Thanks for your thoughts on the subject Thomas.

I guess the Fox 2.6 (Windows) code is not "well written" for many of the reasons that you questioned below.

Having had some time to think it over on the weekend, I've decided that it'd be best to start from scratch and rewrite it in VFP8. There is no pressing deadline for having this app up and running and I'll use it as an excuse to learn some of the newer stuff in 8. I don't know about you, but I like to have a project to work with when learning a new development language/IDE and this one is as good candidate to try it with.

Who knows, if it all works out, I can go to my boss and say we need to upgrade the (old) app to VFP 8.0. When he says how long it'll take, I'll say it done:)

Mike


>Hi Mike,
>
>
>>I tried opening one of the old 2.6 screens
>FPDos or FPWindows ?
>
>>Either one produces some really strange results.
>The converter has some defects. I believe I have "cured" most of them
>for FPW-screens. How abaout sending me one of the "problem" screns ?
>
>>I'm guessing that its because the 2.6 code wasn't "well written"
>Sorry to be not more explicit: Here I meant:
>Is the code worth saving ? not too many bugs / incidents ?
>Is there a division between output and logic or is the code sprinkled into
>snippets of the scrreens and you have to constantly search for a specific function ? Is there duplicate code ? (Are you happy with it ? Is it timesaving to have tested functions?)
>
>>since it >was using a third party code generator (gentab?) for faking a tabbed >container. All of the old screens also had a similar thing to do 3D effects.
>If you are using GenScrnX or another tool you might have more work if you convert the screens... BUT
>
>>I guess the conversion process can't handle these situations very nicely.
>You can run the generated .Spr's in VFP as well (there might be some screen painting issues from different inner workings: 12 screens out of 140 had to be fixed "manually" via explicit changes to .width or something similar:
>done in 2 days!
>
>There are minor other issues as well, but if there is a lot of "business intelligence" (in my app abaout 10 manyears, of which 4 are "outdated") you can't beat a minimal conversion. Rewriting will be done, but incremental, since the app has a high change rate and is vital to the business.
>
>regards
>
>thomas