1. Forum
  2. Newsgroups
  3. comp.lang.c

  • seed

    From fir@fir@grunge.pl to comp.lang.c on Sat Mar 30 22:36:28 2024
    From Newsgroup: comp.lang.c

    i wonder if c should not have a seed kwyword ('type qualifier'? of
    how to call this

    i mean seed is such think you call reallock on ond only that


    char* p = realloc(p, 200); //p is a seed

    this is becouse you may write soem functions that expect seed

    foo(char* p( //seed is expected and only seed

    not normal pointer

    so this is probably needed and will increase typesafety
    --- Synchronet 3.20a-Linux NewsLink 1.114
  • From fir@fir@grunge.pl to comp.lang.c on Sun Mar 31 03:58:22 2024
    From Newsgroup: comp.lang.c

    fir wrote:
    i wonder if c should not have a seed kwyword ('type qualifier'? of
    how to call this

    i mean seed is such think you call reallock on ond only that


    char* p = realloc(p, 200); //p is a seed

    this is becouse you may write soem functions that expect seed

    foo(char* p( //seed is expected and only seed

    not normal pointer

    so this is probably needed and will increase typesafety

    i dont know how it is coz i dont remember that but probably those
    boolean should work after the bitewise and arithmetic - as bitewise and arithmetioc have not much sense ob boolean

    so it should probably be

    1,2 arithmetic and bitwise
    3 relational
    4 bollean

    as to first two probably bitwise should go first

    hovever for me as i say & is both bitwise and boolean (more like
    boolean) so its fortunate it goes late - but ! going early is a mistake imo

    its not what you sey "it is what it is and reason has nothink to do here"

    simply some of this operators produce result to be consumed by others
    and reversing the order has not much sense - boolean should go late
    --- Synchronet 3.20a-Linux NewsLink 1.114