/**************************************************************************** ** ** Copyright (C) 2016 The Qt Company Ltd. ** Contact: https://www.qt.io/licensing/ ** ** This file is part of the documentation of the Qt Toolkit. ** ** $QT_BEGIN_LICENSE:FDL$ ** Commercial License Usage ** Licensees holding valid commercial Qt licenses may use this file in ** accordance with the commercial license agreement provided with the ** Software or, alternatively, in accordance with the terms contained in ** a written agreement between you and The Qt Company. For licensing terms ** and conditions see https://www.qt.io/terms-conditions. For further ** information use the contact form at https://www.qt.io/contact-us. ** ** GNU Free Documentation License Usage ** Alternatively, this file may be used under the terms of the GNU Free ** Documentation License version 1.3 as published by the Free Software ** Foundation and appearing in the file included in the packaging of ** this file. Please review the following information to ensure ** the GNU Free Documentation License version 1.3 requirements ** will be met: https://www.gnu.org/licenses/fdl-1.3.html. ** $QT_END_LICENSE$ ** ****************************************************************************/ /*! \headerfile \title Generic Algorithms \ingroup funclists \brief The header includes the generic, template-based algorithms. Qt provides a number of global template functions in \c that work on containers and perform small tasks to make life easier, such as qDeleteAll(), which invokes \c{operator delete} on all items in a given container or in a given range. You can use these algorithms with any \l {container class} that provides STL-style iterators, including Qt's QList, QLinkedList, QVector, QMap, and QHash classes. Most algorithms take \l {STL-style iterators} as parameters. The algorithms are generic in the sense that they aren't bound to a specific iterator class; you can use them with any iterators that meet a certain set of requirements. Different algorithms can have different requirements for the iterators they accept. For example, qFill() accepts two \l {forward iterators}. The iterator types required are specified for each algorithm. If an iterator of the wrong type is passed (for example, if QList::ConstIterator is passed as an \l {Output Iterators}{output iterator}), you will always get a compiler error, although not necessarily a very informative one. Some algorithms have special requirements on the value type stored in the containers. For example, qDeleteAll() requires that the value type is a non-const pointer type (for example, QWidget *). The value type requirements are specified for each algorithm, and the compiler will produce an error if a requirement isn't met. The generic algorithms can be used on other container classes than those provided by Qt and STL. The syntax of STL-style iterators is modeled after C++ pointers, so it's possible to use plain arrays as containers and plain pointers as iterators. A common idiom is to use qBinaryFind() together with two static arrays: one that contains a list of keys, and another that contains a list of associated values. For example, the following code will look up an HTML entity (e.g., \c &) in the \c name_table array and return the corresponding Unicode value from the \c value_table if the entity is recognized: \snippet code/doc_src_qalgorithms.cpp 2 This kind of code is for advanced users only; for most applications, a QMap- or QHash-based approach would work just as well: \snippet code/doc_src_qalgorithms.cpp 3 \section1 Types of Iterators The algorithms have certain requirements on the iterator types they accept, and these are specified individually for each function. The compiler will produce an error if a requirement isn't met. \section2 Input Iterators An \e{input iterator} is an iterator that can be used for reading data sequentially from a container. It must provide the following operators: \c{==} and \c{!=} for comparing two iterators, unary \c{*} for retrieving the value stored in the item, and prefix \c{++} for advancing to the next item. The Qt containers' iterator types (const and non-const) are all input iterators. \section2 Output Iterators An output iterator is an iterator that can be used for writing data sequentially to a container or to some output stream. It must provide the following operators: unary \c{*} for writing a value (i.e., \c{*it = val}) and prefix \c{++} for advancing to the next item. The Qt containers' non-const iterator types are all output iterators. \section2 Forward Iterators A \e{forward iterator} is an iterator that meets the requirements of both input iterators and output iterators. The Qt containers' non-const iterator types are all forward iterators. \section2 Bidirectional Iterators A \e{bidirectional iterator} is an iterator that meets the requirements of forward iterators but that in addition supports prefix \c{--} for iterating backward. The Qt containers' non-const iterator types are all bidirectional iterators. \section2 Random Access Iterators The last category, \e{random access iterators}, is the most powerful type of iterator. It supports all the requirements of a bidirectional iterator, and supports the following operations: \table \row \li \c{i += n} \li advances iterator \c i by \c n positions \row \li \c{i -= n} \li moves iterator \c i back by \c n positions \row \li \c{i + n} or \c{n + i} \li returns the iterator for the item \c n positions ahead of iterator \c i \row \li \c{i - n} \li returns the iterator for the item \c n positions behind of iterator \c i \row \li \c{i - j} \li returns the number of items between iterators \c i and \c j \row \li \c{i[n]} \li same as \c{*(i + n)} \row \li \c{i < j} \li returns \c true if iterator \c j comes after iterator \c i \endtable QList and QVector's non-const iterator types are random access iterators. \section1 Qt and the STL Algorithms Historically, Qt used to provide functions which were direct equivalents of many STL algorithmic functions. Starting with Qt 5.0, you are instead encouraged to use directly the implementations available in the STL; most of the Qt ones have been deprecated (although they are still available to keep the old code compiling). \section2 Porting guidelines Most of the times, an application using the deprecated Qt algorithmic functions can be easily ported to use the equivalent STL functions. You need to \list 1 \li add the \c{#include } preprocessor directive; \li replace the Qt functions with the STL counterparts, according to the table below. \endlist \table \header \li Qt function \li STL function \row \li qBinaryFind \li \c std::binary_search or \c std::lower_bound \row \li qCopy \li \c std::copy \row \li qCopyBackward \li \c std::copy_backward \row \li qEqual \li \c std::equal \row \li qFill \li \c std::fill \row \li qFind \li \c std::find \row \li qCount \li \c std::count \row \li qSort \li \c std::sort \row \li qStableSort \li \c std::stable_sort \row \li qLowerBound \li \c std::lower_bound \row \li qUpperBound \li \c std::upper_bound \row \li qLess \li \c std::less \row \li qGreater \li \c std::greater \endtable The only cases in which the port may not be straightforward is if the old code relied on template specializations of the qLess() and/or the qSwap() functions, which were used internally by the implementations of the Qt algorithmic functions, but are instead ignored by the STL ones. In case the old code relied on the specialization of the qLess() functor, then a workaround is explicitly passing an instance of the qLess() class to the STL function, for instance like this: \code std::sort(container.begin(), container.end(), qLess()); \endcode Instead, since it's not possible to pass a custom swapper functor to STL functions, the only workaround for a template specialization for qSwap() is providing the same specialization for \c std::swap(). \sa {container classes}, */ /*! \fn OutputIterator qCopy(InputIterator begin1, InputIterator end1, OutputIterator begin2) \relates \deprecated Use \c std::copy instead. Copies the items from range [\a begin1, \a end1) to range [\a begin2, ...), in the order in which they appear. The item at position \a begin1 is assigned to that at position \a begin2; the item at position \a begin1 + 1 is assigned to that at position \a begin2 + 1; and so on. Example: \snippet code/doc_src_qalgorithms.cpp 4 \sa qCopyBackward(), {input iterators}, {output iterators} */ /*! \fn BiIterator2 qCopyBackward(BiIterator1 begin1, BiIterator1 end1, BiIterator2 end2) \relates \deprecated Use \c std::copy_backward instead. Copies the items from range [\a begin1, \a end1) to range [..., \a end2). The item at position \a end1 - 1 is assigned to that at position \a end2 - 1; the item at position \a end1 - 2 is assigned to that at position \a end2 - 2; and so on. Example: \snippet code/doc_src_qalgorithms.cpp 5 \sa qCopy(), {bidirectional iterators} */ /*! \fn bool qEqual(InputIterator1 begin1, InputIterator1 end1, InputIterator2 begin2) \relates \deprecated Use \c std::equal instead. Compares the items in the range [\a begin1, \a end1) with the items in the range [\a begin2, ...). Returns \c true if all the items compare equal; otherwise returns \c false. Example: \snippet code/doc_src_qalgorithms.cpp 6 This function requires the item type (in the example above, QString) to implement \c operator==(). \sa {input iterators} */ /*! \fn void qFill(ForwardIterator begin, ForwardIterator end, const T &value) \relates \deprecated Use \c std::fill instead. Fills the range [\a begin, \a end) with \a value. Example: \snippet code/doc_src_qalgorithms.cpp 7 \sa qCopy(), {forward iterators} */ /*! \fn void qFill(Container &container, const T &value) \relates \deprecated \overload Use \c std::fill instead. This is the same as qFill(\a{container}.begin(), \a{container}.end(), \a value); */ /*! \fn InputIterator qFind(InputIterator begin, InputIterator end, const T &value) \relates \deprecated Use \c std::find instead. Returns an iterator to the first occurrence of \a value in a container in the range [\a begin, \a end). Returns \a end if \a value isn't found. Example: \snippet code/doc_src_qalgorithms.cpp 8 This function requires the item type (in the example above, QString) to implement \c operator==(). If the items in the range are in ascending order, you can get faster results by using qLowerBound() or qBinaryFind() instead of qFind(). \sa qBinaryFind(), {input iterators} */ /*! \fn void qFind(const Container &container, const T &value) \relates \deprecated \overload Use \c std::find instead. This is the same as qFind(\a{container}.constBegin(), \a{container}.constEnd(), \a value); */ /*! \fn void qCount(InputIterator begin, InputIterator end, const T &value, Size &n) \relates \deprecated Use \c std::count instead. Returns the number of occurrences of \a value in the range [\a begin, \a end), which is returned in \a n. \a n is never initialized, the count is added to \a n. It is the caller's responsibility to initialize \a n. Example: \snippet code/doc_src_qalgorithms.cpp 9 This function requires the item type (in the example above, \c int) to implement \c operator==(). \sa {input iterators} */ /*! \fn void qCount(const Container &container, const T &value, Size &n) \relates \deprecated \overload Use \c std::count instead. Instead of operating on iterators, as in the other overload, this function operates on the specified \a container to obtain the number of instances of \a value in the variable passed as a reference in argument \a n. */ /*! \fn void qSwap(T &var1, T &var2) \relates \deprecated Use \c std::swap instead. Exchanges the values of variables \a var1 and \a var2. Example: \snippet code/doc_src_qalgorithms.cpp 10 */ /*! \fn void qSort(RandomAccessIterator begin, RandomAccessIterator end) \relates \deprecated Use \c std::sort instead. Sorts the items in range [\a begin, \a end) in ascending order using the quicksort algorithm. Example: \snippet code/doc_src_qalgorithms.cpp 11 The sort algorithm is efficient on large data sets. It operates in \l {linear-logarithmic time}, O(\e{n} log \e{n}). This function requires the item type (in the example above, \c{int}) to implement \c operator<(). If neither of the two items is "less than" the other, the items are taken to be equal. It is then undefined which one of the two items will appear before the other after the sort. \sa qStableSort(), {random access iterators} */ /*! \fn void qSort(RandomAccessIterator begin, RandomAccessIterator end, LessThan lessThan) \relates \deprecated \overload Use \c std::sort instead. Uses the \a lessThan function instead of \c operator<() to compare the items. For example, here's how to sort the strings in a QStringList in case-insensitive alphabetical order: \snippet code/doc_src_qalgorithms.cpp 12 To sort values in reverse order, pass \l{qGreater()}{qGreater()} as the \a lessThan parameter. For example: \snippet code/doc_src_qalgorithms.cpp 13 If neither of the two items is "less than" the other, the items are taken to be equal. It is then undefined which one of the two items will appear before the other after the sort. An alternative to using qSort() is to put the items to sort in a QMap, using the sort key as the QMap key. This is often more convenient than defining a \a lessThan function. For example, the following code shows how to sort a list of strings case insensitively using QMap: \snippet code/doc_src_qalgorithms.cpp 14 \sa QMap */ /*! \fn void qSort(Container &container) \relates \deprecated \overload Use \c std::sort instead. This is the same as qSort(\a{container}.begin(), \a{container}.end()); */ /*! \fn void qStableSort(RandomAccessIterator begin, RandomAccessIterator end) \relates \deprecated Use \c std::stable_sort instead. Sorts the items in range [\a begin, \a end) in ascending order using a stable sorting algorithm. If neither of the two items is "less than" the other, the items are taken to be equal. The item that appeared before the other in the original container will still appear first after the sort. This property is often useful when sorting user-visible data. Example: \snippet code/doc_src_qalgorithms.cpp 15 The sort algorithm is efficient on large data sets. It operates in \l {linear-logarithmic time}, O(\e{n} log \e{n}). This function requires the item type (in the example above, \c{int}) to implement \c operator<(). \sa qSort(), {random access iterators} */ /*! \fn void qStableSort(RandomAccessIterator begin, RandomAccessIterator end, LessThan lessThan) \relates \deprecated \overload Use \c std::stable_sort instead. Uses the \a lessThan function instead of \c operator<() to compare the items. For example, here's how to sort the strings in a QStringList in case-insensitive alphabetical order: \snippet code/doc_src_qalgorithms.cpp 16 Note that earlier versions of Qt allowed using a lessThan function that took its arguments by non-const reference. From 4.3 and on this is no longer possible, the arguments has to be passed by const reference or value. To sort values in reverse order, pass \l{qGreater()}{qGreater()} as the \a lessThan parameter. For example: \snippet code/doc_src_qalgorithms.cpp 17 If neither of the two items is "less than" the other, the items are taken to be equal. The item that appeared before the other in the original container will still appear first after the sort. This property is often useful when sorting user-visible data. */ /*! \fn void qStableSort(Container &container) \relates \deprecated \overload Use \c std::stable_sort instead. This is the same as qStableSort(\a{container}.begin(), \a{container}.end()); */ /*! \fn RandomAccessIterator qLowerBound(RandomAccessIterator begin, RandomAccessIterator end, const T &value) \relates \deprecated Use \c std::lower_bound instead. Performs a binary search of the range [\a begin, \a end) and returns the position of the first occurrence of \a value. If no such item is found, returns the position where it should be inserted. The items in the range [\a begin, \e end) must be sorted in ascending order; see qSort(). Example: \snippet code/doc_src_qalgorithms.cpp 18 This function requires the item type (in the example above, \c{int}) to implement \c operator<(). qLowerBound() can be used in conjunction with qUpperBound() to iterate over all occurrences of the same value: \snippet code/doc_src_qalgorithms.cpp 19 \sa qUpperBound(), qBinaryFind() */ /*! \fn RandomAccessIterator qLowerBound(RandomAccessIterator begin, RandomAccessIterator end, const T &value, LessThan lessThan) \relates \deprecated \overload Use \c std::lower_bound instead. Uses the \a lessThan function instead of \c operator<() to compare the items. Note that the items in the range must be sorted according to the order specified by the \a lessThan object. */ /*! \fn void qLowerBound(const Container &container, const T &value) \relates \deprecated \overload Use \c std::lower_bound instead. For read-only iteration over containers, this function is broadly equivalent to qLowerBound(\a{container}.begin(), \a{container}.end(), value). However, since it returns a const iterator, you cannot use it to modify the container; for example, to insert items. */ /*! \fn RandomAccessIterator qUpperBound(RandomAccessIterator begin, RandomAccessIterator end, const T &value) \relates \deprecated Use \c std::upper_bound instead. Performs a binary search of the range [\a begin, \a end) and returns the position of the one-past-the-last occurrence of \a value. If no such item is found, returns the position where the item should be inserted. The items in the range [\a begin, \e end) must be sorted in ascending order; see qSort(). Example: \snippet code/doc_src_qalgorithms.cpp 20 This function requires the item type (in the example above, \c{int}) to implement \c operator<(). qUpperBound() can be used in conjunction with qLowerBound() to iterate over all occurrences of the same value: \snippet code/doc_src_qalgorithms.cpp 21 \sa qLowerBound(), qBinaryFind() */ /*! \fn RandomAccessIterator qUpperBound(RandomAccessIterator begin, RandomAccessIterator end, const T &value, LessThan lessThan) \relates \deprecated \overload Use \c std::upper_bound instead. Uses the \a lessThan function instead of \c operator<() to compare the items. Note that the items in the range must be sorted according to the order specified by the \a lessThan object. */ /*! \fn void qUpperBound(const Container &container, const T &value) \relates \deprecated \overload Use \c std::upper_bound instead. This is the same as qUpperBound(\a{container}.begin(), \a{container}.end(), \a value); */ /*! \fn RandomAccessIterator qBinaryFind(RandomAccessIterator begin, RandomAccessIterator end, const T &value) \relates \deprecated Use \c std::binary_search or \c std::lower_bound instead. Performs a binary search of the range [\a begin, \a end) and returns the position of an occurrence of \a value. If there are no occurrences of \a value, returns \a end. The items in the range [\a begin, \a end) must be sorted in ascending order; see qSort(). If there are many occurrences of the same value, any one of them could be returned. Use qLowerBound() or qUpperBound() if you need finer control. Example: \snippet code/doc_src_qalgorithms.cpp 22 This function requires the item type (in the example above, QString) to implement \c operator<(). \sa qLowerBound(), qUpperBound(), {random access iterators} */ /*! \fn RandomAccessIterator qBinaryFind(RandomAccessIterator begin, RandomAccessIterator end, const T &value, LessThan lessThan) \relates \deprecated \overload Use \c std::binary_search or \c std::lower_bound instead. Uses the \a lessThan function instead of \c operator<() to compare the items. Note that the items in the range must be sorted according to the order specified by the \a lessThan object. */ /*! \fn void qBinaryFind(const Container &container, const T &value) \relates \deprecated \overload Use \c std::binary_search or \c std::lower_bound instead. This is the same as qBinaryFind(\a{container}.begin(), \a{container}.end(), \a value); */ /*! \fn void qDeleteAll(ForwardIterator begin, ForwardIterator end) \relates Deletes all the items in the range [\a begin, \a end) using the C++ \c delete operator. The item type must be a pointer type (for example, \c{QWidget *}). Example: \snippet code/doc_src_qalgorithms.cpp 23 Notice that qDeleteAll() doesn't remove the items from the container; it merely calls \c delete on them. In the example above, we call clear() on the container to remove the items. This function can also be used to delete items stored in associative containers, such as QMap and QHash. Only the objects stored in each container will be deleted by this function; objects used as keys will not be deleted. \sa {forward iterators} */ /*! \fn void qDeleteAll(const Container &c) \relates \overload This is the same as qDeleteAll(\a{c}.begin(), \a{c}.end()). */ /*! \fn LessThan qLess() \relates \deprecated Use \c std::less instead. Returns a functional object, or functor, that can be passed to qSort() or qStableSort(). Example: \snippet code/doc_src_qalgorithms.cpp 24 \sa {qGreater()}{qGreater()} */ /*! \fn LessThan qGreater() \relates \deprecated Use \c std::greater instead. Returns a functional object, or functor, that can be passed to qSort() or qStableSort(). Example: \snippet code/doc_src_qalgorithms.cpp 25 \sa {qLess()}{qLess()} */ /*! \fn uint qPopulationCount(quint8 v) \relates \since 5.2 Returns the number of bits set in \a v. This number is also called the Hamming Weight of \a v. */ /*! \fn uint qPopulationCount(quint16 v) \relates \since 5.2 \overload */ /*! \fn uint qPopulationCount(quint32 v) \relates \since 5.2 \overload */ /*! \fn uint qPopulationCount(quint64 v) \relates \since 5.2 \overload */ /*! \fn uint qCountTrailingZeroBits(quint8 v) \relates \since 5.6 Returns the number of consecutive zero bits in \a v, when searching from the LSB. For example, qCountTrailingZeroBits(1) returns 0 and qCountTrailingZeroBits(8) returns 3. */ /*! \fn uint qCountTrailingZeroBits(quint16 v) \relates \since 5.6 \overload */ /*! \fn uint qCountTrailingZeroBits(quint32 v) \relates \since 5.6 \overload */ /*! \fn uint qCountTrailingZeroBits(quint64 v) \relates \since 5.6 \overload */ /*! \fn uint qCountLeadingZeroBits(quint8 v) \relates \since 5.6 Returns the number of consecutive zero bits in \a v, when searching from the MSB. For example, qCountLeadingZeroBits(quint8(1)) returns 7 and qCountLeadingZeroBits(quint8(8)) returns 4. */ /*! \fn uint qCountLeadingZeroBits(quint16 v) \relates \since 5.6 Returns the number of consecutive zero bits in \a v, when searching from the MSB. For example, qCountLeadingZeroBits(quint16(1)) returns 15 and qCountLeadingZeroBits(quint16(8)) returns 12. */ /*! \fn uint qCountLeadingZeroBits(quint32 v) \relates \since 5.6 Returns the number of consecutive zero bits in \a v, when searching from the MSB. For example, qCountLeadingZeroBits(quint32(1)) returns 31 and qCountLeadingZeroBits(quint32(8)) returns 28. */ /*! \fn uint qCountLeadingZeroBits(quint64 v) \relates \since 5.6 Returns the number of consecutive zero bits in \a v, when searching from the MSB. For example, qCountLeadingZeroBits(quint64(1)) returns 63 and qCountLeadingZeroBits(quint64(8)) returns 60. */