nx_fetchfail_gate.nx
buildroot/runtime/nx_fetchfail_gate.nx
about
nx_fetchfail_gate.nx -- teeth for the CONTENT axis and the plane that carries it.
IN-PROCESS BY CONSTRUCTION: it imports nx_mirrorintegrity_lib.nx and calls mi_content_class directly,
so it exercises THE SAME ruler nx_fetchfail runs instead of parsing that program's stdout. It has no
NOT-DEPLOYED failure mode.
EVERY FIXTURE IS ASSEMBLED AT RUNTIME UNDER /tmp/fetchfail_gate/, never checked in and never shared
with a production beat, and the prose floor is passed as an ARGUMENT rather than read from the
shipped conf -- so this gate cannot be turned green or red by an edit to production data.
Setup unlinks before it creates: A GATE THAT IS NOT IDEMPOTENT REPORTS ON ITS FIRST RUN AND LIES
ABOUT EVERY RUN AFTER.
THE FIRST REAL TOOTH IS ANTI-VACUITY. The trivial wrong implementation is "flag every body", which
passes every detection tooth below. Only bodies that MUST NOT be flagged can refute it, and the one
that actually matters is a page whose PROSE DISCUSSES bot walls in the very words the rules match on.
A DETECTOR THAT CANNOT TELL A PAGE ABOUT A DEFECT FROM THE DEFECT IS A FALSE-POSITIVE GENERATOR.
license_tier: ORIGINAL expect_exit: 0
dependencies 2 imports · 0 importers
imports: nx_mirrorintegrity_lib.nxnx_gate_verdict.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| 21 | const FG_DIR: *u8 = "/tmp/fetchfail_gate\x00" |
| 22 | const FG_MODE644: i64 = 420 |
| 23 | const FG_MODE755: i64 = 493 |
| 27 | const FG_FLOOR: i64 = 64 |
functions
| 29 | func fg_put(path: *u8, data: *u8) -> i64 called by 1: main |
| 39 | func fg_path(buf: *u8, name: *u8) -> i64 |
| 47 | func main(argc: i64, argv: *i64) -> i64 |