c284d86885
An image query has no text to match, so it is vector-only whatever the caller asked for, and full-text search embeds nothing, so it needs no agreement on embedders.
88 lines
2.8 KiB
Python
88 lines
2.8 KiB
Python
import pytest
|
|
|
|
from haiku.rag.client import HaikuRAG
|
|
from haiku.rag.client.search import _rank
|
|
from haiku.rag.store.models import Chunk
|
|
|
|
|
|
@pytest.fixture
|
|
def exploding_reranker(monkeypatch):
|
|
"""A reranker that cannot be built, so any access fails loudly."""
|
|
|
|
def boom(self):
|
|
raise AssertionError("reranker built for a query that cannot use it")
|
|
|
|
monkeypatch.setattr(HaikuRAG, "reranker", property(boom), raising=True)
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_image_ranking_never_builds_the_reranker(
|
|
temp_db_path, exploding_reranker
|
|
):
|
|
"""Local rerankers load model weights on construction, so an image query,
|
|
which has no text to score against, must not touch one."""
|
|
async with HaikuRAG(temp_db_path, create=True) as rag:
|
|
candidates = [
|
|
(Chunk(id="a", document_id="d", content="one"), 0.9),
|
|
(Chunk(id="b", document_id="d", content="two"), 0.8),
|
|
]
|
|
|
|
ranked = await _rank(rag, b"image-bytes", candidates, limit=1)
|
|
|
|
assert [c.id for c, _ in ranked] == ["a"]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_image_fetch_never_builds_the_reranker(
|
|
temp_db_path, exploding_reranker, monkeypatch
|
|
):
|
|
async with HaikuRAG(temp_db_path, create=True) as rag:
|
|
seen = {}
|
|
|
|
async def fake_search(
|
|
query, limit, search_type="hybrid", filter=None, query_vector=None
|
|
):
|
|
seen["limit"] = limit
|
|
return []
|
|
|
|
async def fake_embed_image(self, image):
|
|
return [0.1] * 8
|
|
|
|
monkeypatch.setattr(rag.chunk_repository, "search", fake_search)
|
|
monkeypatch.setattr(type(rag.embedder), "embed_image", fake_embed_image)
|
|
monkeypatch.setattr(
|
|
type(rag.embedder), "supports_images", property(lambda self: True)
|
|
)
|
|
|
|
await rag.search(b"image-bytes", limit=5)
|
|
|
|
# No over-fetch: nothing will re-rank these.
|
|
assert seen["limit"] == 5
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_an_image_query_ignores_a_full_text_search_type(
|
|
temp_db_path, monkeypatch
|
|
):
|
|
"""An image has no text to match, so it searches by vector whatever the
|
|
caller asked for."""
|
|
async with HaikuRAG(temp_db_path, create=True) as rag:
|
|
seen = {}
|
|
|
|
async def fake_search(query, limit, search_type, filter, query_vector):
|
|
seen.update({"search_type": search_type, "query_vector": query_vector})
|
|
return []
|
|
|
|
async def fake_embed_image(self, image):
|
|
return [0.1] * 8
|
|
|
|
monkeypatch.setattr(rag.chunk_repository, "search", fake_search)
|
|
monkeypatch.setattr(type(rag.embedder), "embed_image", fake_embed_image)
|
|
monkeypatch.setattr(
|
|
type(rag.embedder), "supports_images", property(lambda self: True)
|
|
)
|
|
|
|
await rag.search(b"image-bytes", search_type="fts")
|
|
|
|
assert seen["search_type"] == "vector"
|
|
assert seen["query_vector"] == [0.1] * 8
|