wrong BeginString in generated FIX50 files

[gbirchmeier]

FIX50 generated files are serializing with the wrong BeginString (FIX.5.0 instead of FIX.T or whatever).

Task 1: Make an acceptance test that causes test app to send an application-level message. It should fail because the BeginString is wrong.

Task 2: Fix the BeginString and make the test pass.

[ruaanviljoen]

Is it possible for someone to direct me to the source files or project at least where this is happening? I need to use this quite urgently, and would either like to try and 1) fix it or 2) get a workaround until it is fixed.

I am very new to QuickFix so I have been struggling to find the place where this beginString is actually set to FIX.5.0 or w/e.

[gbirchmeier]

We think in the generated Message class.

Specifically, QuickFIX.NET\Message\FIX50\Message.cs.

Given that Peter van Lonkhuysen is not having the same problem, it appears that maybe the issue isn't quite what we think it is. I will try to look into this over the weekend.

[gbirchmeier]

In my own experimentation, I'm seeing the following:

1. Message.ToString() prints a message that starts with 8=FIX.5.0. I will check with the other QFs to see if this is consistent.

2. When a FIX application serializes a message, it uses the correct BeginString (8=FIXT.1.1).

So, while it appears there may be some error in the ToString command as Ruaan is using it, this should be ok when the app is actually sending messages to the counterparty.

[gbirchmeier]

QF/J outputs "FIXT.1.1" in its to-string messages for FIX 5. I think QF/C++ does too, though I only looked at source and didn't write a test.

It appears that, when sending a message, the begin string is overwritten based on session. During ToString, this doesn't happen.