[JITLink][LoongArch] Support R_LARCH_ALIGN relaxation (#122259)

[llvm-project.git] / clang-tools-extra / docs / clang-tidy / checks / bugprone / unchecked-optional-access.rst

.. title:: clang-tidy - bugprone-unchecked-optional-access

bugprone-unchecked-optional-access
==================================

*Note*: This check uses a flow-sensitive static analysis to produce its
results. Therefore, it may be more resource intensive (RAM, CPU) than the
average clang-tidy check.

This check identifies unsafe accesses to values contained in
``std::optional<T>``, ``absl::optional<T>``, ``base::Optional<T>``,
``folly::Optional<T>``, ``bsl::optional``, or
``BloombergLP::bdlb::NullableValue`` objects. Below we will refer to all these
types collectively as ``optional<T>``.

An access to the value of an ``optional<T>`` occurs when one of its ``value``,
``operator*``, or ``operator->`` member functions is invoked.  To align with
common misconceptions, the check considers these member functions as equivalent,
even though there are subtle differences related to exceptions versus undefined
behavior. See *Additional notes*, below, for more information on this topic.

An access to the value of an ``optional<T>`` is considered safe if and only if
code in the local scope (for example, a function body) ensures that the
``optional<T>`` has a value in all possible execution paths that can reach the
access. That should happen either through an explicit check, using the
``optional<T>::has_value`` member function, or by constructing the
``optional<T>`` in a way that shows that it unambiguously holds a value (e.g
using ``std::make_optional`` which always returns a populated
``std::optional<T>``).

Below we list some examples, starting with unsafe optional access patterns,
followed by safe access patterns.

Unsafe access patterns
~~~~~~~~~~~~~~~~~~~~~~

Access the value without checking if it exists
----------------------------------------------

The check flags accesses to the value that are not locally guarded by
existence check:

.. code-block:: c++

   void f(std::optional<int> opt) {
     use(*opt); // unsafe: it is unclear whether `opt` has a value.
   }

Access the value in the wrong branch
------------------------------------

The check is aware of the state of an optional object in different
branches of the code. For example:

.. code-block:: c++

   void f(std::optional<int> opt) {
     if (opt.has_value()) {
     } else {
       use(opt.value()); // unsafe: it is clear that `opt` does *not* have a value.
     }
   }

Assume a function result to be stable
-------------------------------------

The check is aware that function results might not be stable. That is,
consecutive calls to the same function might return different values.
For example:

.. code-block:: c++

   void f(Foo foo) {
     if (foo.opt().has_value()) {
       use(*foo.opt()); // unsafe: it is unclear whether `foo.opt()` has a value.
     }
   }

Exception: accessor methods
```````````````````````````

The check assumes *accessor* methods of a class are stable, with a heuristic to
determine which methods are accessors. Specifically, parameter-free ``const``
methods are treated as accessors. Note that this is not guaranteed to be safe
-- but, it is widely used (safely) in practice, and so we have chosen to treat
it as generally safe. Calls to non ``const`` methods are assumed to modify
the state of the object and affect the stability of earlier accessor calls.

Rely on invariants of uncommon APIs
-----------------------------------

The check is unaware of invariants of uncommon APIs. For example:

.. code-block:: c++

   void f(Foo foo) {
     if (foo.HasProperty("bar")) {
       use(*foo.GetProperty("bar")); // unsafe: it is unclear whether `foo.GetProperty("bar")` has a value.
     }
   }

Check if a value exists, then pass the optional to another function
-------------------------------------------------------------------

The check relies on local reasoning. The check and value access must
both happen in the same function. An access is considered unsafe even if
the caller of the function performing the access ensures that the
optional has a value. For example:

.. code-block:: c++

   void g(std::optional<int> opt) {
     use(*opt); // unsafe: it is unclear whether `opt` has a value.
   }

   void f(std::optional<int> opt) {
     if (opt.has_value()) {
       g(opt);
     }
   }

Safe access patterns
~~~~~~~~~~~~~~~~~~~~

Check if a value exists, then access the value
----------------------------------------------

The check recognizes all straightforward ways for checking if a value
exists and accessing the value contained in an optional object. For
example:

.. code-block:: c++

   void f(std::optional<int> opt) {
     if (opt.has_value()) {
       use(*opt);
     }
   }


Check if a value exists, then access the value from a copy
----------------------------------------------------------

The criteria that the check uses is semantic, not syntactic. It
recognizes when a copy of the optional object being accessed is known to
have a value. For example:

.. code-block:: c++

   void f(std::optional<int> opt1) {
     if (opt1.has_value()) {
       std::optional<int> opt2 = opt1;
       use(*opt2);
     }
   }


Ensure that a value exists using common macros
----------------------------------------------

The check is aware of common macros like ``CHECK`` and ``DCHECK``. Those can be
used to ensure that an optional object has a value. For example:

.. code-block:: c++

   void f(std::optional<int> opt) {
     DCHECK(opt.has_value());
     use(*opt);
   }

Ensure that a value exists, then access the value in a correlated branch
------------------------------------------------------------------------

The check is aware of correlated branches in the code and can figure out
when an optional object is ensured to have a value on all execution
paths that lead to an access. For example:

.. code-block:: c++

   void f(std::optional<int> opt) {
     bool safe = false;
     if (opt.has_value() && SomeOtherCondition()) {
       safe = true;
     }
     // ... more code...
     if (safe) {
       use(*opt);
     }
   }

Stabilize function results
~~~~~~~~~~~~~~~~~~~~~~~~~~

Since function results are not assumed to be stable across calls, it is best to
store the result of the function call in a local variable and use that variable
to access the value. For example:

.. code-block:: c++

   void f(Foo foo) {
     if (const auto& foo_opt = foo.opt(); foo_opt.has_value()) {
       use(*foo_opt);
     }
   }

Do not rely on uncommon-API invariants
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

When uncommon APIs guarantee that an optional has contents, do not rely on it --
instead, check explicitly that the optional object has a value. For example:

.. code-block:: c++

   void f(Foo foo) {
     if (const auto& property = foo.GetProperty("bar")) {
       use(*property);
     }
   }

instead of the `HasProperty`, `GetProperty` pairing we saw above.

Do not rely on caller-performed checks
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

If you know that all of a function's callers have checked that an optional
argument has a value, either change the function to take the value directly or
check the optional again in the local scope of the callee. For example:

.. code-block:: c++

   void g(int val) {
     use(val);
   }

   void f(std::optional<int> opt) {
     if (opt.has_value()) {
       g(*opt);
     }
   }

and

.. code-block:: c++

   struct S {
     std::optional<int> opt;
     int x;
   };

   void g(const S &s) {
     if (s.opt.has_value() && s.x > 10) {
       use(*s.opt);
   }

   void f(S s) {
     if (s.opt.has_value()) {
       g(s);
     }
   }

Additional notes
~~~~~~~~~~~~~~~~

Aliases created via ``using`` declarations
------------------------------------------

The check is aware of aliases of optional types that are created via
``using`` declarations. For example:

.. code-block:: c++

   using OptionalInt = std::optional<int>;

   void f(OptionalInt opt) {
     use(opt.value()); // unsafe: it is unclear whether `opt` has a value.
   }

Lambdas
-------

The check does not currently report unsafe optional accesses in lambdas.
A future version will expand the scope to lambdas, following the rules
outlined above. It is best to follow the same principles when using
optionals in lambdas.

Access with ``operator*()`` vs. ``value()``
-------------------------------------------

Given that ``value()`` has well-defined behavior (either throwing an exception
or terminating the program), why treat it the same as ``operator*()`` which
causes undefined behavior (UB)? That is, why is it considered unsafe to access
an optional with ``value()``, if it's not provably populated with a value?  For
that matter, why is ``CHECK()`` followed by ``operator*()`` any better than
``value()``, given that they are semantically equivalent (on configurations that
disable exceptions)?

The answer is that we assume most users do not realize the difference between
``value()`` and ``operator*()``. Shifting to ``operator*()`` and some form of
explicit value-presence check or explicit program termination has two
advantages:

  * Readability. The check, and any potential side effects like program
    shutdown, are very clear in the code. Separating access from checks can
    actually make the checks more obvious.

  * Performance. A single check can cover many or even all accesses within
    scope. This gives the user the best of both worlds -- the safety of a
    dynamic check, but without incurring redundant costs.