Is there a way to recreate the structure of the data dictionary (filedict.dbf/mdx) for a file that you just have the source for and database for?  The only way I can think of is to by hand; compile, find fields not defined, figure out what they should be (types, and length) then define all of them again...  this is kinda time consuming...
 
I also understand that I can export all the information for the *.mdx and *.dbf info for the fields that are needed, I'm just wondering if this also works like I think:
 
1.  Create program/define fields
2.  Export FD's into XFR_*  files
3.  Copy said XFR_* files to server
4.  Import XFR_* files
 
Now the FDs will be updated with the new fields in the TAS6 prog that was created?
 
If there is no way to have the fields recreated in the above problem, I can see if I can get a hold of the original FDs (but I do not think I can...)
 
Thanks,
-ITman
 
P.S.  This forum is immensely helpful!

You need to be REALLY careful changing data structures because it potentially makes you an "orphan" in terms of any future updates.  If possible, the first choice would be to take advantage of existing unused fields and files.  Second, add additional files to the database that have the structure you need and can be linked to existing files, and finally as a last resort, change the structure of existing files.
 
There is a TAS6 program that runs the first time you load Evo that reads the DBA data dictionary and constructs the codebase dictionary but my understanding is, it already has the structure of the files and fields defined and is only reading FILELOC.B to establish the paths to the different companies.  There is a "clean" set of DBF/MDX files available at http://www.istechsupport.com/datadict.asp