key_iterator Class

Detailed Description

QHash::key_iterator is essentially the same as QHash::const_iterator with the difference that operator*() and operator->() return a key instead of a value.

For most uses QHash::iterator and QHash::const_iterator should be used, you can easily access the key by calling QHash::iterator::key():

 for (QHash<int, QString>::const_iterator it = hash.cbegin(), end = hash.cend(); it != end; ++it) {
     cout << "The key: " << it.key() << Qt::endl
     cout << "The value: " << it.value() << Qt::endl;
     cout << "Also the value: " << (*it) << Qt::endl;
 }

However, to have interoperability between QHash's keys and STL-style algorithms we need an iterator that dereferences to a key instead of a value. With QHash::key_iterator we can apply an algorithm to a range of keys without having to call QHash::keys(), which is inefficient as it costs one QHash iteration and memory allocation to create a temporary QList.

 // Inefficient, keys() is expensive
 QList<int> keys = hash.keys();
 int numPrimes = std::count_if(keys.cbegin(), keys.cend(), isPrimeNumber);
 qDeleteAll(hash2.keys());

 // Efficient, no memory allocation needed
 int numPrimes = std::count_if(hash.keyBegin(), hash.keyEnd(), isPrimeNumber);
 qDeleteAll(hash2.keyBegin(), hash2.keyEnd());

QHash::key_iterator is const, it's not possible to modify the key.

The default QHash::key_iterator constructor creates an uninitialized iterator. You must initialize it using a QHash function like QHash::keyBegin() or QHash::keyEnd().

Warning: Iterators on implicitly shared containers do not work exactly like STL-iterators. You should avoid copying a container while iterators are active on that container. For more information, read Implicit sharing iterator problem.