That turned out to be the wrong paper. I'm looking for a paper that describes the (early) dialect of C that let you do stuff like this: struct w { char lo, hi; }; int x; char b = x.lo; I can't find my hardcopy so was looking for a pdf. ron On Mon, Apr 2, 2018 at 10:36 AM David du Colombier <0intro(a)gmail.com> wrote:

That reminds me of the multicharacter constant vs 'byte' index used in speak.c, I didn't realize this was an intended 'feature' for faking tuples (and allowing to index by either the first or second element) in early C, it seemed a bit of a hack to me. I was hoping there was some elegant way to achieve nearly the same thing in modern C, but I didn't find anything obvious short of string comparisons and an array with a byte and a pointer to a string. On 4/2/2018 6:53 PM, ron minnich wrote: That turned out to be the wrong paper. I'm looking for a paper that describes the (early) dialect of C that let you do stuff like this: struct w { char lo, hi; }; int x; char b = x.lo; I can't find my hardcopy so was looking for a pdf. ron On Mon, Apr 2, 2018 at 10:36 AM David du Colombier <0intro@gmail.com<mailto:0intro@gmail.com>> wrote: https://9p.io/cm/cs/who/dmr/cacm.html More generally, just replace "cm.bell-labs.com<http://cm.bell-labs.com>" with "9p.io<http://9p.io>". -- David du Colombier -- Jonathan Gevaryahu AKA Lord Nightmare jgevaryahu@gmail.com<mailto:jgevaryahu@gmail.com> jgevaryahu@hotmail.com<mailto:jgevaryahu@hotmail.com>