Open LiamM32 opened 6 months ago
I have my own forked repository now. If I get around to it, I may make this and commit it.
Should the overloaded versions of the functions be placed right underneath the declarations for the current ones (which use const(char)*
), or should they all be placed in their own block of lines later in the file?
I have my own forked repository now. If I get around to it, I may make this and commit it.
Should the overloaded versions of the functions be placed right underneath the declarations for the current ones (which use
const(char)*
), or should they all be placed in their own block of lines later in the file?
Id write a script that scans for function headers with char* functions, pulls out the lines and then add that as an string[] enum
then make a mixin tempate makestringoverload(alias stringtype,alias tostringz)
theres probably to much debate to use Phobos toStringz directly and it wouldnt be that much harder to make it genertic
Some functions here have
const(char)*
as an argument type, which is used to pass strings to C functions. Calling these functions using D strings instead sometimes, but not always works.I propose that new versions of these functions are added that have D strings as arguments. They will all convert the D strings to C-style strings using
getStringz
fromstd.string
, and then call the original version of the function. Those that havecost(char)*
as their return type will instead havestring
as their return type when using a D string as an argument.