silverdr_at_inet.com.pl
Date: 2008-01-07 11:45:52
On 2008-01-07, at 11:06, <ruud.baltissen@abp.nl>
<ruud.baltissen@abp.nl> wrote:
>
> [...]
>
> But I cannot use this try-and-error method all the time because 1) it
> eats up valuable time
Yes - it does :-(
> and 2) I end up with a routine that I don't
> understand myself.
Don't worry - that's the way the whole modern software industry
works ;-)
> Worse, if things go wrong, where should I look for
> ???
Ooops! I forgot we are supposed to be old-fashioned rather than modern.
>
> If nobody has any sample sources, it just means that the development
> takes longer then I hoped.
I still many sources from the things I did but would have the same
problem - would have to go completely through the DOS routines and
analyse them step by step as all the things I did with the 1541 were
either completely high or completely low-level. Nothing in-between
like you need now. I presume most people have similar experiences as
such things were simply things that very few people needed/did before.
[...]
>> It'll never be anything else, no?
>
> Like a normal directory, it should be expandable. Hmmm, what about
> directories that once contained hundreds of files and now are empty?
> Another challenge I think :)
I think on the harddrive it is not that much of a problem. You have to
allocate (and expand) as much as is needed to handle all the entries
but then when you delete the files you actually mark them as deleted
(as CBM DOS does) and you don't absolutely have to free the blocks.
Yes, it's sub-optimal but it's IMHO good enough at least for the
beginning.
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.