Thu 2003-02-13 09:52, mark lewis (1:3634/12) wrote to andrew clarke:

 ac> Binary config files are another matter, and really depends on
 ac> what people still want support for,

ml> i don't see where they are another matter... the code should work as
ml> it was unless there have been updates to those binary formats...
ml> they should work straight out just as they did before... hummm...
ml> unless there's some sort of byte alignment problem...

It may or may not work, but chances are it won't due to byte alignment
problems (among other things), in which case timEd will probably just
segfault, and I won't know why unless I can duplicate the problem, which
means having people sending me their entire config files, which is
presumably not something they will want to do, and not really something I
want to spend a lot of time on.

 ac> as I don't intend to re-add code for all the obsolete config
 ac> file formats that timEd 1.10 supported,

ml> re-add? why chop them out in the first place?

Well, there's no point compiling code that presumably nobody is ever going
to use, eg. code for FastEcho 1.41 config files.  There's was also support
for something called xMail, which I've never heard of.

 ac> particularly since I have no way of testing those.

ml> that's why other folk download, run and report back... we/developers
ml> don't always have access to everything we write for... there are
ml> times that we have to write "blind" based only on what info we can
ml> gather and sample data file and let others report the situation... i
ml> know, its hard to do but if i, a lowly self-taught pascal coder can
ml> do it, i _know_ that others can do it too > 

Others can do it.  I have better things to do.

 ac> So, if something is simple to add, and test, I will probably
 ac> do it.  In the mean time if you can export your binary config
 ac> file to AREAS.BBS or SQUISH.CFG formats then do that.

ml> exporting to those formats looses some of the benefits of reading
ml> the binary directly... some necessary options cannot be expressed in
ml> those formats... 

The point was it's better than nothing.

-- mail{at}ozzmosis.com

--- timEd/Cygwin 1.11.a5