🚀 go-pugleaf

In article <BsFG1G.EG8@fc.sde.hp.com> Gary Fritz <fritz@hpfcbig.sde.hp.com> writes:
>jim@n5ial.chi.il.us (Jim Graham) writes:
>: I finally ended up with a simple process where the filesize is determined,
>: a buffer large enough to hold the file is malloc'ed (in smaller chunks to
>: avoid one block of data > 64k...geez, you'd think the people writing C
>: compilers could handle that in the background and make a more correctly
>: behaved C compiler).
>
>Don't blame C, blame MSDOS and the underlying Intel hardware.

like I said, ``you'd think [they] could handle that in the background''
                                                    ^^^^^^^^^^^^^^^^^

I don't really care about dos's limitations (well, ok, I have to care....
but I don't want to have to care).  my point is, you'd think that a
commercial C compiler like tc++ would SILENTLY organize things so that
you could (as far as you're concerned, anyways) malloc a data structure
larger than 64k.

of course, there would be some inherent dangers involved (just like there
are with optimizing compilers), and this would need to be controlled by
a command line switch just like -O for optimizers.

oh well, then again, they forgot to include all of the standard C library
functions, so why should they do something like the above?  :-(

anyways, gotta run.
   --jim

\input std_disclaimer                     % 73 DE N5IAL (/9)
------------------------------------------------------------------------------
"[...] the plural of `mongoose' ought to be `polygoose'." [Jargon file 2.9.10]

INTERNET:  jim@n5ial.chi.il.us  |  grahj@gagme.chi.il.us  |  j.graham@ieee.org
UUCP:  gagme!n5ial!jim@clout.chi.il.us
AMATEUR RADIO:  n5ial@n9hsi (Chicago.IL.US.Earth)
------------------------------------------------------------------------------