HackerTube

CppCon 2015: Andrei Alexandrescu “Declarative Control Flow"

CppCon2 HN recommenders · recommended in 2 months, 2016–2022
YouTube
talk · advanced · evergreen · 2016–2022 on HN

Andrei Alexandrescu's CppCon 2015 talk on handling exceptional control flow declaratively in C++. It uses scope guards with success and failure variants, built on std::uncaught_exceptions, to replace ad hoc RAII classes and rethrowing try/catch blocks.

Pointed to in threads on defer, scope_exit and RAII as the clearest treatment of scope guards.

C, C++, Rust, Go & systems languages#scopeguard#raii#uncaught_exceptions#cppcon

5 comments from Hacker News

@uep2018-02 · in “Scope exit in C++11 (2012)”

Andrei Alexandrescu has a fun talk about implementing scope_exit using a new C++17 function: std::uncaught_exceptions (which is a tweak to std::uncaught_exception). Fundamentally, it doesn't really solve the destructor throwing exceptions issue though. ▶ this video

Recommends View on HN
@Ace172016-07 · in “Memory management in C programs (2014)”

Google is an example that writing production code that doesn't use exceptions is possible, not that this is the right thing to do. They said that their decision was primarily legacy-motivated (in short, they were fearing that introducing exceptions in a non-RAII codebase would lead us to a long period of instability). I'll add that in some situations (e.g when freeing resources) the control flow gets very convoluted, and you don't want to deal with directly. See: Andrei Alexandrescu - Declarative Control Flow ( ▶ this video )

Recommends View on HN
@einpoklum2022-10 · in “A Tour of C++ (Third edition)”

A count of the number of exceptions in-flight was added, replacing the boolean indication of whether an exception is in-flight. This allows the scope guard construct proposed by Andrei Alexandrescu to distinguish between a scope being left on success and on an exception being thrown. YouTube video about scope guard: ▶ this video Description of the change to the standard: link

Cites View on HN
@watmough2022-06 · in “Ask HN: What should have been the term for RAII?”

I think scope or similar is good. Also, see related video here CppCon 2015: Andrei Alexandrescu “Declarative Control Flow" ▶ this video

Cites View on HN
@dnautics2020-12 · in “A Defer Mechanism for C”

yes, you always run defer. I don't see it in this proposal, but Most languages/frameworks that implement a defer also implement a "error defer" which only gets triggered on some sort of labeled early exit scenario, and some implement "success defer": (i believe this is scopeguard: ▶ this video )

Cites View on HN

Up next