Sterling Trader and C++

[clearinghouse]

The manual says to avoid C++ altogether because it's just a bad idea. However, I'm wondering if anyone out there has actually worked with Sterling using C++. It just seems like an exercise in heavy COM/ATL programming, starting with an #import of their typelib in VC++.

1) My dilemma is that I have a sizable code base for placing orders all in C++, and Sterling throws a wrench my way since I'll have to write a bridge (or, compile everything with C++/CLI and use interop services.) I'm just curious if anyone else out there has solved this problem or has a commercial solution that functions as a bridge to the usual code.

2) The second question I have is why they think the XML functions are better than the standard functions? I'm trying to understand what it is about the mechanics of the underlying API that would make this true, seeing as parsing XML sounds very time consuming as it is.

Thank you.

 

[tradelink]

there is a c++ interface in tradelink for standard client functions, so you can talk to any of brokers including sterling.

look at tradelink.org in source->trunk->brokerservers->tradelibfast->TLClient_WM.cpp

 

[clearinghouse]

Quote from tradelink:

there is a c++ interface in tradelink for standard client functions, so you can talk to any of brokers including sterling.

look at tradelink.org in source->trunk->brokerservers->tradelibfast->TLClient_WM.cpp

I look at your stuff before I do anything -- the code base is a great set of examples for how to achieve very specific things. I keep tabs on your code.

I saw how you were doing things so I was afraid I'd have to do something similar (or adapt yours), so I wondered if anyone was out there integrating the code directly without having some kind of IPC mechanism inbetween.

The part that stood out to me was that they already said there were performance issues in their document, then offered this XML alternative. So if I use their XML-mode AND IPC, just how much am I shooting myself in the foot here?