eee8f4c8a4
- New: src/rag/engine/ — in-process hybrid search (FTS5 BM25 + sqlite-vec + LLM rerank) - New: src/rag/qmd/ — compatibility layer (qmd_query, qmd_chat, qmd_chat_stream, qmd_index_*) - New: src/ingest/stub_writer.py — .md stubs for binary files (videos, archives) - New: scripts/deploy.sh + scripts/pull_models.sh + Makefile + .env.example - Removed: LightRAG, sentence-transformers embedding via separate package, rag_standalone/ - Removed: @nousresearch/qmd npm dep (package not published); Node.js from Dockerfile - Updated: tests/ (46 passed), docker-compose, .dockerignore, config.yaml, README Engine: in-process Python (no daemon, no npm), sentence-transformers 384-dim, RRF fusion (k=60), BM25 + vector with numpy fallback. WebSocket API unchanged. Deploy: 'git clone' + 'make init' + 'make pull-models MODELS_SOURCE=...' + 'make up'. Models (5.83 GB) live outside git; pulled via rsync from dev host.
68 lines
2.3 KiB
Python
68 lines
2.3 KiB
Python
"""Ручной smoke-test движка через CLI: ingest → query → exit."""
|
||
import asyncio
|
||
import sys
|
||
import tempfile
|
||
from pathlib import Path
|
||
|
||
ROOT = Path(__file__).resolve().parent.parent
|
||
sys.path.insert(0, str(ROOT))
|
||
|
||
from src.rag.engine import get_or_create_engine # noqa: E402
|
||
|
||
|
||
SAMPLES = {
|
||
"plan.md": (
|
||
"# План 3-го этажа\n\n"
|
||
"План 3-го этажа жилого дома. Оси: А, Б, В, Г. Размеры между осями А и Б: 5400 мм.\n"
|
||
"Квартиры: 301, 302, 303.\n"
|
||
),
|
||
"auth.md": (
|
||
"# Авторизация\n\n"
|
||
"Авторизация работает через JWT-токены с TTL 24 часа.\n"
|
||
"Refresh-токен живёт 30 дней.\n"
|
||
),
|
||
"schedule.md": (
|
||
"# График работ\n\n"
|
||
"Строительство начинается 1 июня 2026. Срок сдачи — 30 ноября 2027.\n"
|
||
),
|
||
}
|
||
|
||
|
||
async def main() -> int:
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
coll = Path(tmp) / "demo" / "qmd_collections" / "demo"
|
||
coll.mkdir(parents=True)
|
||
eng = get_or_create_engine(coll)
|
||
eng.warmup()
|
||
|
||
# Ingest
|
||
for name, text in SAMPLES.items():
|
||
result = eng.index_text(text, source_path=name)
|
||
print(f" + {name}: chunks={result.chunks_indexed} vectors={result.vectors_indexed} skipped={result.skipped}")
|
||
|
||
# Status
|
||
print("\nStatus:", eng.status())
|
||
|
||
# Search BM25
|
||
print('\n--- BM25 "авторизация JWT" ---')
|
||
for h in eng.search("авторизация JWT"):
|
||
print(f" {h.score:.3f} {h.file_path}: {h.snippet(60)}")
|
||
|
||
# Search vector
|
||
print('\n--- Vector "как устроен вход в систему" ---')
|
||
for h in eng.vsearch("как устроен вход в систему"):
|
||
print(f" {h.score:.3f} {h.file_path}: {h.snippet(60)}")
|
||
|
||
# Hybrid query
|
||
print('\n--- Hybrid "когда сдача объекта" ---')
|
||
for h in eng.query("когда сдача объекта", limit=3, use_rerank=False):
|
||
print(f" {h.score:.3f} {h.source_annotation()}: {h.snippet(60)}")
|
||
|
||
eng.close()
|
||
print("\nAll good [OK]")
|
||
return 0
|
||
|
||
|
||
if __name__ == "__main__":
|
||
raise SystemExit(asyncio.run(main()))
|