nx_matstruct.nx

nx_matstruct.nx -- ★THE STRUCTURAL MATERIAL REGISTRY: what a part is MADE OF, and what that implies. ★THE GAP THIS CLOSES (seq1417, measured 2026-07-30): the ecosystem HAS a real structural primitive -- nx_stress computes sigma = F/A in integer Pascals and returns SEALED verdicts SAFE/WORKING/YIELDED/ FRACTURED -- but it imports nx_material_profile.nx and inherits ITS enum, which is the SLICER'S FILAMENT REGISTRY (PLA, PETG, ABS, ASA, PC, TPU, NYLON, PA-CF, PEEK, carrying hotend temps and retraction). So the entire structural-material universe was 3D-PRINTER FILAMENT. 'This bracket is 1018 steel' was INEXPRESSIBLE, and an underbuilt verdict on a car was impossible no matter how good the geometry got. ★★WHY THIS IS A NEW ORGAN AND NOT SIX MORE ROWS IN NX_MAT_*: that enum is OWNED BY THE PRINT LANE and is backed by a struct of hotend temperature, cooling and pressure advance. Adding steel to it would owe a hotend temperature for steel. A filament class and a structural-material class are DIFFERENT THINGS THAT SHARE A NAME; binding a general primitive to a domain-specific registry was the category error. This registry keeps the print lane's data intact and simply stops being the only answer. ★★★NOTHING IS STRIPPED (rule 25 -- the answer is always better data, never fewer options): indices 0..8 ARE the nine filaments, in nx_stress's exact order, carrying its exact shipped yields and ultimates, so this is a strict SUPERSET and a caller can migrate by index. T6 proves that correspondence value by value against the literals nx_stress ships, so a transcription slip cannot pass. nx_matstruct selftest license_tier: ORIGINAL expect_exit: 0 No hw writes (Rule 26).