>>Let me start by saying we've found a workaround, so I'm posting this because I want to know the answer, not because I need a solution.
>>
>>A vertical market application I'm working on has a routine that copies a couple of files from one folder to another, changing the name in the process. It's been working for quite some time, but last week, some (but not all) users started having problems. The error log shows that there's an extra folder in the path for the source file. That is, there's a variable with a value like:
>>
>>

>>C:\MyFolder\MyFile.FRX
>>

>>
>>When COPY FILE is handed this variable, the error message says it can't find:
>>
>>

>>C:\MyFolder\Reports\MyFile.FRX
>>

>>
>>In trying to track this down this morning, we actually replaced the variable with a hard-coded path to the file, so the command was like:
>>
>>

>>COPY FILE C:\MyFolder\MyFile.FRX TO C:\MyOtherFolder\MyOtherFile.FRX
>>

>>
>>We still got an error that showed Reports in the path.
>>
>>What can I add? These aren't actually on C; the folders are on a mapped drive. The source file exists and we have no problem copying it via Explorer.
>>
>>As I said, we've found a work-around. We're using the CopyFile API function and it works as expected.
>>
>>Anybody have a clue what would cause this?
>>
>>Tamar
>
>Hi Tamar
>
>My sole idea is the mapping.
>Does it work with the UNC path?

We didn't think to try that, since the whole application set-up is built around using mapped paths (on a terminal server ... fortunately, the IT part isn't my responsibility).

I think Greg's answer about the reports being in the project as well is the answer, though.

Tamar