Anirudh1950's picture
Publish project from GitHub
1bd8853 verified
Raw History Blame Contribute Delete
3.38 kB
import chromadb
import uuid
from typing import List, Dict, Any
class MemoryManager:
def __init__(self, db_path: str = "./chroma_db", collection_name: str = "scraper_memory"):
print(f"Initializing Long-Term Memory (ChromaDB) at '{db_path}'...")
try:
self.client = chromadb.PersistentClient(path=db_path)
self.collection = self.client.get_or_create_collection(name=collection_name)
except Exception as e:
print(f"Error initializing memory: {e}")
self.client = None
self.collection = None
def store(self, items: List[Dict[str, Any]], topic: str):
"""
Stores scraped items into Long-Term Memory.
"""
if not self.collection or not items:
return
documents = []
metadatas = []
ids = []
print(f"Memorizing {len(items)} items for topic '{topic}'...")
for item in items:
# Create a rich text representation for embedding
content = f"{item.get('title', '')}\n{item.get('text', '') or item.get('snippet', '')}\n{item.get('content', '')}"
if not content.strip():
continue
documents.append(content)
# Metadata must be simple types
meta = {
"source": item.get('source', 'unknown'),
"url": item.get('url') or item.get('link') or "none",
"topic": topic,
"timestamp": str(item.get("timestamp", ""))
}
metadatas.append(meta)
# Unique ID
ids.append(f"{topic}_{str(uuid.uuid4())[:8]}")
try:
self.collection.add(
documents=documents,
metadatas=metadatas,
ids=ids
)
print(" [Memory] Successfully stored items.")
except Exception as e:
print(f" [Memory] Error storing items: {e}")
def recall(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
"""
Recalls relevant information from memory.
"""
if not self.collection:
return []
print(f"Recalling memories for: '{query}'...")
try:
results = self.collection.query(
query_texts=[query],
n_results=n_results
)
# Format results back to list of dicts
recalled_items = []
if results['documents']:
for i, doc in enumerate(results['documents'][0]):
meta = results['metadatas'][0][i]
recalled_items.append({
"content": doc,
"source": meta.get("source"),
"url": meta.get("url"),
"memory_type": "long_term_recall"
})
if recalled_items:
print(f" [Memory] Recalled {len(recalled_items)} past facts.")
else:
print(" [Memory] No relevant memories found.")
return recalled_items
except Exception as e:
print(f" [Memory] Error recalling: {e}")
return []