[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
- Subject: "Daisy-Chaining" of C functions with lua_insert stack data: necessary to clean up stack before return?
- From: bil til <biltil52@...>
- Date: Sun, 13 Nov 2022 21:35:42 +0100
Sorry for a further maybe a bit "greedy" qeustion concerning my C code.
I have some parts of my code, where I want to "daisy-chain" a Lua
function to some other C "sub-function" of me, and for this I need to
add some additional parameters on the Lua stack from time to time, e.
g. I use my own version of luab_print like this:
int luaB_print (lua_State *L) {
lua_pushnil( L);
lua_insert( L, 1);
lua_pushnil( L);
lua_insert( L, 2);
lua_pushnil( L);
lua_insert( L, 3);
return com_print_i( L, WRITEMODE_print );
}
(my internal function "com_print_i" needs three additional parameter
as first function parameters... to keep it easy I just inserted NIL
here 3 times at the start of the the stack).
In a first approach, I cleaned up the stack again tediously in
com_print_i, before I return 0 (luaB_print should not return
parameters, so I just return 0):
for( int i= 0; i< 3; i++)
lua_remove( L, 1);
return 0;
}
... just after thinking a bit about this (and as this is quite a
cumbersome and tedious procedure, if I have to do this at several
points of my C code), I thought that from my understanding of
"invoking a function with stack", this should be NOT necessary, as on
return the internal Lua machine anyway should strip the stack to
exactly the same point where luaB_print was invoked.
So I skipped this "lua_remove" for loop, and I do not see anything bad
happening so far... .
Am I correct here? Or do I overlook any possibly nerving side effects?
(I use yield / resume inside com_print_i)?
(if a person with deeper understanding of the "Lua internals" could
look into this, I would be very happy...).