Sometimes in event-oriented programming one would swallow events of the same type, which are already within a pipeline (e.g. refresh window, move mouse pointer, etc.) But that would usually be a queue (FIFO) and not a stack (LIFO) ...   -----Ursprüngliche Nachricht----- Von:Jason McKesson via Std-Proposals <std-proposals_at_[hidden]> Gesendet:Mi 03.01.2024 21:02 Betreff:Re: [std-proposals] New Data structure. An:std-proposals_at_[hidden]; CC:Jason McKesson <jmckesson_at_[hidden]>; On Wed, Jan 3, 2024 at 2:48 PM organicoman via Std-Proposals <std-proposals_at_[hidden]> wrote: > > Hello Gents, > I would like to suggest a new data structure that combines 'std::unordered_set' and a 'std::stack' behavior. > I will refere to it as 'std::hashed_stack'. > > Important methods of 'std::hashed_stack' container are: > >  -'push': to insert a key, and if it already exists nothing will happen. > >  -'pop': to extract the last key inserted. But not the last key *pushed* if it was already in the container? That sounds *really* weird. What's the use case for this data structure? In fact, I'm not sure when I would want either behavior. If I'm using a thing that calls itself a "stack," the idea that `push` and `pop` should be balanced is non-negotiable. That's what makes it a *stack*. The possibility that pushing two things and poping once could result in a balanced stack is... bizarre. If I push a thing twice, I should have to pop it twice, even if it's the same thing and takes up the same space in the stack's storage. -- Std-Proposals mailing list Std-Proposals_at_[hidden] https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals