380 lines
15 KiB
Python
380 lines
15 KiB
Python
"""Unit tests: importer directory walk + sha256 delta logic.
|
|
|
|
The walk tests are pure filesystem (``tmp_path``); the delta tests run
|
|
against the local compose Postgres (preferred — a real vector table),
|
|
skipping with clear instructions when the stack is not up.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import asyncio
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
from sqlalchemy import func, select
|
|
|
|
from app.models import Chunk, Document
|
|
from app.rag.importer import (
|
|
EXCLUDED_DIRS,
|
|
import_sources,
|
|
iter_importable_files,
|
|
)
|
|
from app.rag.llm import EmbeddingError
|
|
from tests.fakes import FakeEmbedder
|
|
|
|
#: A9 default extension set as dotted suffixes (what the importer passes to
|
|
#: the walker when no override is configured).
|
|
DEFAULT_EXTS = frozenset({".md", ".markdown", ".txt", ".yaml", ".yml", ".json", ".py"})
|
|
|
|
|
|
class _PoisonEmbedder(FakeEmbedder):
|
|
"""Fails (like a real endpoint) on any text containing 'poison'."""
|
|
|
|
async def embed(self, texts: list[str]) -> list[list[float]]:
|
|
if any("poison" in t for t in texts):
|
|
raise EmbeddingError("embeddings endpoint refused the input (simulated)")
|
|
return await super().embed(texts)
|
|
|
|
|
|
class _CapEmbedder(FakeEmbedder):
|
|
"""Simulates the endpoint's ~1024-token input cap at ~1.1 chars/token:
|
|
any single text over 1000 chars is rejected (URL-dense worst case)."""
|
|
|
|
async def embed(self, texts: list[str]) -> list[list[float]]:
|
|
if any(len(t) > 1000 for t in texts):
|
|
raise EmbeddingError(
|
|
"a single 1100-char chunk exceeded the endpoint's per-request "
|
|
"input token cap — lower BOR_CHUNK_TARGET_CHARS and re-import"
|
|
)
|
|
return await super().embed(texts)
|
|
|
|
|
|
def _cleanup_source(db, source: str) -> None:
|
|
for doc in db.scalars(select(Document).where(Document.source == source)).all():
|
|
db.delete(doc)
|
|
db.commit()
|
|
|
|
|
|
def test_iter_importable_files_excludes_noncontent_dirs_and_hidden(tmp_path: Path) -> None:
|
|
"""Well-known non-content dirs, hidden (dot-) dirs/files, and non-A9
|
|
extensions are all skipped; the A9 formats pass."""
|
|
root = tmp_path / "proj"
|
|
for d in (
|
|
"notes/sub",
|
|
".venv/lib",
|
|
"node_modules/x",
|
|
".git",
|
|
"__pycache__",
|
|
".pytest_cache",
|
|
"dist",
|
|
"build",
|
|
".esphome/.espressif", # vendored hidden cache — the real A9 case
|
|
):
|
|
(root / d).mkdir(parents=True)
|
|
files = {
|
|
# content that must be found:
|
|
"README.md": "readme",
|
|
"notes/sub/deep.md": "deep",
|
|
"compose.yaml": "services: {}",
|
|
"legacy.YML": "a: b", # case-insensitive suffix
|
|
"notes/sub/agent.py": "x = 1",
|
|
"config.json": "{}",
|
|
"README.txt": "plain",
|
|
"notes/sub/deep.markdown": "md2",
|
|
# must be skipped:
|
|
".venv/lib/junk.md": "junk",
|
|
"node_modules/x/j.md": "j",
|
|
".git/c.md": "g",
|
|
"__pycache__/c.md": "p",
|
|
".pytest_cache/c.md": "pc",
|
|
"dist/d.md": "d",
|
|
"build/b.md": "b",
|
|
".esphome/.espressif/secret.md": "vendor",
|
|
".secret.md": "hidden file", # dot-prefixed FILE, not just dir
|
|
"notes/sub/notes.csv": "a,b", # not an A9 format
|
|
"notes/sub/file.md.bak": "x",
|
|
}
|
|
for rel, text in files.items():
|
|
(root / rel).write_text(text)
|
|
|
|
found = {p.relative_to(root).as_posix() for p in iter_importable_files(root, DEFAULT_EXTS)}
|
|
assert found == {
|
|
"README.md",
|
|
"notes/sub/deep.md",
|
|
"compose.yaml",
|
|
"legacy.YML",
|
|
"notes/sub/agent.py",
|
|
"config.json",
|
|
"README.txt",
|
|
"notes/sub/deep.markdown",
|
|
}
|
|
|
|
|
|
def test_iter_importable_files_missing_dir_yields_nothing(tmp_path: Path) -> None:
|
|
assert iter_importable_files(tmp_path / "definitely-missing", DEFAULT_EXTS) == []
|
|
|
|
|
|
def test_iter_importable_files_respects_custom_extension_filter(tmp_path: Path) -> None:
|
|
"""A narrower filter (e.g. md only) excludes the other A9 formats."""
|
|
root = tmp_path / "filtered"
|
|
root.mkdir()
|
|
(root / "a.md").write_text("a")
|
|
(root / "b.yaml").write_text("a: b")
|
|
(root / "c.py").write_text("x = 1")
|
|
found = {p.name for p in iter_importable_files(root, frozenset([".md"]))}
|
|
assert found == {"a.md"}
|
|
|
|
|
|
def test_excluded_dirs_match_plan_anchor_a9() -> None:
|
|
assert {
|
|
".venv", "node_modules", ".git", "__pycache__", ".pytest_cache", "dist", "build"
|
|
} == EXCLUDED_DIRS
|
|
|
|
|
|
def test_added_then_unchanged_then_updated_then_pruned(db, tmp_path: Path) -> None:
|
|
root = tmp_path / "src"
|
|
root.mkdir()
|
|
(root / "a.md").write_text("# A\n\nalpha\n\n## Sub\n\nmore alpha\n")
|
|
(root / "b.md").write_text("# B\n\nbeta\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
s1 = asyncio.run(import_sources([root], llm, session=db))
|
|
assert (s1.files, s1.added, s1.unchanged, s1.updated, s1.pruned) == (2, 2, 0, 0, 0)
|
|
# a.md has two sections (2 chunks), b.md one (1 chunk).
|
|
assert s1.chunks == 3
|
|
# Embeddings are stored with the configured dimension.
|
|
n = db.scalar(
|
|
select(func.count())
|
|
.select_from(Chunk)
|
|
.join(Document, Document.id == Chunk.document_id)
|
|
.where(Document.source == root.name)
|
|
)
|
|
assert n == 3
|
|
for c in db.scalars(
|
|
select(Chunk)
|
|
.join(Document, Document.id == Chunk.document_id)
|
|
.where(Document.source == root.name)
|
|
).all():
|
|
assert c.embedding is not None and len(c.embedding) == 768
|
|
|
|
s2 = asyncio.run(import_sources([root], llm, session=db))
|
|
assert s2.added == 0 and s2.unchanged == 2
|
|
|
|
(root / "a.md").write_text("# A\n\nalpha CHANGED\n")
|
|
s3 = asyncio.run(import_sources([root], llm, session=db))
|
|
assert s3.updated == 1 and s3.unchanged == 1
|
|
doc = db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "a.md")
|
|
)
|
|
assert doc is not None and "CHANGED" in doc.content
|
|
|
|
(root / "a.md").unlink()
|
|
s4 = asyncio.run(import_sources([root], llm, session=db, prune=True))
|
|
assert s4.pruned == 1
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "a.md")
|
|
) is None
|
|
# Chunks of the pruned document are gone (FK cascade).
|
|
n_after = db.scalar(
|
|
select(func.count())
|
|
.select_from(Chunk)
|
|
.join(Document, Document.id == Chunk.document_id)
|
|
.where(Document.source == root.name)
|
|
)
|
|
assert n_after == 1
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_embedding_failure_is_logged_and_import_continues(db, tmp_path: Path) -> None:
|
|
"""A file the embedding endpoint refuses must not abort the whole KB:
|
|
its rows are rolled back, the error is counted, and other files import."""
|
|
root = tmp_path / "mixed"
|
|
root.mkdir()
|
|
(root / "bad.md").write_text("# Bad\n\npoison content that the endpoint refuses\n")
|
|
(root / "good.md").write_text("# Good\n\nperfectly fine content\n")
|
|
try:
|
|
summary = asyncio.run(import_sources([root], _PoisonEmbedder(), session=db))
|
|
assert summary.files == 2
|
|
assert summary.errors == 1
|
|
assert summary.added == 1 # only good.md
|
|
# bad.md left no row and no orphan chunks behind (rolled back).
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "bad.md")
|
|
) is None
|
|
assert db.scalar(
|
|
select(func.count())
|
|
.select_from(Chunk)
|
|
.join(Document, Document.id == Chunk.document_id)
|
|
.where(Document.source == root.name, Document.path == "bad.md")
|
|
) == 0
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "good.md")
|
|
) is not None
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_oversized_chunk_triggers_adaptive_rechunk(db, tmp_path: Path) -> None:
|
|
"""A URL-dense paragraph the endpoint rejects must be re-chunked smaller
|
|
for that file only — the import still succeeds."""
|
|
root = tmp_path / "dense"
|
|
root.mkdir()
|
|
# One ~1165-char paragraph: under the 1200-char hard cap, over the
|
|
# simulated token cap. The retry at 600 chars must split it.
|
|
para = "see https://example.com/" + "a" * 1100
|
|
(root / "dense.md").write_text(f"# D\n\n{para}\n")
|
|
try:
|
|
summary = asyncio.run(import_sources([root], _CapEmbedder(), session=db))
|
|
assert summary.errors == 0
|
|
assert summary.added == 1
|
|
doc = db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "dense.md")
|
|
)
|
|
assert doc is not None
|
|
assert len(doc.chunks) >= 2 # re-chunked smaller than the hard cap
|
|
assert all(len(c.content) <= 1000 for c in doc.chunks)
|
|
assert all(c.embedding is not None for c in doc.chunks)
|
|
# The content survives the split.
|
|
assert "".join(c.content for c in doc.chunks).count("a" * 500) >= 1
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_missing_source_dir_is_skipped(db, tmp_path: Path) -> None:
|
|
llm = FakeEmbedder()
|
|
summary = asyncio.run(import_sources([tmp_path / "missing"], llm, session=db))
|
|
assert summary.files == 0 and summary.added == 0
|
|
|
|
|
|
def test_limit_caps_files_and_disables_prune(db, tmp_path: Path) -> None:
|
|
root = tmp_path / "limited"
|
|
root.mkdir()
|
|
for name in ("a.md", "b.md", "c.md"):
|
|
(root / name).write_text(f"# {name}\n\nbody {name}\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
summary = asyncio.run(import_sources([root], llm, limit=2, session=db, prune=True))
|
|
assert summary.files == 2 and summary.added == 2
|
|
# c.md was never walked, so it must NOT be pruned (prune disabled
|
|
# under --limit) — and nothing else disappears either.
|
|
assert summary.pruned == 0
|
|
assert db.scalar(
|
|
select(func.count()).select_from(Document).where(Document.source == root.name)
|
|
) == 2
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_limit_must_be_positive(db, tmp_path: Path) -> None:
|
|
with pytest.raises(ValueError):
|
|
asyncio.run(import_sources([tmp_path], FakeEmbedder(), limit=0, session=db))
|
|
|
|
|
|
def test_prune_is_scoped_to_the_given_sources(db, tmp_path: Path) -> None:
|
|
src_x = tmp_path / "SourceX"
|
|
src_y = tmp_path / "SourceY"
|
|
src_x.mkdir()
|
|
src_y.mkdir()
|
|
(src_x / "x.md").write_text("# X\n\nx body\n")
|
|
(src_y / "y.md").write_text("# Y\n\ny body\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
asyncio.run(import_sources([src_x, src_y], llm, session=db))
|
|
# Re-import ONLY source Y (y.md removed) with prune: source X's doc
|
|
# must survive — prune never touches sources not passed to this run.
|
|
(src_y / "y.md").unlink()
|
|
summary = asyncio.run(import_sources([src_y], llm, session=db, prune=True))
|
|
assert summary.pruned == 1
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == "SourceX", Document.path == "x.md")
|
|
) is not None
|
|
finally:
|
|
_cleanup_source(db, "SourceX")
|
|
_cleanup_source(db, "SourceY")
|
|
|
|
|
|
def test_chunk_positions_and_titles(db, tmp_path: Path) -> None:
|
|
root = tmp_path / "titled"
|
|
root.mkdir()
|
|
(root / "multi.md").write_text("# Real Title\n\n## One\n\na\n\n## Two\n\nb\n")
|
|
(root / "noh1.md").write_text("## Only heading\n\nbody\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
asyncio.run(import_sources([root], llm, session=db))
|
|
titles = {
|
|
d.path: d.title
|
|
for d in db.scalars(select(Document).where(Document.source == root.name)).all()
|
|
}
|
|
assert titles["multi.md"] == "Real Title" # H1 wins
|
|
assert titles["noh1.md"] == "noh1" # …else the file stem
|
|
doc = db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "multi.md")
|
|
)
|
|
assert doc is not None
|
|
positions = sorted(c.position for c in doc.chunks)
|
|
assert positions == list(range(len(doc.chunks))) and len(doc.chunks) >= 2
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_multi_format_import_counts_per_format_and_titles_stem(db, tmp_path: Path) -> None:
|
|
"""A9 formats all import; the summary records per-format counts, and
|
|
non-markdown titles come from the file stem (a ``#`` line is a comment
|
|
there, not a heading)."""
|
|
root = tmp_path / "multi"
|
|
(root / "svc").mkdir(parents=True)
|
|
(root / "guide.md").write_text("# Real Heading\n\nbody\n")
|
|
(root / "svc" / "compose.yaml").write_text("# a comment\nservices:\n gitlab: {}\n")
|
|
(root / "svc" / "agent.py").write_text("# docstring-like comment\ndef ping():\n return 1\n")
|
|
(root / "inventory.json").write_text('{"hosts": []}\n')
|
|
(root / "notes.txt").write_text("plain text notes\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
summary = asyncio.run(import_sources([root], llm, session=db))
|
|
assert summary.files == 5
|
|
assert summary.added == 5
|
|
assert summary.formats == {"md": 1, "yaml": 1, "py": 1, "json": 1, "txt": 1}
|
|
# PLAN §9 summary line: counts, highest first, ext:name pairs.
|
|
assert summary.format_counts() == "json:1,md:1,py:1,txt:1,yaml:1"
|
|
|
|
titles = {
|
|
d.path: d.title
|
|
for d in db.scalars(select(Document).where(Document.source == root.name)).all()
|
|
}
|
|
assert titles["guide.md"] == "Real Heading" # markdown keeps the H1
|
|
assert titles["svc/compose.yaml"] == "compose" # …comment is not a heading
|
|
assert titles["svc/agent.py"] == "agent"
|
|
assert titles["inventory.json"] == "inventory"
|
|
assert titles["notes.txt"] == "notes"
|
|
finally:
|
|
_cleanup_source(db, root.name)
|
|
|
|
|
|
def test_prune_removes_files_now_excluded_by_format_filter(db, tmp_path: Path) -> None:
|
|
"""Previously-imported junk leaves the index: a file that no longer
|
|
matches the A9 extension filter is pruned on the next ``prune=True`` run.
|
|
This is how dot-dir READMEs imported before the scope fix get cleaned up."""
|
|
root = tmp_path / "cleanup"
|
|
root.mkdir()
|
|
(root / "keep.md").write_text("# Keep\n\nkept\n")
|
|
(root / "junk.md.bak").write_text("old junk that was once imported\n")
|
|
llm = FakeEmbedder()
|
|
try:
|
|
# Seed: import both files as if they were valid at the time.
|
|
(root / "junk.md").write_text("old junk that was once imported\n")
|
|
(root / "junk.md.bak").unlink()
|
|
asyncio.run(import_sources([root], llm, session=db))
|
|
# Rename the junk out of the A9 formats, then prune.
|
|
(root / "junk.md").rename(root / "junk.md.bak")
|
|
summary = asyncio.run(import_sources([root], llm, session=db, prune=True))
|
|
assert summary.pruned == 1
|
|
assert summary.unchanged == 1 # keep.md survived
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "junk.md")
|
|
) is None
|
|
assert db.scalar(
|
|
select(Document).where(Document.source == root.name, Document.path == "keep.md")
|
|
) is not None
|
|
finally:
|
|
_cleanup_source(db, root.name)
|