Include the current addon, terrain integrations, tools, regression tests and recorded development evidence.
2 lines
1.3 KiB
Plaintext
2 lines
1.3 KiB
Plaintext
Concrete response to native diagnostic reasons8/nodes128/allowed59/managers0: known cargo was pruned correctly but exhausted the old shared queue budget. Two independent processed-node budgets:128 non-deposit structural identities (includes pruned crew and unknown inventory) and128 positively identified deposit cargo roots. Sum queue hardcap256, unique/cycle rejection retained; deposit/character subtrees never traversed. Structural129 still incomplete with exactly128 GetChildren calls; queued129th node is rejected before descent. Any individual or queue overflow remains incomplete, never treated as known empty. Existing installed-slot reciprocal exemptions and manager/weapon/muzzle/magazine limits remain unchanged. WORKBENCH reason8 now queue-limit256;1024 structural-limit128;2048 cargo-limit128. Diagnostic records structural/cargo counts in the same scan and emits one additional bounded row only in existing explicit failed fixture. Actual-method tests: old source fails realistic128deposit cargo plus mounted graph; new passes, all original assertions pass both default+WORKBENCH, three mixed ordering cases exercise each budget, cargo native weapon components never sampled. Two native fixture regression suites also pass. No new tests registry; no target/native writes.
|