• Joerg Joergonson (12/12) Jun 15 2016 Suppose I have a loop where I execute two functions:
• Andrea Fontana (2/14) Jun 16 2016 Something like: https://dpaste.dzfl.pl/79b77c934825 ?
• Rene Zwanenburg (8/12) Jun 16 2016 Storing the state is your best bet. Based on your recent post

Joerg Joergonson <JJoergonson gmail.com> writes:

Suppose I have a loop where I execute two functions:

for(...)
{
    if (x) Do1(x);
    if (y) Do2(y);
}

The problem is, I really always want to execute all the Do2's 
first then the Do1's. As is, we could get any order of calls.

Suppose I can't run the loop twice for performance reasons(there 
is other stuff in it) and I don't want to store the state and 
call info then sort them out afterwards.

Is there an efficient lazy way to make this happen?

Andrea Fontana <nospam example.com> writes:

On Thursday, 16 June 2016 at 01:57:19 UTC, Joerg Joergonson wrote:
 Suppose I have a loop where I execute two functions:

 for(...)
 {
    if (x) Do1(x);
    if (y) Do2(y);
 }

 The problem is, I really always want to execute all the Do2's 
 first then the Do1's. As is, we could get any order of calls.

 Suppose I can't run the loop twice for performance 
 reasons(there is other stuff in it) and I don't want to store 
 the state and call info then sort them out afterwards.

 Is there an efficient lazy way to make this happen?

Something like: https://dpaste.dzfl.pl/79b77c934825 ?

Rene Zwanenburg <renezwanenburg gmail.com> writes:

On Thursday, 16 June 2016 at 01:57:19 UTC, Joerg Joergonson wrote:
 Is there an efficient lazy way to make this happen?

No, I don't see how that would work.

 Suppose I can't run the loop twice for performance 
 reasons(there is other stuff in it) and I don't want to store 
 the state and call info then sort them out afterwards.

Storing the state is your best bet. Based on your recent post 
about OpenGL I assume this is for the same project?

If so, you can reuse the storage buffer between frames. Generally 
speaking it's the allocations that are slow, copying some state 
to a buffer shouldn't be expensive. Use separate buffers for the 
Do1 and Do2 calls so you don't have to sort.