Re: [cc65] Common macros for dirent.d_type

> yet every program i remember that determines what files on a disk it can use
> for whatever, does that by looking at the filename (except geos maybe? :)).
> the filetype is always implied.

So what is the problem here? After all, the disk drives find the files without 
the type being specified in the OPEN, no matter the file type. So it's not a 
matter of open (not creating) at all.

So what about creating files...:

> well, as i said, its not so much about what is specified somewhere, but what
> ppl did since 1985. and as for that, "SEQ file" has even become a synonym
> for "petsci text, maybe with C/G graphics" in large parts of the commodore
> world. there is a reason for why eg jiffy dos even has a wedge command to
> display seq files as text - it was incredibly common at some point.

I don't know about "large part of the Commodore world". However, what are you 
talking about? BASIC programs that run on the CBM machines? Machine language 
programs that run on the CBM? Are we going to compile them with cc65? I guess 
not, so there is no compatibility issue here.

So that leaves "legacy" C programs compiled on/for the PET. Those compilers 
probably wouldn't have a default file type either, but the user defining the file 
type with the file name extension (like adding ",U").

Just please tell me the name of some "legacy" C-compiled programs for the CBM 
that would not compile on cc65 due to that file type? It probably won't compile 
due to many other incompatibilities of the old C compilers. They would have to 
be adopted anyway.

So in my not so humble opinion the C compiler should not add/use a default file 
type at all. Let the floppy handle it. And if someone needs special file types, 
the file name extension is the way - so there is not even a way to set a 
special global variable just to change the default file type.

André

----------------------------------------------------------------------
To unsubscribe from the list send mail to majordomo@musoftware.de with
the string "unsubscribe cc65" in the body(!) of the mail.