The simple question, that hit me the other day, is why don’t we just throw std::error_code or std::error_conditon.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | // throw/catch std::error_code's try { throw std::make_error_code( std::errc::not_enough_memory ); } catch ( std::error_code ec ) { // querying the code caught if (ec == std::errc::not_enough_memory) { std::clog << "\nNot enough memory" ; } } catch (std::error_condition ecn) { // ... } |
Let me try and calmly observe.
- This is not using RTTI.
- This is not dynamic types catching.
- This is throw and catch by value.
What are the drawbacks of this for new standard C++ projects? There must be some, but right now I am having difficulties to find them drawbacks.
For throwing this from public API, this renders the API not “std::exception friendly”. But right now, it seems we have the right not to care. That is the sentiment in the C++ community at large.
Is this leading to fast(er) code , even if /EH switch is used?
