mirror of
https://github.com/lllin000/PaperForge.git
synced 2026-07-22 06:50:53 +00:00
66c2db427e
schema.py: CREATE_VEC_FULLTEXT_META was referenced at line 328 but never defined. Adds the missing DDL constant so the v5 schema migration does not raise NameError on fresh databases. tests: add embedding test suite with four tests covering vec0 resume detection, empty-table fallback, force drop, and delete-then-write ordering verification.
42 lines
1.3 KiB
Python
42 lines
1.3 KiB
Python
"""Seed a minimal vault with memory DB and vec0 meta rows for embed resume tests."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
|
|
from paperforge.memory.db import ensure_vec_extension, get_connection, get_memory_db_path
|
|
from paperforge.memory.schema import ensure_schema
|
|
|
|
|
|
@pytest.fixture
|
|
def seeded_vault(tmp_path: Path) -> Path:
|
|
"""Build a minimal vault with memory DB populated with fake vec0 meta rows."""
|
|
vault = tmp_path / "testvault"
|
|
vault.mkdir(parents=True)
|
|
|
|
# Minimal config so paperforge paths resolve
|
|
(vault / "System" / "PaperForge").mkdir(parents=True)
|
|
cfg = {
|
|
"exporter_dir": "System/PaperForge/exporter",
|
|
"zotero_storage": "System/PaperForge/storage",
|
|
"memory_db": "System/PaperForge/memory/paperforge.db",
|
|
}
|
|
(vault / "paperforge.json").write_text(json.dumps(cfg), encoding="utf-8")
|
|
|
|
# Create memory DB with vec0 schema
|
|
db_path = get_memory_db_path(vault)
|
|
conn = get_connection(db_path)
|
|
ensure_vec_extension(conn)
|
|
ensure_schema(conn)
|
|
|
|
# Insert a fake row into vec_fulltext_meta so resume finds "existing data"
|
|
conn.execute(
|
|
"INSERT INTO vec_fulltext_meta(rowid, paper_id, chunk_index, text) VALUES (1, 'test_paper', 0, 'hello world')"
|
|
)
|
|
conn.commit()
|
|
conn.close()
|
|
|
|
return vault
|