Frama-C-discuss mailing list archives

This page gathers the archives of the old Frama-C-discuss archives, that was hosted by Inria's gforge before its demise at the end of 2020. To search for mails newer than September 2020, please visit the page of the new mailing list on Renater.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Frama-c-discuss] Issue to prove a precondition for user call

  • Subject: [Frama-c-discuss] Issue to prove a precondition for user call
  • From: dmentre at linux-france.org (David MENTRE)
  • Date: Tue, 24 Mar 2009 11:27:45 +0100
Hello,

I have following (simplified) code:
===
#define _GNU_SOURCE

#include <stdio.h>
#include <stdlib.h>

#define MAX_STRING_SIZE 255
#define MAX_VAL 10

int num;

/*@ requires \valid(stdin);
    assigns \nothing;
    ensures \result >= 0 && \result < num;
 */
int get(void)
{
  char buf[MAX_STRING_SIZE];
  char *s = NULL;

  while (1) {
    s = fgets(buf, MAX_STRING_SIZE, stdin);

    if (s != NULL) {
      int choice;

      choice = atoi(s);
      if (choice >= 0 && choice < num) {
        s = fgets(buf, MAX_STRING_SIZE, stdin);
        if (s != NULL && strncmp(s, "y", 1) == 0) {
          return choice;
        } else {
        }
      } else {
      }
    } else {
    }
  }
}

/*@ requires \valid(stdin);
 */
int main(void)
{
  int a;

  num = MAX_VAL - 1;
  a = get();

  return a;
}
====

I'm launching Jessie with:
  "frama-c -jessie-analysis -jessie-std-stubs -jessie-atp alt-ergo
-jessie-cpu-limit 2 -jessie-gui fgets-issue.c"

I am failing to prove "get_ensures_default_po_7" which is
"valid_string(s_0_0_0, result1, result0)". I am puzzled by this proof
obligation.

Could somebody explain to me what is exactly needed? What are the
three arguments to valid_string?

I looked into Jessie prolog but found only "#define FRAMA_C_STRING
__declspec(valid_string)" which is not very verbose. :-)

Many thanks in advance for any help,
Sincerely yours,
david