Re: Threads across programming languages

On Fri, 3 May 2024 08:45:58 +0200, Bonita Montero wrote:

Am 03.05.2024 um 01:21 schrieb Lawrence D'Oliveiro:
 

I/O performance certainly is possible with Python, and it has the high-
performance production-quality frameworks to prove it.

 
I thought about high performance code with >= 1e5 IOs/s.
That's not possible with Python.

Sure it is. Try the “stress_test” script I wrote here
<https://gitlab.com/ldo/inotipy_examples>. It can easily generate more I/O
events than the Linux kernel can cope with, on whatever machine you’re on.

Try it with something that has actual lexically-bound local variables
in it:
 
     def factory(count : int) :
 
         def counter() :
             nonlocal count count += 1 return count
         #end counter
 
     #begin
         return counter
     #end factory
 
     f1 = factory(3)
     f2 = factory(30) print(f1())
     print(f2())
     print(f1())
     print(f2())
 
output:
 
     4
     31
     5
     32

 
That should be similar:
 
#include <iostream>
#include <functional>
 
using namespace std;
 
function<int ()> factory()
{
return []
{
static int count = 0;
return ++count;
};
}
 
int main()
{
auto
f1 = factory(),
f2 = factory();
cout << f1() << endl;
cout << f2() << endl;
cout << f1() << endl;
cout << f2() << endl;
}

Ahem, and what is the output from your C++ version?

(Hint: I don’t think it’s correct.)