TL
No problem, I see the issues that you mention, yes it does remove existing API, and you can no longer pick which function to call from library. Ideally though, one...
This approach can be interesting for applications allocation large blocks of data, but it is highly debatable whether there is a need for this kind of resource checking on such...
I looked over this PR again, and the changes and overhead is minimal as you also claim. Thumbs up from me.
Thanks for the suggestion. free is out of the question, because it normally means heap deallocation, which is something that should be done in addition to destruction only if the...
I don't care much for the naming of the create/destroy idiom. In general the element to be "created" already exist, but needs to be initialized. The opposite operation doesn't normally...
Edited: You could also ask why are functions and type names in STC mainly from C++ STL? C works more similar to Rust than C++: Rust has default move semantics...
I will keep drop, as I like it better than destroy and other alternatives, and it is used in Rust as destructor name. There is a c_FLT_SKIPWHILE macro which works...
The library is primarily intended as header-only and therefore it need no installation. However, the non-templated objects (particularly cstr, cregex) would be nice to have accessible as a library, as...
This search is already "illegal", so it is unspecified how data are organized internally. The last update I did is quite clean and short: it jumps with one ctest struct...