Jim,

I tested the files you sent to me. What I discovered was that you have to base your OLEPUBLIC class on Session rather than Custom. In fact, I tried several other VFP base classes. The only one that worked was Session. I'll add this to my list of rules.


>Craig,
>
>My replies are inline.
>
>>There are two rules for creating a UFL in VFP:
>>
>>1. The name of the DLL must begin with CRUFL. This means that your project should also start with CRUFL. By default, the name of the project will be used for the DLL. I can't tell from your postings if you've done that.
>>
>
>The project is named CRUFLTST.PJX and the DLL is CRUFLTST.DLL. On the Servers tab of the Project Info the Project Name is CRUFLTST.
>
>>2. You must declare the parameter and return value data types for the public methods. You've done that in your code.
>>
>
>Yep.
>
>>The last thing to remember is that Crystal will read the registry when it loads, looking for CRUFL. If you build the DLL while Crystal is running, you'll have to Exit, then reload it.
>>
>
>I learned this (and a few other gotchas) early on and have studiously avoided running VFP7 and CR at the same time.
>
>I have a test system with removable hard drives, so it is easy for me to swap hard drives for testing. I have installed and re-installed both VFP7 and CR so many times I do it in my dreams. At this point I am certain it is not something peculiar to my system. It must be one of three things: VFP's DLL creation, CR9's U2LCom.dll and CR8.5's currently patched U2LCom.dll, or something I have not figured out about how to create UFL's. The fact that the original install of CR8.5's U2LCom.dll works makes me very suspicious of CR, but the fact you are having no problem accessing UFLs with VFP7 and CR9 make me suspect there is something about which I am being obtuse or the CR documentation is being unclear.
>
>I could e-mail you a zipped copy of the 3 files in my test project and see what happens when you build and access it on your system. That might help point the way. It is only 1.6k bytes.
>
>I appreciate your help in chasing this down.
>
>...Jim