1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
//! **The server, whole** (DESIGN §8.5, VISION §4.8, REMOTE §8, §12):
//! [`Engine::serve`], which is what a bare `yog` *is*.
//!
//! It is here and not in `main.rs` for that file's own standing reason
//! (bl-f6fe): `main.rs` is coverage-excluded, so anything that lives there is
//! free to drift and no test can notice. What kept this arm there was that it
//! never returned — it parked forever — and bl-269a is exactly what dissolved
//! that: with a stop the loop ends, so the whole face is an ordinary function a
//! test drives to completion.
//!
//! There is one face now (bl-7942), so V5.4's *"nothing here is a second
//! implementation"* has nothing left to hold apart: `main.rs` is one call into
//! this one, which is exactly what a coverage-excluded file should hold of a
//! face.
use Engine;
use crateSystemClock;
use crateEnv;
use Arc;
/// What a `yog` that could not become the engine exits with (bl-1d9b) — the
/// two boot refusals' one code. It is a *usage* class, `sysexits.h`'s `EX_*`
/// range like [`world::seat::REFUSED`](crate::world::seat::REFUSED), and it is
/// deliberately not `0`: a supervisor restarting a unit, a script chaining a
/// launch, and a human reading `echo $?` all learn the same thing from it, and
/// none of them learned anything before.
pub const NO_ENGINE: i32 = 69;