Files
AILang/bench/orchestrator-stats/2026-05-14-iter-pd.1.json
T
Brummel ffa80326a3 iter pd.1: core API split — load_modules_with + build_workspace + implicit_imports threading
Carved load_workspace_with's inline DFS+inject+validate block into two
new public fns: load_modules_with (DFS only) and build_workspace
(validation + registry, takes implicit_imports: &[&str]). Threaded the
new parameter through validate_canonical_type_names + check_class_ref
+ check_type_con_name; the four hardcoded literal-"prelude" fallback
blocks in the diagnostic helpers became iter-loops over implicit_imports.
load_workspace_with survives as a 3-line shim composing the new fns +
the still-in-place prelude inject, so surface (ailang-surface) and all
downstream crates compile + test unchanged. Retired pub fn
load_workspace (zero production callers; 9 in-mod test callers
migrated to load_workspace_with(&entry, load_one); load_one + the
load_module import gated under cfg(test) since they're now test-only).
Production literal-"prelude" count in workspace.rs dropped 8 -> 4 (all
4 in the shim's inject path; none in the diagnostic helpers, which
was the spec's pd.1 target).

pd.2 will move the prelude embed + the inject step into ailang-surface
and retire the shim; pd.3 will delete examples/prelude.ail.json.

cargo test --workspace green; bench/check.py + compile_check.py +
cross_lang.py all exit 0.
2026-05-14 12:25:12 +02:00

20 lines
316 B
JSON

{
"iter_id": "pd.1",
"date": "2026-05-14",
"mode": "standard",
"outcome": "DONE",
"tasks_total": 6,
"tasks_completed": 6,
"reloops_per_task": {
"1": 0,
"2": 0,
"3": 0,
"4": 0,
"5": 0,
"6": 0
},
"review_loops_spec": 0,
"review_loops_quality": 0,
"blocked_reason": null
}