namespace change, it shouldn't be
very hard to change on my Code to match.  Are there any other known
VB.net issues?

Thanks for all your help
Rob Kulseth

-----Original Message-----
From: OM...@th... [mailto:OM...@th...]
Sent: Friday, January 10, 2003 4:21 PM
To: Rob Kulseth
Cc: qui...@li...;
qui...@li...
Subject: Re: [Quickfix-developers] VB.NET



Well I think we basically need to come up with a new namespace.  I can
send you what was sent to me (although I'm not sure what version it is
based off of).  John is using the namespace FixDotNet, but I'm thinking
of going with QuickFIX or something (more like the java api).

--oren



|---------+----------------------------------------------->
|         |           "Rob Kulseth" <ro...@ri...>   |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/10/2003 11:32 AM                 |
|         |                                               |
|---------+----------------------------------------------->
 
>-----------------------------------------------------------------------
-----------------------|
  |
|
  |       To:       <qui...@li...>
|
  |       cc:
|
  |       Subject:  [Quickfix-developers] VB.NET
|
 
>-----------------------------------------------------------------------
-----------------------|




Are there plans to make a change to the standard QuickFix Code to make
it useable in VB.NET, or should this change be done ourselves.

Thanks
Rob Kulseth


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers







-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

RE: [Quickfix-developers] VB.NET

[Rob K. <ro...@ri...>]

I have made the changes using the namespace "FixNet" in my copy of quick =
fix.  The change was easy to make.  All I did was search for the text =
"Fix::" and replace it with "FixNet::" and change all the namespace =
statements to read "namespace FixNet".  This appears to be working fine. =
 If you guys decide to make a standard namespace change, it shouldn't be =
very hard to change on my Code to match.  Are there any other known =
VB.net issues?

Thanks for all your help
Rob Kulseth

-----Original Message-----
From: OM...@th... [mailto:OM...@th...]
Sent: Friday, January 10, 2003 4:21 PM
To: Rob Kulseth
Cc: qui...@li...;
qui...@li...
Subject: Re: [Quickfix-developers] VB.NET



Well I think we basically need to come up with a new namespace.  I can =
send
you what was sent to me (although I'm not sure what version it is based =
off
of).  John is using the namespace FixDotNet, but I'm thinking of going =
with
QuickFIX or something (more like the java api).

--oren



|---------+----------------------------------------------->
|         |           "Rob Kulseth" <ro...@ri...>   |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/10/2003 11:32 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  =
>------------------------------------------------------------------------=
----------------------|
  |                                                                      =
                        |
  |       To:       <qui...@li...>          =
                        |
  |       cc:                                                            =
                        |
  |       Subject:  [Quickfix-developers] VB.NET                         =
                        |
  =
>------------------------------------------------------------------------=
----------------------|




Are there plans to make a change to the standard QuickFix Code to make =
it
useable in VB.NET, or should this change be done ourselves.

Thanks
Rob Kulseth


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld =3D Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

Re: [Quickfix-developers] VB.NET

[<OM...@th...>]

Well I think we basically need to come up with a new namespace.  I can send
you what was sent to me (although I'm not sure what version it is based off
of).  John is using the namespace FixDotNet, but I'm thinking of going with
QuickFIX or something (more like the java api).

--oren



|---------+----------------------------------------------->
|         |           "Rob Kulseth" <ro...@ri...>   |
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/10/2003 11:32 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       <qui...@li...>                                  |
  |       cc:                                                                                    |
  |       Subject:  [Quickfix-developers] VB.NET                                                 |
  >----------------------------------------------------------------------------------------------|




Are there plans to make a change to the standard QuickFix Code to make it
useable in VB.NET, or should this change be done ourselves.

Thanks
Rob Kulseth


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] RE: Preliminary support for FIX 4.3 in CVS

[El-Abid, B. <ba...@se...>]

> After we finish up the final touches with 4.3 we are moving on to
> encryption as our next major feature.  Anybody planning to use encryption,
> please let us know which algoritms you want supported.  We will pick the
> most wanted one as the pilot for initial encryption support.

 Hello all,

 We're planning to use the blowfish cryptographic algorithm since 
 our first trading partner uses it.

 Thanks,

Bassam El-Abid
Software Engineer
Sector Data, LLC
222 Sutter Street, Suite 450
San Francisco, CA  94108
Main: 415.782.0454 x245
Fax:  415.782.0474
Mob:  415.652.2390
email:ba...@se...


_____________________________________________________
Sector Data, LLC, is not affiliated with Sector, Inc., or SIAC

[Quickfix-developers] VB.NET

[Rob K. <ro...@ri...>]

Are there plans to make a change to the standard QuickFix Code to make =
it useable in VB.NET, or should this change be done ourselves.

Thanks
Rob Kulseth

RE: [Quickfix-developers] VB.net

[<OM...@th...>]

This sounds to me very likely.  John Lopez recently sent me some code that
fixes the problem and it looks like changing the namespace for the .NET
classes is exactly what he did.

--oren


|---------+----------------------------------------------->
|         |           "Vitor Castro" <vc...@hi...>|
|         |           Sent by:                            |
|         |           qui...@li...ur|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           01/10/2003 06:33 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       "'Rob Kulseth'" <ro...@ri...>                                          |
  |       cc:       "QuickFIX" <qui...@li...>                       |
  |       Subject:  RE: [Quickfix-developers] VB.net                                             |
  >----------------------------------------------------------------------------------------------|




Hi Rob,

I also ran into this problem while evaluating QF. I believe it has to do
with the fact that VB is not case sensitive. As you can see there is a
C++ namespace named FIX where the Application class in defined. There is
also another Managed C++ namespace named Fix where the .Net Application
Wrapper is declared and which is the one you want to use. As VB is not
case sensitive it tries to implement the C++ interface which is not
possible.
I'm not completely sure of this as I was using C# at the time but I hope
it helps.

Regards


-----Original Message-----
From: qui...@li...
[mailto:qui...@li...] On Behalf Of
Rob Kulseth
Sent: segunda-feira, 6 de Janeiro de 2003 19:46
To: qui...@li...
Subject: [Quickfix-developers] VB.net


I am trying to Implement the Fix.Application interface in a VB.net class
file and the IDE keeps telling me that the "Fix.Application is not
accessible in this context because it is 'Private'".  I looked at the
interface definition, and it is defiantly public.  Has anyone
successfully implemented the application interface in VB.net?  Any clue
how to do this in VB.net?

Here is my class definition.  Pretty simple.

Public Class FixConnection : Implements Fix.Application

End Class



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf _______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers