Spaces:
Sleeping
Sleeping
File size: 7,095 Bytes
d19fdec 25c85ff d19fdec 25c85ff d19fdec 99da31c d19fdec 99da31c d19fdec 99da31c d19fdec 99da31c d19fdec | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 | """
SQLiteDocumentRepository for encrypted chunk storage.
Design decisions:
- Fernet encryption/decryption happens in this repository layer.
- Thread safety: a per-instance Lock guards every sqlite3 call. Connections are opened
per-operation (safest pattern for multi-threaded use without a connection pool).
- Schema uses INSERT OR REPLACE (UPSERT) so ingest remains idempotent.
"""
import json
import logging
import sqlite3
import threading
from src.core.exceptions import InfrastructureError
logger = logging.getLogger(__name__)
_SCHEMA = """
CREATE TABLE IF NOT EXISTS document_chunks (
chunk_id TEXT PRIMARY KEY,
encrypted_text TEXT NOT NULL,
source TEXT DEFAULT 'unknown',
chunk_index INTEGER DEFAULT 0,
extra_metadata TEXT DEFAULT '{}'
)
"""
class SQLiteDocumentRepository:
def __init__(self, db_path: str, encryption_manager):
self.db_path = db_path
self.encryption_manager = encryption_manager
self._lock = threading.Lock()
self._init_db()
# ------------------------------------------------------------------
# Internal helpers
# ------------------------------------------------------------------
def _init_db(self) -> None:
try:
with self._lock, sqlite3.connect(self.db_path) as conn:
conn.execute(_SCHEMA)
conn.commit()
logger.info("SQLite document store ready at '%s'.", self.db_path)
except Exception as e:
raise InfrastructureError(f"SQLite initialization failed: {e}") from e
def _conn(self) -> sqlite3.Connection:
conn = sqlite3.connect(self.db_path)
conn.row_factory = sqlite3.Row
return conn
# ------------------------------------------------------------------
# Public interface (mirrors DocumentRepository)
# ------------------------------------------------------------------
def save_document_chunk(self, chunk_id: str, original_text: str, metadata: dict) -> bool:
"""Encrypt and upsert chunk payload into SQLite."""
try:
encrypted_text = self.encryption_manager.encrypt_data(original_text)
source = metadata.get("source", "unknown")
chunk_index = metadata.get("chunk_index", 0)
extra = {
k: v
for k, v in metadata.items()
if k not in {
"chunk_id", "source", "chunk_index",
"encrypted_text", "decrypted_text", "text",
}
and v is not None
}
with self._lock, self._conn() as conn:
conn.execute(
"""
INSERT INTO document_chunks
(chunk_id, encrypted_text, source, chunk_index, extra_metadata)
VALUES (?, ?, ?, ?, ?)
ON CONFLICT(chunk_id) DO UPDATE SET
encrypted_text = excluded.encrypted_text,
source = excluded.source,
chunk_index = excluded.chunk_index,
extra_metadata = excluded.extra_metadata
""",
(chunk_id, encrypted_text, source, chunk_index, json.dumps(extra)),
)
conn.commit()
logger.info("Chunk '%s' encrypted and saved.", chunk_id)
return True
except Exception as e:
logger.error("Chunk save failed (%s): %s", chunk_id, e)
return False
def get_document_chunks(self, chunk_ids: list[str]) -> list[dict]:
if not chunk_ids:
return []
placeholders = ",".join("?" for _ in chunk_ids)
try:
with self._lock, self._conn() as conn:
rows = conn.execute(
f"SELECT * FROM document_chunks WHERE chunk_id IN ({placeholders})",
chunk_ids,
).fetchall()
by_id = {}
for row in rows:
result = dict(row)
result["decrypted_text"] = self.encryption_manager.decrypt_data(
result.pop("encrypted_text")
)
extra = json.loads(result.pop("extra_metadata", "{}"))
result.update(extra)
by_id[result["chunk_id"]] = result
return [by_id[cid] for cid in chunk_ids if cid in by_id]
except Exception as e:
logger.error("Batch chunk read failed: %s", e)
raise InfrastructureError(f"SQLite batch read failed: {e}") from e
def has_document_chunk(self, chunk_id: str) -> bool:
try:
with self._lock, self._conn() as conn:
row = conn.execute(
"SELECT 1 FROM document_chunks WHERE chunk_id = ? LIMIT 1", (chunk_id,)
).fetchone()
return row is not None
except Exception as e:
logger.error("Chunk existence check failed (%s): %s", chunk_id, e)
return False
def delete_document_chunk(self, chunk_id: str) -> None:
try:
with self._lock, self._conn() as conn:
conn.execute(
"DELETE FROM document_chunks WHERE chunk_id = ?", (chunk_id,)
)
conn.commit()
except Exception as e:
raise InfrastructureError(f"SQLite delete failed for {chunk_id}: {e}") from e
def delete_chunk_ids(self, chunk_ids: list[str]) -> None:
if not chunk_ids:
return
placeholders = ",".join("?" for _ in chunk_ids)
try:
with self._lock, self._conn() as conn:
conn.execute(
f"DELETE FROM document_chunks WHERE chunk_id IN ({placeholders})",
chunk_ids,
)
conn.commit()
except Exception as e:
raise InfrastructureError(f"SQLite bulk delete failed: {e}") from e
def delete_all_chunks(self) -> None:
try:
with self._lock, self._conn() as conn:
conn.execute("DELETE FROM document_chunks")
conn.commit()
except Exception as e:
raise InfrastructureError(f"SQLite full reset failed: {e}") from e
def list_chunk_ids(self) -> set[str]:
try:
with self._lock, self._conn() as conn:
rows = conn.execute(
"SELECT chunk_id FROM document_chunks"
).fetchall()
return {row[0] for row in rows}
except Exception as e:
raise InfrastructureError(f"SQLite list IDs failed: {e}") from e
def ping(self) -> bool:
"""Lightweight liveness check — verifies the DB file is accessible and schema exists."""
try:
with self._lock, self._conn() as conn:
conn.execute("SELECT 1 FROM document_chunks LIMIT 1")
return True
except Exception as e:
logger.error("SQLite ping failed: %s", e)
return False
|