{"id":109503,"date":"2024-03-08T07:00:00","date_gmt":"2024-03-08T15:00:00","guid":{"rendered":"https:\/\/devblogs.microsoft.com\/oldnewthing\/?p=109503"},"modified":"2024-03-11T10:14:55","modified_gmt":"2024-03-11T17:14:55","slug":"20240308-00","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/oldnewthing\/20240308-00\/?p=109503","title":{"rendered":"How can I force a copy of a C++ value?"},"content":{"rendered":"

Last time, we saw a case where we wanted to pass a copy of an object to an operator that accepted objects by reference<\/a>. We solved this by forcing a copy and using the copy.<\/p>\n

\/\/ Original\r\nco_await m_pendingAction;\r\n\r\n\/\/ Alternative 1: Copy to a local\r\nauto pendingAction = m_pendingAction;\r\nco_await pendingAction;\r\n\r\n\/\/ Alternative 2: Copy to a temporary\r\nco_await winrt::IAsyncAction(m_pendingAction);\r\n<\/pre>\n
Our problem was motivated by the co_await<\/code> operator, but this problem afflicts function calls, too.<\/p>\n
void consume(Widget const& widget, std::function<void()> callback);\r\n\r\nauto widget = std::make_unique<Widget>();\r\nconsume(*widget, [&] {\r\n    widget.reset();\r\n});\r\n<\/pre>\n
In the above contrived example, we create a widget<\/code> and pass it by reference to the consume<\/code> method, but in the callback, we destroy the widget! You probably aren’t crazy enough to do this on purpose, but it might happen by accident due to reentrancy or a complex interaction between components. Again, the solution is to pass a copy.<\/p>\n
\/\/ Pass a copy of the widget to consume in case the\r\n\/\/ original gets destroyed.\r\n\r\n\/\/ Alternative 1: Copy to a local\r\nauto widget_copy = *widget;\r\nconsume(widget_copy, [&] {\r\n    widget.reset();\r\n});\r\n\r\n\/\/ Alternative 2: Copy to a temporary\r\nconsume(Widget(*widget), [&] {\r\n    widget.reset();\r\n});\r\n<\/pre>\n
If only there were a convenient way to make a temporary copy of something. The first alternative requires you to name a local variable, and the lifetime of that local variable will extend to the of the block, which may be longer than you were expecting.<\/p>\n
The second alternative creates the temporary inline, but you have to spell out the type name, and that type name can get quite unwieldy.<\/p>\n
Maybe there’s a third way that gives us the inline temporary but also avoids typing out the type name. Here’s a start:<\/p>\n
template<typename T>\r\nstd::decay_t<T> copy_of(T&& t)\r\n{ return std::forward<T>(t); }\r\n\r\nco_await copy_of(m_pendingAction);\r\n\r\nconsume(copy_of(*widget), [&] {\r\n    widget.reset();\r\n});\r\n<\/pre>\n
The copy_of<\/code> takes an arbitrary reference and returns a copy of whatever it refers to. There are some sneaky bits here.<\/p>\n
We can’t declare the parameter as copy_of(T const& t)<\/code> because T<\/code> might not have a copy constructor from T const&<\/code>.<\/p>\n
We use std::decay_t<\/code> in case T<\/code> has a cv-qualifier. (And to get rid of any reference declarators on T<\/code> that may have been introduced by the forwarding reference.)<\/p>\n
Technically, the std::forward<T><\/code> is not necessary starting in C++17-maybe-20-question-mark?\u00b9 thanks to  P18250R0<\/a> which introduced implicit moves from rvalue reference parameters into return values. Basically, it says that if you return<\/code> a parameter that is an rvalue reference, then the compiler is allowed to move from that parameter instead of its normal behavior of treating it as an lvalue. Nonetheless, it doesn’t hurt to be explicit about it, particularly if you aren’t confident that everyone will be using a C++17 compiler that incorporates P18250R0, or if you want your code to work downlevel to C++11.<\/p>\n
There’s also a problem if T<\/code>‘s copy constructor or move copy constructor is explicit. We’ll have to use an explicit conversion.<\/p>\n
And I forgot to put a constexpr<\/code> on the function, so it can be compile-time evaluated; and I forgot to put a noexcept<\/code> clause, which probably should follow  the noexcept(noexcept)<\/code> idiom<\/a>, leaving us with<\/p>\n
template<typename T>\r\nstd::decay_t<T> copy_of(T&& t)\r\nnoexcept(noexcept(\r\n    std::decay_t<T>(std::forward<T>(t))\r\n    ))\r\n{ return\r\n    std::decay_t<T>(std::forward<T>(t))\r\n; } \r\n<\/pre>\n
But wait, we’re not done yet.<\/p>\n
In  P0849R8<\/a>, C++23 added “auto<\/code> decay copy” as a core language feature, so starting in C++23, you can write auto(x)<\/code> to make a copy of x<\/code>, removing the need for the copy_of<\/code> function entirely.<\/p>\n
\/\/ Await a copy because Cancel()\r\n\/\/ can modify it while awaiting.\r\nco_await auto(m_pendingAction);\r\n\r\n\/\/ Consume a copy because the lambda modifies it\r\n\/\/ from the callback.\r\nconsume(auto(*widget), [&] {\r\n    widget.reset();\r\n});\r\n<\/pre>\n
It took over a decade, but we finally got there.<\/p>\n
\u00b9 P1825R0 was submitted as a C++17 defect report, which makes it retroactive to C++17, but it naturally takes time for compilers to implement the required changes, so the confidence level doesn’t really kick in until C++20.<\/p>\n","protected":false},"excerpt":{"rendered":"
Forcing a copy with a minimum of typing.<\/p>\n","protected":false},"author":1069,"featured_media":111744,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1],"tags":[25],"class_list":["post-109503","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-oldnewthing","tag-code"],"acf":[],"blog_post_summary":"
Forcing a copy with a minimum of typing.<\/p>\n","_links":{"self":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/109503","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/users\/1069"}],"replies":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/comments?post=109503"}],"version-history":[{"count":0,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/109503\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media\/111744"}],"wp:attachment":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media?parent=109503"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/categories?post=109503"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/tags?post=109503"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}