In response to Mike.
I got your entoke.exe file, but it on a Win10 computer and it only works on
32-bit computers. Have a Win95 computer here, so maybe I'll give it a go on
it. If Win95 isn't 16-bit....
Looking forward to Bob's version!

 Curtis

On Thu, Oct 21, 2021 at 1:18 PM John R. Hogerhuis <[email protected]> wrote:

> The closest thing to a perfect solution... that just works... is for the
> disk service to present the proper extension instead of the incorrect one.
> That's what Laddiealpha does. It peeks at beginning of the file to see if
> it's a binary file and automatically presents a correct extension to TSDOS.
> Whether the user knows how to load that file is a separate concern. But I
> figure people will simply ask on list.
>
> Other ways to go:
>
> Do not present the file at all in the directory if a mismatch is detected.
> The user will see it as a mystery but they can read their documentation or
> ask on list and the mystery will be resolved. Another idea would be to
> rename it in some way to be clearly bad like "FILE.><" So it's not just
> missing.
>
> Present the filename as is. But when at the protocol level the file is
> opened or on the first read then an error is returned based on inspection
> of the file. TSDOS or whatever will report the error and the user may
> figure it out or they will ask on list. I haven't explored this but
> theoretically it should be feasible.
>
> Embed a tokenizer in the file server to automatically make the content
> match the extension. This won't always work. And it would have to be laptop
> model aware since 100 vs NEC tokens.
>
> -- John.
>
>

Reply via email to