File size: 9,716 Bytes
a54fd97 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 | """
Vector Store - Multi-View Indexing (Section 3.1)
Implements three-layer indexing I(m_k):
- Semantic Layer: s_k = E_dense(m_k) - Dense vector similarity
- Lexical Layer: l_k = E_sparse(m_k) - BM25 keyword matching (Tantivy FTS)
- Symbolic Layer: r_k = E_sym(m_k) - Metadata filtering (SQL)
"""
from typing import List, Optional, Dict, Any
import lancedb
import pyarrow as pa
from models.memory_entry import MemoryEntry
from utils.embedding import EmbeddingModel
import config
import os
import re
class VectorStore:
"""
Multi-View Indexing - Storage and retrieval for memory units (Section 3.1)
Three-layer indexing I(m_k):
1. Semantic Layer: Dense embeddings for conceptual similarity
2. Lexical Layer: Tantivy FTS for exact keyword matching
3. Symbolic Layer: SQL-based metadata filtering
"""
def __init__(
self,
db_path: str = None,
embedding_model: EmbeddingModel = None,
table_name: str = None,
storage_options: Optional[Dict[str, Any]] = None
):
self.db_path = db_path or config.LANCEDB_PATH
self.embedding_model = embedding_model or EmbeddingModel()
self.table_name = table_name or config.MEMORY_TABLE_NAME
self.table = None
self._fts_initialized = False
# Detect if using cloud storage (GCS, S3, Azure)
self._is_cloud_storage = self.db_path.startswith(("gs://", "s3://", "az://"))
# Connect to database
if self._is_cloud_storage:
self.db = lancedb.connect(self.db_path, storage_options=storage_options)
else:
os.makedirs(self.db_path, exist_ok=True)
self.db = lancedb.connect(self.db_path)
self._init_table()
def _init_table(self):
"""Initialize table schema and FTS index."""
schema = pa.schema([
pa.field("entry_id", pa.string()),
pa.field("lossless_restatement", pa.string()),
pa.field("keywords", pa.list_(pa.string())),
pa.field("timestamp", pa.string()),
pa.field("location", pa.string()),
pa.field("persons", pa.list_(pa.string())),
pa.field("entities", pa.list_(pa.string())),
pa.field("topic", pa.string()),
pa.field("vector", pa.list_(pa.float32(), self.embedding_model.dimension))
])
if self.table_name not in self.db.table_names():
self.table = self.db.create_table(self.table_name, schema=schema)
print(f"Created new table: {self.table_name}")
else:
self.table = self.db.open_table(self.table_name)
print(f"Opened existing table: {self.table_name}")
def _init_fts_index(self):
"""Initialize Full-Text Search index on lossless_restatement column."""
if self._fts_initialized:
return
try:
if self._is_cloud_storage:
# Use native FTS for cloud storage (Tantivy only works with local filesystem)
self.table.create_fts_index(
"lossless_restatement",
use_tantivy=False,
replace=True
)
print("FTS index created (native mode for cloud storage)")
else:
# Use Tantivy FTS for local storage (better performance)
self.table.create_fts_index(
"lossless_restatement",
use_tantivy=True,
tokenizer_name="en_stem",
replace=True
)
print("FTS index created (Tantivy mode)")
self._fts_initialized = True
except Exception as e:
print(f"FTS index creation skipped: {e}")
def _results_to_entries(self, results: List[dict]) -> List[MemoryEntry]:
"""Convert LanceDB results to MemoryEntry objects."""
entries = []
for r in results:
try:
entries.append(MemoryEntry(
entry_id=r["entry_id"],
lossless_restatement=r["lossless_restatement"],
keywords=list(r.get("keywords") or []),
timestamp=r.get("timestamp") or None,
location=r.get("location") or None,
persons=list(r.get("persons") or []),
entities=list(r.get("entities") or []),
topic=r.get("topic") or None
))
except Exception as e:
print(f"Warning: Failed to parse result: {e}")
continue
return entries
def add_entries(self, entries: List[MemoryEntry]):
"""Batch add memory entries."""
if not entries:
return
restatements = [entry.lossless_restatement for entry in entries]
vectors = self.embedding_model.encode_documents(restatements)
data = []
for entry, vector in zip(entries, vectors):
data.append({
"entry_id": entry.entry_id,
"lossless_restatement": entry.lossless_restatement,
"keywords": entry.keywords,
"timestamp": entry.timestamp or "",
"location": entry.location or "",
"persons": entry.persons,
"entities": entry.entities,
"topic": entry.topic or "",
"vector": vector.tolist()
})
self.table.add(data)
print(f"Added {len(entries)} memory entries")
# Initialize FTS index after first data insertion
if not self._fts_initialized:
self._init_fts_index()
def semantic_search(self, query: str, top_k: int = 5) -> List[MemoryEntry]:
"""
Semantic Layer Search - Dense vector similarity (Section 3.1)
s_k = E_dense(m_k)
"""
try:
if self.table.count_rows() == 0:
return []
query_vector = self.embedding_model.encode_single(query, is_query=True)
results = self.table.search(query_vector.tolist()).limit(top_k).to_list()
return self._results_to_entries(results)
except Exception as e:
print(f"Error during semantic search: {e}")
return []
def keyword_search(self, keywords: List[str], top_k: int = 3) -> List[MemoryEntry]:
"""
Lexical Layer Search - BM25 keyword matching (Section 3.1)
l_k = E_sparse(m_k)
"""
try:
if not keywords or self.table.count_rows() == 0:
return []
# LanceDB auto-detects string input as FTS query when FTS index exists
query = " ".join(keywords)
try:
results = self.table.search(query).limit(top_k).to_list()
except Exception:
# Tantivy-style query syntax is brittle for raw benchmark text;
# fall back to a whitespace query made of plain word tokens.
safe_terms = re.findall(r"\w+", query)
if not safe_terms:
return []
safe_query = " ".join(safe_terms)
results = self.table.search(safe_query).limit(top_k).to_list()
return self._results_to_entries(results)
except Exception as e:
print(f"Error during keyword search: {e}")
return []
def structured_search(
self,
persons: Optional[List[str]] = None,
timestamp_range: Optional[tuple] = None,
location: Optional[str] = None,
entities: Optional[List[str]] = None,
top_k: Optional[int] = None
) -> List[MemoryEntry]:
"""
Symbolic Layer Search - Metadata filtering (Section 3.1)
r_k = E_sym(m_k), filters by timestamps, entities, persons
"""
try:
if self.table.count_rows() == 0:
return []
if not any([persons, timestamp_range, location, entities]):
return []
conditions = []
if persons:
values = ", ".join([f"'{p}'" for p in persons])
conditions.append(f"array_has_any(persons, make_array({values}))")
if location:
safe_location = location.replace("'", "''")
conditions.append(f"location LIKE '%{safe_location}%'")
if entities:
values = ", ".join([f"'{e}'" for e in entities])
conditions.append(f"array_has_any(entities, make_array({values}))")
if timestamp_range:
start_time, end_time = timestamp_range
conditions.append(f"timestamp >= '{start_time}' AND timestamp <= '{end_time}'")
where_clause = " AND ".join(conditions)
query = self.table.search().where(where_clause, prefilter=True)
if top_k:
query = query.limit(top_k)
results = query.to_list()
return self._results_to_entries(results)
except Exception as e:
print(f"Error during structured search: {e}")
return []
def get_all_entries(self) -> List[MemoryEntry]:
"""Get all memory entries."""
results = self.table.to_arrow().to_pylist()
return self._results_to_entries(results)
def optimize(self):
"""Optimize table after bulk insertions for better query performance."""
self.table.optimize()
print("Table optimized")
def clear(self):
"""Clear all data and reinitialize table."""
self.db.drop_table(self.table_name)
self._fts_initialized = False
self._init_table()
print("Database cleared")
|