Re: Code guidelines

On 03/09/2024 13:23, Thiago Adams wrote:
...

 I will give a sample
 In my code I have
 if (obj->member1 &&
     obj->member1->member2 &&
     obj->member1->member2->member3)
{
 }
 The contract is
   * obj->member1                   CAN be null
  * obj->member1->member2          CANNOT be null
  * obj->member1->member2->member3 CAN be null
 So I can write just
 if (obj->member1 &&
     obj->member1->member2->member3)
{
 }
  but...maybe, is better to be a little redundant here?
I think I prefer to leave "obj->member1->member2 && " even if I know
it should not be null.
 if (obj->member1 &&
     obj->member1->member2 &&
     obj->member1->member2->member3)
{
 }
 

I also have a interesting sample with linked list
struct item {
      char* title;
      struct item* next;  //can be null
};
struct list {
     struct item* head; //can be null
     struct item* tail; //can be null
};
void list_push_back(struct list* list, struct item* p_item)
{
    if (list->head == NULL) {
       list->head = p_item;
       //tail also should be null
    }
    else {
       //When the head is null tail is also null
       //if head is not null then tail is not null
       //So tail is not null here.. assert works as assume
       assert(list->tail != nullptr);
       //next can be null or not null..however for the tail node
       //next IS/MUST BE null
       assert(list->tail->next == nullptr);
       list->tail->next = p_item;
    }
    list->tail = p_item;
}
assert in this case works as "assume" and "check".
For instance,
We assume  "assert(list->tail != nullptr);" because we access it directly here
list->tail->next = p_item;
We also assume "assert(list->tail->next == nullptr)" otherwise we could have a memory leak overriding next.