Re: [Quickfix-developers] Changed code to suppress RefMsgType, RefTagID and SessionRejectReason tags in FIX 4.1 Reject Message

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

Bernard,

One other thing I would suggest is to change the check from:

if(beginString != FIX::BeginString_FIX41)

to:

if(beginString >= FIX::BeginString_FIX42)

That way earlier versions like 4.0 will also be sent without those fields.
In order to do this you will also need to add a new operator overload for >
= to the StringField class.  Alternatively you can just get the field out
as a string ( using getValue() ) and do the comparison against that.

Thanks for the patch.

--oren



|---------+----------------------------------------------->
|         |           "Bernard Spanger"                   |
|         |           <bsp...@gx...>             |
|         |           Sent by:                            |
|         |           qui...@li...|
|         |           ceforge.net                         |
|         |                                               |
|         |                                               |
|         |           05/10/2002 10:28 AM                 |
|         |                                               |
|---------+----------------------------------------------->
  >----------------------------------------------------------------------------------------------|
  |                                                                                              |
  |       To:       <qui...@li...>                                  |
  |       cc:                                                                                    |
  |       Subject:  [Quickfix-developers] Changed code to suppress RefMsgType, RefTagID and      |
  |        SessionRejectReason tags in FIX 4.1 Reject Message                                    |
  >----------------------------------------------------------------------------------------------|




If anyone is using QuickFIX for a 4.1 session, following are the changes I
made to Session.cpp to suppress the RefMsgType, RefTagID and
SessionRejectReason tags in the Reject Message.  Before this change, a
reject message caused each side to continue to exchange infinite reject
messages.  This seems to take care of the problem.

             - Bernard

----------------------------------------------------------------------------

----

  void Session::generateReject( const Message& message, int err, int field
)
  {
    Message reject;
    reject.getHeader().setField(MsgType("3"));
    fill( reject.getHeader() );

             FIX::BeginString beginString;    // Changed by Bernard on
5/10/02.
             message.getHeader().getField(beginString);

    MsgSeqNum msgSeqNum;
    MsgType msgType;
    message.getHeader().getField(msgType);
    message.getHeader().getField(msgSeqNum);
             if (beginString != FIX::BeginString_FIX41) {  // Changed by
Bernard on
5/10/02.
                 reject.setField(RefMsgType(msgType));
                 reject.setField(SessionRejectReason(err));
             }
    reject.setField(RefSeqNum(msgSeqNum));
    if( msgType != "A" && msgType != "4")
      m_pStore->incrNextTargetMsgSeqNum();

    switch( err )
      {
      case 0:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField(Text("Invalid tag number")); break;
      case 1:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField(Text("Required tag missing")); break;
      case 2:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField(Text("Tag not defined for this message type"));
        break;
      case 4:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField(Text("Tag specified without a value")); break;
      case 5:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField
          (Text("Value is incorrect (out of range) for this tag"));
        break;
      case 6:
                         if (beginString != FIX::BeginString_FIX41)  //
Changed by Bernard on
5/10/02.
                                     reject.setField(RefTagID(field));
        reject.setField(Text("Incorrect data format for value")); break;
      case 9:
        reject.setField(Text("CompID problem")); break;
      case 10:
        reject.setField(Text("SendingTime accuracy problem")); break;
      case 11:
        reject.setField(Text("Invalid MsgType")); break;
      };

    sendRaw(reject);
  }

  void Session::generateReject( const Message& message, const std::string&
str )
  {
    Message reject;
    reject.getHeader().setField(MsgType("3"));
    fill( reject.getHeader() );

             FIX::BeginString beginString;  // Changed by Bernard on
5/10/02.
             message.getHeader().getField(beginString);

    MsgType msgType;
    MsgSeqNum msgSeqNum;
    message.getHeader().getField(msgType);
    message.getHeader().getField(msgSeqNum);
             if (beginString != FIX::BeginString_FIX41)  // Changed by
Bernard on
5/10/02.
                 reject.setField(RefMsgType(msgType));
    reject.setField(RefSeqNum(msgSeqNum));

    if( msgType != "A" && msgType != "4")
      m_pStore->incrNextTargetMsgSeqNum();
    reject.setField(Text(str));
    sendRaw(reject);
  }

------------------------------------------
Bernard Spanger
bsp...@gx...
G. X. Clarke & Co.
10 Exchange Place
Jersey City, NJ 07302
201-200-3607



_______________________________________________________________

Have big pipes? SourceForge.net is looking for download mirrors. We supply
the hardware. You get the recognition. Email Us: ban...@so...
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers