I suggest you ...

allow user-supplied function semantics to override function inputs

For functions which the user has supplied function-wide semantics, I would like to see a flag/mode that allows the provided semantics to essentially "override" the inputs to the function.

For example, if I provide a "_n_p" semantic for a function, I don't want to see a message for possible use of NULL pointer generated within the function; I have already provided a semantic that the parameter is not NULL. Similarly with integral parameters, if I provide a semantic that "1n < 2" and then use that parameter to index into a 2 element array, I don't want to see "possible index out of range" by a billion because the input is considered "unknown".

This is *clearly* predicated on the semantics already being verified as called, as they are through messages 418, 419, 422, 426 and the like.

1 vote
Vote
Sign in
Check!
(thinking…)
Reset
or sign in with
  • facebook
  • google
    Password icon
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    Michael Metivier shared this idea  ·   ·  Admin →

    0 comments

    Sign in
    Check!
    (thinking…)
    Reset
    or sign in with
    • facebook
    • google
      Password icon
      Signed in as (Sign out)
      Submitting...

      Feedback and Knowledge Base