facade.qbk
来自「Boost provides free peer-reviewed portab」· QBK 代码 · 共 619 行 · 第 1/2 页
QBK
619 行
[section:facade Iterator Facade]While the iterator interface is rich, there is a core subset of theinterface that is necessary for all the functionality. We haveidentified the following core behaviors for iterators:* dereferencing* incrementing* decrementing* equality comparison* random-access motion* distance measurementIn addition to the behaviors listed above, the core interface elementsinclude the associated types exposed through iterator traits:`value_type`, `reference`, `difference_type`, and`iterator_category`.Iterator facade uses the Curiously Recurring TemplatePattern (CRTP) [Cop95]_ so that the user can specify the behaviorof `iterator_facade` in a derived class. Former designs usedpolicy objects to specify the behavior, but that approach wasdiscarded for several reasons:1. the creation and eventual copying of the policy object may create overhead that can be avoided with the current approach.2. The policy object approach does not allow for custom constructors on the created iterator types, an essential feature if `iterator_facade` should be used in other library implementations.3. Without the use of CRTP, the standard requirement that an iterator's `operator++` returns the iterator type itself would mean that all iterators built with the library would have to be specializations of `iterator_facade<...>`, rather than something more descriptive like `indirect_iterator<T*>`. Cumbersome type generator metafunctions would be needed to build new parameterized iterators, and a separate `iterator_adaptor` layer would be impossible.[h2 Usage]The user of `iterator_facade` derives his iterator class from aspecialization of `iterator_facade` and passes the derivediterator class as `iterator_facade`\ 's first template parameter.The order of the other template parameters have been carefullychosen to take advantage of useful defaults. For example, whendefining a constant lvalue iterator, the user can pass aconst-qualified version of the iterator's `value_type` as`iterator_facade`\ 's `Value` parameter and omit the`Reference` parameter which follows.The derived iterator class must define member functions implementingthe iterator's core behaviors. The following table describesexpressions which are required to be valid depending on the categoryof the derived iterator type. These member functions are describedbriefly below and in more detail in the iterator facaderequirements.[table Core Interface [ [Expression] [Effects] ] [ [`i.dereference()`] [Access the value referred to] [ [`i.equal(j)`] [Compare for equality with `j`] ] [ [`i.increment()`] [Advance by one position] ] [ [`i.decrement()`] [Retreat by one position] ] [ [`i.advance(n)`] [Advance by `n` positions] [ [`i.distance_to(j)`] [Measure the distance to `j`] ]][/ .. Should we add a comment that a zero overhead implementation of iterator_facade is possible with proper inlining?]In addition to implementing the core interface functions, an iteratorderived from `iterator_facade` typically defines severalconstructors. To model any of the standard iterator concepts, theiterator must at least have a copy constructor. Also, if the iteratortype `X` is meant to be automatically interoperate with anotheriterator type `Y` (as with constant and mutable iterators) thenthere must be an implicit conversion from `X` to `Y` or from `Y`to `X` (but not both), typically implemented as a conversionconstructor. Finally, if the iterator is to model Forward TraversalIterator or a more-refined iterator concept, a default constructor isrequired.[h2 Iterator Core Access]`iterator_facade` and the operator implementations need to be ableto access the core member functions in the derived class. Making thecore member functions public would expose an implementation detail tothe user. The design used here ensures that implementation details donot appear in the public interface of the derived iterator type.Preventing direct access to the core member functions has twoadvantages. First, there is no possibility for the user to accidentlyuse a member function of the iterator when a member of the value_typewas intended. This has been an issue with smart pointerimplementations in the past. The second and main advantage is thatlibrary implementers can freely exchange a hand-rolled iteratorimplementation for one based on `iterator_facade` without fear ofbreaking code that was accessing the public core member functionsdirectly.In a naive implementation, keeping the derived class' core memberfunctions private would require it to grant friendship to`iterator_facade` and each of the seven operators. In order toreduce the burden of limiting access, `iterator_core_access` isprovided, a class that acts as a gateway to the core member functionsin the derived iterator class. The author of the derived class onlyneeds to grant friendship to `iterator_core_access` to make his coremember functions available to the library.`iterator_core_access` will be typically implemented as an emptyclass containing only private static member functions which invoke theiterator core member functions. There is, however, no need tostandardize the gateway protocol. Note that even if`iterator_core_access` used public member functions it would notopen a safety loophole, as every core member function preserves theinvariants of the iterator.[h2 `operator\[\]`]The indexing operator for a generalized iterator presents specialchallenges. A random access iterator's `operator[]` is onlyrequired to return something convertible to its `value_type`.Requiring that it return an lvalue would rule out currently-legalrandom-access iterators which hold the referenced value in a datamember (e.g. |counting|_), because `*(p+n)` is a referenceinto the temporary iterator `p+n`, which is destroyed when`operator[]` returns... |counting| replace:: `counting_iterator`Writable iterators built with `iterator_facade` implement thesemantics required by the preferred resolution to `issue 299`_ andadopted by proposal n1550_: the result of `p[n]` is an objectconvertible to the iterator's `value_type`, and `p[n] = x` isequivalent to `*(p + n) = x` (Note: This result object may beimplemented as a proxy containing a copy of `p+n`). This approachwill work properly for any random-access iterator regardless of theother details of its implementation. A user who knows more aboutthe implementation of her iterator is free to implement an`operator[]` that returns an lvalue in the derived iteratorclass; it will hide the one supplied by `iterator_facade` fromclients of her iterator... _n1550: http://anubis.dkuug.dk/JTC1/SC22/WG21/docs/papers/2003/n1550.html.. _`issue 299`: http://anubis.dkuug.dk/jtc1/sc22/wg21/docs/lwg-active.html#299.. _`operator arrow`:[h2 `operator->`]The `reference` type of a readable iterator (and today's inputiterator) need not in fact be a reference, so long as it isconvertible to the iterator's `value_type`. When the `value_type`is a class, however, it must still be possible to access membersthrough `operator->`. Therefore, an iterator whose `reference`type is not in fact a reference must return a proxy containing a copyof the referenced value from its `operator->`.The return types for `iterator_facade`\ 's `operator->` and`operator[]` are not explicitly specified. Instead, those typesare described in terms of a set of requirements, which must besatisfied by the `iterator_facade` implementation... [Cop95] [Coplien, 1995] Coplien, J., Curiously Recurring Template Patterns, C++ Report, February 1995, pp. 24-27.[section:facade_reference Reference] template < class Derived , class Value , class CategoryOrTraversal , class Reference = Value& , class Difference = ptrdiff_t > class iterator_facade { public: typedef remove_const<Value>::type value_type; typedef Reference reference; typedef Value\* pointer; typedef Difference difference_type; typedef /* see below__ \*/ iterator_category; reference operator\*() const; /* see below__ \*/ operator->() const; /* see below__ \*/ operator[](difference_type n) const; Derived& operator++(); Derived operator++(int); Derived& operator--(); Derived operator--(int); Derived& operator+=(difference_type n); Derived& operator-=(difference_type n); Derived operator-(difference_type n) const; protected: typedef iterator_facade iterator_facade\_; }; // Comparison operators template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type // exposition operator ==(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type operator !=(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type operator <(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type operator <=(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type operator >(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> typename enable_if_interoperable<Dr1,Dr2,bool>::type operator >=(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); // Iterator difference template <class Dr1, class V1, class TC1, class R1, class D1, class Dr2, class V2, class TC2, class R2, class D2> /* see below__ \*/ operator-(iterator_facade<Dr1,V1,TC1,R1,D1> const& lhs, iterator_facade<Dr2,V2,TC2,R2,D2> const& rhs); // Iterator addition template <class Dr, class V, class TC, class R, class D> Derived operator+ (iterator_facade<Dr,V,TC,R,D> const&, typename Derived::difference_type n); template <class Dr, class V, class TC, class R, class D> Derived operator+ (typename Derived::difference_type n, iterator_facade<Dr,V,TC,R,D> const&);__ `iterator category`___ `operator arrow`___ brackets___ minus_.. _`iterator category`:The `iterator_category` member of `iterator_facade` is.. parsed-literal:: *iterator-category*\ (CategoryOrTraversal, value_type, reference)where *iterator-category* is defined as follows:.. include:: facade_iterator_category.rstThe `enable_if_interoperable` template used above is for expositionpurposes. The member operators should only be in an overload setprovided the derived types `Dr1` and `Dr2` are interoperable, meaning that at least one of the types is convertible to the other. The`enable_if_interoperable` approach uses SFINAE to take the operatorsout of the overload set when the types are not interoperable. The operators should behave *as-if* `enable_if_interoperable`were defined to be: template <bool, typename> enable_if_interoperable_impl {}; template <typename T> enable_if_interoperable_impl<true,T> { typedef T type; };
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?