blob: 7c96256c72e7ffc0bafc3e24ff4b5541875bf3ad [file] [log] [blame]
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
<title>Development</title>
<link rel="stylesheet" href="../boostbook.css" type="text/css">
<meta name="generator" content="DocBook XSL Stylesheets V1.75.2">
<link rel="home" href="../index.html" title="Chapter 1. Boost.Optional">
<link rel="up" href="../index.html" title="Chapter 1. Boost.Optional">
<link rel="prev" href="../index.html" title="Chapter 1. Boost.Optional">
<link rel="next" href="synopsis.html" title="Synopsis">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
<table cellpadding="2" width="100%"><tr>
<td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../../../../boost.png"></td>
<td align="center"><a href="../../../../../index.html">Home</a></td>
<td align="center"><a href="../../../../libraries.htm">Libraries</a></td>
<td align="center"><a href="http://www.boost.org/users/people.html">People</a></td>
<td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td>
<td align="center"><a href="../../../../../more/index.htm">More</a></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="../index.html"><img src="../../../../../doc/html/images/prev.png" alt="Prev"></a><a accesskey="u" href="../index.html"><img src="../../../../../doc/html/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../../../doc/html/images/home.png" alt="Home"></a><a accesskey="n" href="synopsis.html"><img src="../../../../../doc/html/images/next.png" alt="Next"></a>
</div>
<div class="section" title="Development">
<div class="titlepage"><div><div><h2 class="title" style="clear: both">
<a name="boost_optional.development"></a><a class="link" href="development.html" title="Development">Development</a>
</h2></div></div></div>
<div class="toc"><dl>
<dt><span class="section"><a href="development.html#boost_optional.development.the_models">The models</a></span></dt>
<dt><span class="section"><a href="development.html#boost_optional.development.the_semantics">The semantics</a></span></dt>
<dt><span class="section"><a href="development.html#boost_optional.development.the_interface">The Interface</a></span></dt>
</dl></div>
<div class="section" title="The models">
<div class="titlepage"><div><div><h3 class="title">
<a name="boost_optional.development.the_models"></a><a class="link" href="development.html#boost_optional.development.the_models" title="The models">The models</a>
</h3></div></div></div>
<p>
In C++, we can <span class="emphasis"><em>declare</em></span> an object (a variable) of type
<code class="computeroutput"><span class="identifier">T</span></code>, and we can give this variable
an <span class="emphasis"><em>initial value</em></span> (through an <span class="emphasis"><em>initializer</em></span>.
(c.f. 8.5)). When a declaration includes a non-empty initializer (an initial
value is given), it is said that the object has been initialized. If the
declaration uses an empty initializer (no initial value is given), and neither
default nor value initialization applies, it is said that the object is
<span class="bold"><strong>uninitialized</strong></span>. Its actual value exist but
has an <span class="emphasis"><em>indeterminate initial value</em></span> (c.f. 8.5.9). <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code> intends
to formalize the notion of initialization (or lack of it) allowing a program
to test whether an object has been initialized and stating that access to
the value of an uninitialized object is undefined behavior. That is, when
a variable is declared as <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
and no initial value is given, the variable is <span class="emphasis"><em>formally</em></span>
uninitialized. A formally uninitialized optional object has conceptually
no value at all and this situation can be tested at runtime. It is formally
<span class="emphasis"><em>undefined behavior</em></span> to try to access the value of an
uninitialized optional. An uninitialized optional can be assigned a value,
in which case its initialization state changes to initialized. Furthermore,
given the formal treatment of initialization states in optional objects,
it is even possible to reset an optional to <span class="emphasis"><em>uninitialized</em></span>.
</p>
<p>
In C++ there is no formal notion of uninitialized objects, which means that
objects always have an initial value even if indeterminate. As discussed
on the previous section, this has a drawback because you need additional
information to tell if an object has been effectively initialized. One of
the typical ways in which this has been historically dealt with is via a
special value: <code class="computeroutput"><span class="identifier">EOF</span></code>, <code class="computeroutput"><span class="identifier">npos</span></code>, -1, etc... This is equivalent to
adding the special value to the set of possible values of a given type. This
super set of <code class="computeroutput"><span class="identifier">T</span></code> plus some
<span class="emphasis"><em>nil_t</em></span>&#8212;were <code class="computeroutput"><span class="identifier">nil_t</span></code>
is some stateless POD-can be modeled in modern languages as a <span class="bold"><strong>discriminated
union</strong></span> of T and nil_t. Discriminated unions are often called <span class="emphasis"><em>variants</em></span>.
A variant has a <span class="emphasis"><em>current type</em></span>, which in our case is either
<code class="computeroutput"><span class="identifier">T</span></code> or <code class="computeroutput"><span class="identifier">nil_t</span></code>.
Using the <a href="../../../../variant/index.html" target="_top">Boost.Variant</a>
library, this model can be implemented in terms of <code class="computeroutput"><span class="identifier">boost</span><span class="special">::</span><span class="identifier">variant</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">,</span><span class="identifier">nil_t</span><span class="special">&gt;</span></code>.
There is precedent for a discriminated union as a model for an optional value:
the <a href="http://www.haskell.org/" target="_top">Haskell</a> <span class="bold"><strong>Maybe</strong></span>
built-in type constructor. Thus, a discriminated union <code class="computeroutput"><span class="identifier">T</span><span class="special">+</span><span class="identifier">nil_t</span></code>
serves as a conceptual foundation.
</p>
<p>
A <code class="computeroutput"><span class="identifier">variant</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">,</span><span class="identifier">nil_t</span><span class="special">&gt;</span></code> follows naturally from the traditional
idiom of extending the range of possible values adding an additional sentinel
value with the special meaning of <span class="emphasis"><em>Nothing</em></span>. However,
this additional <span class="emphasis"><em>Nothing</em></span> value is largely irrelevant
for our purpose since our goal is to formalize the notion of uninitialized
objects and, while a special extended value can be used to convey that meaning,
it is not strictly necessary in order to do so.
</p>
<p>
The observation made in the last paragraph about the irrelevant nature of
the additional <code class="computeroutput"><span class="identifier">nil_t</span></code> with
respect to <span class="underline">purpose</span> of <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code> suggests
an alternative model: a <span class="emphasis"><em>container</em></span> that either has a
value of <code class="computeroutput"><span class="identifier">T</span></code> or nothing.
</p>
<p>
As of this writing I don't know of any precedence for a variable-size fixed-capacity
(of 1) stack-based container model for optional values, yet I believe this
is the consequence of the lack of practical implementations of such a container
rather than an inherent shortcoming of the container model.
</p>
<p>
In any event, both the discriminated-union or the single-element container
models serve as a conceptual ground for a class representing optional&#8212;i.e.
possibly uninitialized&#8212;objects. For instance, these models show the <span class="emphasis"><em>exact</em></span>
semantics required for a wrapper of optional values:
</p>
<p>
Discriminated-union:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
<span class="bold"><strong>deep-copy</strong></span> semantics: copies of the variant
implies copies of the value.
</li>
<li class="listitem">
<span class="bold"><strong>deep-relational</strong></span> semantics: comparisons
between variants matches both current types and values
</li>
<li class="listitem">
If the variant's current type is <code class="computeroutput"><span class="identifier">T</span></code>,
it is modeling an <span class="emphasis"><em>initialized</em></span> optional.
</li>
<li class="listitem">
If the variant's current type is not <code class="computeroutput"><span class="identifier">T</span></code>,
it is modeling an <span class="emphasis"><em>uninitialized</em></span> optional.
</li>
<li class="listitem">
Testing if the variant's current type is <code class="computeroutput"><span class="identifier">T</span></code>
models testing if the optional is initialized
</li>
<li class="listitem">
Trying to extract a <code class="computeroutput"><span class="identifier">T</span></code> from
a variant when its current type is not <code class="computeroutput"><span class="identifier">T</span></code>,
models the undefined behavior of trying to access the value of an uninitialized
optional
</li>
</ul></div>
<p>
Single-element container:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
<span class="bold"><strong>deep-copy</strong></span> semantics: copies of the container
implies copies of the value.
</li>
<li class="listitem">
<span class="bold"><strong>deep-relational</strong></span> semantics: comparisons
between containers compare container size and if match, contained value
</li>
<li class="listitem">
If the container is not empty (contains an object of type <code class="computeroutput"><span class="identifier">T</span></code>), it is modeling an <span class="emphasis"><em>initialized</em></span>
optional.
</li>
<li class="listitem">
If the container is empty, it is modeling an <span class="emphasis"><em>uninitialized</em></span>
optional.
</li>
<li class="listitem">
Testing if the container is empty models testing if the optional is initialized
</li>
<li class="listitem">
Trying to extract a <code class="computeroutput"><span class="identifier">T</span></code> from
an empty container models the undefined behavior of trying to access the
value of an uninitialized optional
</li>
</ul></div>
</div>
<div class="section" title="The semantics">
<div class="titlepage"><div><div><h3 class="title">
<a name="boost_optional.development.the_semantics"></a><a class="link" href="development.html#boost_optional.development.the_semantics" title="The semantics">The semantics</a>
</h3></div></div></div>
<p>
Objects of type <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
are intended to be used in places where objects of type <code class="computeroutput"><span class="identifier">T</span></code>
would but which might be uninitialized. Hence, <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>'s
purpose is to formalize the additional possibly uninitialized state. From
the perspective of this role, <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
can have the same operational semantics of <code class="computeroutput"><span class="identifier">T</span></code>
plus the additional semantics corresponding to this special state. As such,
<code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code> could
be thought of as a <span class="emphasis"><em>supertype</em></span> of <code class="computeroutput"><span class="identifier">T</span></code>.
Of course, we can't do that in C++, so we need to compose the desired semantics
using a different mechanism. Doing it the other way around, that is, making
<code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code> a
<span class="emphasis"><em>subtype</em></span> of <code class="computeroutput"><span class="identifier">T</span></code>
is not only conceptually wrong but also impractical: it is not allowed to
derive from a non-class type, such as a built-in type.
</p>
<p>
We can draw from the purpose of <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
the required basic semantics:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
<span class="bold"><strong>Default Construction:</strong></span> To introduce a formally
uninitialized wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Direct Value Construction via copy:</strong></span> To
introduce a formally initialized wrapped object whose value is obtained
as a copy of some object.
</li>
<li class="listitem">
<span class="bold"><strong>Deep Copy Construction:</strong></span> To obtain a new
yet equivalent wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Direct Value Assignment (upon initialized):</strong></span>
To assign a value to the wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Direct Value Assignment (upon uninitialized):</strong></span>
To initialize the wrapped object with a value obtained as a copy of some
object.
</li>
<li class="listitem">
<span class="bold"><strong>Assignment (upon initialized):</strong></span> To assign
to the wrapped object the value of another wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Assignment (upon uninitialized):</strong></span> To initialize
the wrapped object with value of another wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Deep Relational Operations (when supported by the
type T):</strong></span> To compare wrapped object values taking into account
the presence of uninitialized states.
</li>
<li class="listitem">
<span class="bold"><strong>Value access:</strong></span> To unwrap the wrapped object.
</li>
<li class="listitem">
<span class="bold"><strong>Initialization state query:</strong></span> To determine
if the object is formally initialized or not.
</li>
<li class="listitem">
<span class="bold"><strong>Swap:</strong></span> To exchange wrapped objects. (with
whatever exception safety guarantees are provided by <code class="computeroutput"><span class="identifier">T</span></code>'s
swap).
</li>
<li class="listitem">
<span class="bold"><strong>De-initialization:</strong></span> To release the wrapped
object (if any) and leave the wrapper in the uninitialized state.
</li>
</ul></div>
<p>
Additional operations are useful, such as converting constructors and converting
assignments, in-place construction and assignment, and safe value access
via a pointer to the wrapped object or null.
</p>
</div>
<div class="section" title="The Interface">
<div class="titlepage"><div><div><h3 class="title">
<a name="boost_optional.development.the_interface"></a><a class="link" href="development.html#boost_optional.development.the_interface" title="The Interface">The Interface</a>
</h3></div></div></div>
<p>
Since the purpose of optional is to allow us to use objects with a formal
uninitialized additional state, the interface could try to follow the interface
of the underlying <code class="computeroutput"><span class="identifier">T</span></code> type
as much as possible. In order to choose the proper degree of adoption of
the native <code class="computeroutput"><span class="identifier">T</span></code> interface, the
following must be noted: Even if all the operations supported by an instance
of type <code class="computeroutput"><span class="identifier">T</span></code> are defined for
the entire range of values for such a type, an <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
extends such a set of values with a new value for which most (otherwise valid)
operations are not defined in terms of <code class="computeroutput"><span class="identifier">T</span></code>.
</p>
<p>
Furthermore, since <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
itself is merely a <code class="computeroutput"><span class="identifier">T</span></code> wrapper
(modeling a <code class="computeroutput"><span class="identifier">T</span></code> supertype),
any attempt to define such operations upon uninitialized optionals will be
totally artificial w.r.t. <code class="computeroutput"><span class="identifier">T</span></code>.
</p>
<p>
This library chooses an interface which follows from <code class="computeroutput"><span class="identifier">T</span></code>'s
interface only for those operations which are well defined (w.r.t the type
<code class="computeroutput"><span class="identifier">T</span></code>) even if any of the operands
are uninitialized. These operations include: construction, copy-construction,
assignment, swap and relational operations.
</p>
<p>
For the value access operations, which are undefined (w.r.t the type <code class="computeroutput"><span class="identifier">T</span></code>) when the operand is uninitialized, a
different interface is chosen (which will be explained next).
</p>
<p>
Also, the presence of the possibly uninitialized state requires additional
operations not provided by <code class="computeroutput"><span class="identifier">T</span></code>
itself which are supported by a special interface.
</p>
<a name="boost_optional.development.the_interface.lexically_hinted_value_access_in_the_presence_of_possibly_untitialized_optional_objects__the_operators___and___gt_"></a><h5>
<a name="id555569"></a>
<a class="link" href="development.html#boost_optional.development.the_interface.lexically_hinted_value_access_in_the_presence_of_possibly_untitialized_optional_objects__the_operators___and___gt_">Lexically-hinted
Value Access in the presence of possibly untitialized optional objects: The
operators * and -&gt;</a>
</h5>
<p>
A relevant feature of a pointer is that it can have a <span class="bold"><strong>null
pointer value</strong></span>. This is a <span class="emphasis"><em>special</em></span> value which
is used to indicate that the pointer is not referring to any object at all.
In other words, null pointer values convey the notion of inexistent objects.
</p>
<p>
This meaning of the null pointer value allowed pointers to became a <span class="emphasis"><em>de
facto</em></span> standard for handling optional objects because all you have
to do to refer to a value which you don't really have is to use a null pointer
value of the appropriate type. Pointers have been used for decades&#8212;from
the days of C APIs to modern C++ libraries&#8212;to <span class="emphasis"><em>refer</em></span>
to optional (that is, possibly inexistent) objects; particularly as optional
arguments to a function, but also quite often as optional data members.
</p>
<p>
The possible presence of a null pointer value makes the operations that access
the pointee's value possibly undefined, therefore, expressions which use
dereference and access operators, such as: <code class="computeroutput"><span class="special">(</span>
<span class="special">*</span><span class="identifier">p</span> <span class="special">=</span> <span class="number">2</span> <span class="special">)</span></code>
and <code class="computeroutput"><span class="special">(</span> <span class="identifier">p</span><span class="special">-&gt;</span><span class="identifier">foo</span><span class="special">()</span> <span class="special">)</span></code>, implicitly
convey the notion of optionality, and this information is tied to the <span class="emphasis"><em>syntax</em></span>
of the expressions. That is, the presence of operators <code class="computeroutput"><span class="special">*</span></code>
and <code class="computeroutput"><span class="special">-&gt;</span></code> tell by themselves
&#8212;without any additional context&#8212; that the expression will be undefined
unless the implied pointee actually exist.
</p>
<p>
Such a <span class="emphasis"><em>de facto</em></span> idiom for referring to optional objects
can be formalized in the form of a concept: the <a href="../../../../utility/OptionalPointee.html" target="_top">OptionalPointee</a>
concept. This concept captures the syntactic usage of operators <code class="computeroutput"><span class="special">*</span></code>, <code class="computeroutput"><span class="special">-&gt;</span></code>
and conversion to <code class="computeroutput"><span class="keyword">bool</span></code> to convey
the notion of optionality.
</p>
<p>
However, pointers are good to <span class="underline">refer</span>
to optional objects, but not particularly good to handle the optional objects
in all other respects, such as initializing or moving/copying them. The problem
resides in the shallow-copy of pointer semantics: if you need to effectively
move or copy the object, pointers alone are not enough. The problem is that
copies of pointers do not imply copies of pointees. For example, as was discussed
in the motivation, pointers alone cannot be used to return optional objects
from a function because the object must move outside from the function and
into the caller's context.
</p>
<p>
A solution to the shallow-copy problem that is often used is to resort to
dynamic allocation and use a smart pointer to automatically handle the details
of this. For example, if a function is to optionally return an object <code class="computeroutput"><span class="identifier">X</span></code>, it can use <code class="computeroutput"><span class="identifier">shared_ptr</span><span class="special">&lt;</span><span class="identifier">X</span><span class="special">&gt;</span></code>
as the return value. However, this requires dynamic allocation of <code class="computeroutput"><span class="identifier">X</span></code>. If <code class="computeroutput"><span class="identifier">X</span></code>
is a built-in or small POD, this technique is very poor in terms of required
resources. Optional objects are essentially values so it is very convenient
to be able to use automatic storage and deep-copy semantics to manipulate
optional values just as we do with ordinary values. Pointers do not have
this semantics, so are inappropriate for the initialization and transport
of optional values, yet are quite convenient for handling the access to the
possible undefined value because of the idiomatic aid present in the <a href="../../../../utility/OptionalPointee.html" target="_top">OptionalPointee</a> concept
incarnated by pointers.
</p>
<a name="boost_optional.development.the_interface.optional_lt_t_gt__as_a_model_of_optionalpointee"></a><h5>
<a name="id555831"></a>
<a class="link" href="development.html#boost_optional.development.the_interface.optional_lt_t_gt__as_a_model_of_optionalpointee">Optional&lt;T&gt;
as a model of OptionalPointee</a>
</h5>
<p>
For value access operations <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;&gt;</span></code> uses operators <code class="computeroutput"><span class="special">*</span></code>
and <code class="computeroutput"><span class="special">-&gt;</span></code> to lexically warn
about the possibly uninitialized state appealing to the familiar pointer
semantics w.r.t. to null pointers.
</p>
<div class="warning" title="Warning"><table border="0" summary="Warning">
<tr>
<td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="../../../../../doc/html/images/warning.png"></td>
<th align="left">Warning</th>
</tr>
<tr><td align="left" valign="top"><p>
However, it is particularly important to note that <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;&gt;</span></code> objects are not pointers. <span class="underline"><code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;&gt;</span></code> is not, and does not model, a pointer</span>.
</p></td></tr>
</table></div>
<p>
For instance, <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;&gt;</span></code>
does not have shallow-copy so does not alias: two different optionals never
refer to the <span class="emphasis"><em>same</em></span> value unless <code class="computeroutput"><span class="identifier">T</span></code>
itself is a reference (but may have <span class="emphasis"><em>equivalent</em></span> values).
The difference between an <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
and a pointer must be kept in mind, particularly because the semantics of
relational operators are different: since <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
is a value-wrapper, relational operators are deep: they compare optional
values; but relational operators for pointers are shallow: they do not compare
pointee values. As a result, you might be able to replace <code class="computeroutput"><span class="identifier">optional</span><span class="special">&lt;</span><span class="identifier">T</span><span class="special">&gt;</span></code>
by <code class="computeroutput"><span class="identifier">T</span><span class="special">*</span></code>
on some situations but not always. Specifically, on generic code written
for both, you cannot use relational operators directly, and must use the
template functions <a href="../../../../utility/OptionalPointee.html#equal" target="_top"><code class="computeroutput"><span class="identifier">equal_pointees</span><span class="special">()</span></code></a>
and <a href="../../../../utility/OptionalPointee.html#less" target="_top"><code class="computeroutput"><span class="identifier">less_pointees</span><span class="special">()</span></code></a>
instead.
</p>
</div>
</div>
<table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr>
<td align="left"></td>
<td align="right"><div class="copyright-footer">Copyright © 2003 -2007 Fernando Luis Cacciola Carballal<p>
Distributed under the Boost Software License, Version 1.0. (See accompanying
file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)
</p>
</div></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="../index.html"><img src="../../../../../doc/html/images/prev.png" alt="Prev"></a><a accesskey="u" href="../index.html"><img src="../../../../../doc/html/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../../../doc/html/images/home.png" alt="Home"></a><a accesskey="n" href="synopsis.html"><img src="../../../../../doc/html/images/next.png" alt="Next"></a>
</div>
</body>
</html>