OSLib 6.11 Released

[Chris Rutter]

On Wed, 27 Sep 2000, Jonathan Coxhead wrote:

>    That seems very nice all round. Any downside?

Only that I think I prefer the merge-into-one-directory solution.

The namespace of unqualified headers is clean (i.e. there won't be files
`Computer' clashing with those in `Core'), so this is feasible.

Of course, even if the 77 file limit had never existed, would this
segregation be desireable?  Personally, I very rarely go looking at
the source tree through the filesystem, and when I do, it almost
always takes a bit of mindless hunting-down time to find the file
I'm after.  It would be simpler, if perhaps less elegant, I feel, to
have them all in a single directory.

The whole 77 limit is just an embarassment: raFS convincingly solves
the problem on those discs or operating systems where the problem
still remains, however: I feel it reasonable to tell the user `install
this on a proper filesystem', and let them sort out whether they wish
to use a soft-load FileCore, X-Files, raFS, NFS, or whatever -- that
isn't something OSLib should get involved in (other than maybe
suggesting where to find some example programs).  The people installing
OSLib should be capable, competent software engineers: people more than
capable of rigging up an NFS server, an raFS directory, or whatever -- if
they haven't done already, which I believe the overwhelming majority of
technically-savvy RISC OS users have.

I'm fairly strongly convinced that unless the `Core' &c. directories
feature in the #include statement, they shouldn't exist in the
filesystem either.  If there's a good reason (and not just expediency
of a geriatric filing system) to segregate them, then I believe they
probably /should/ feature in the #include directive, and that the
namespaces should be defined as discrete, so that portions of one
module belonging in one part of the library can be implement separately
from another.

The fact that this isn't especially necessary -- we have a non-clashing
namespace defined for us by RISC OS right from the start, anyway --
suggests to me that in fact the headers /should not/ be segregated.

c.