feat(runtime): implement single-article consolidation runtime and modularize codebase

This commit is contained in:
2026-08-24 00:14:07 -03:00
parent e1e0be1353
commit 23de7d8fe7
176 changed files with 266754 additions and 10179 deletions
@@ -0,0 +1,23 @@
"""Multi-extractor golden-set quality tests across the 20 reference units."""
import json
from pathlib import Path
from src.runtime.candidate.parser import build_candidates_payload
from src.runtime.hygiene.harness import execute_deterministic_hygiene_fallback
def test_golden_set_all_20_reference_cases_process_cleanly():
ref_dir = Path("evals/reference_20")
files = list(ref_dir.glob("article_*.json"))
assert len(files) == 20, f"Expected 20 reference unit files, found {len(files)}"
for f in sorted(files):
data = json.loads(f.read_text(encoding="utf-8"))
payload = build_candidates_payload(data)
# Verify deterministic extraction works for all 20 units
md, meta = execute_deterministic_hygiene_fallback(payload)
assert len(md) > 0
assert meta["title"] is not None
assert meta["kept_block_count"] >= 1