Spaces:
Build error
Build error
talexm commited on
Commit ·
759c15a
1
Parent(s): 3ec398b
adjustments
Browse files
app.py
CHANGED
|
@@ -4,34 +4,40 @@ from pathlib import Path
|
|
| 4 |
from PIL import Image
|
| 5 |
from rag_sec.document_search_system import DocumentSearchSystem
|
| 6 |
from chainguard.blockchain_logger import BlockchainLogger
|
| 7 |
-
from rag_sec.document_search_system import main
|
| 8 |
-
import streamlit as st
|
| 9 |
-
from rag_sec.document_retriver import DocumentRetriever
|
| 10 |
|
| 11 |
# Blockchain Logger
|
| 12 |
blockchain_logger = BlockchainLogger()
|
| 13 |
|
| 14 |
# Initialize DocumentSearchSystem
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
|
| 25 |
# Directory for storing uploaded files
|
| 26 |
UPLOAD_DIR = "uploaded_files"
|
| 27 |
os.makedirs(UPLOAD_DIR, exist_ok=True)
|
| 28 |
|
| 29 |
# Streamlit Layout
|
| 30 |
-
st.title("Memora: Secure File Upload with Blockchain & Neo4j")
|
| 31 |
st.subheader("Securely upload, organize, and query your files")
|
| 32 |
|
| 33 |
-
# File Upload
|
| 34 |
-
uploaded_files = st.file_uploader(
|
|
|
|
|
|
|
| 35 |
|
| 36 |
if uploaded_files:
|
| 37 |
for uploaded_file in uploaded_files:
|
|
@@ -41,7 +47,7 @@ if uploaded_files:
|
|
| 41 |
f.write(uploaded_file.getbuffer())
|
| 42 |
st.success(f"File saved locally: {file_path}")
|
| 43 |
|
| 44 |
-
# Display file details
|
| 45 |
if uploaded_file.type.startswith('image'):
|
| 46 |
image = Image.open(uploaded_file)
|
| 47 |
st.image(image, caption=uploaded_file.name, use_column_width=True)
|
|
@@ -50,14 +56,14 @@ if uploaded_files:
|
|
| 50 |
album = st.text_input(f"Album for {uploaded_file.name}", "Default Album")
|
| 51 |
tags = st.text_input(f"Tags for {uploaded_file.name} (comma-separated)", "")
|
| 52 |
|
| 53 |
-
# Log Metadata
|
| 54 |
if st.button(f"Log Metadata for {uploaded_file.name}"):
|
| 55 |
metadata = {"file_name": uploaded_file.name, "tags": tags.split(','), "album": album}
|
| 56 |
blockchain_details = blockchain_logger.log_data(metadata)
|
| 57 |
blockchain_hash = blockchain_details.get("block_hash", "N/A")
|
| 58 |
|
| 59 |
-
# Use Neo4jHandler from DocumentSearchSystem to log transaction
|
| 60 |
-
system.neo4j_handler.
|
| 61 |
st.write(f"Metadata logged successfully! Blockchain Details: {blockchain_details}")
|
| 62 |
|
| 63 |
# Blockchain Integrity Validation
|
|
@@ -65,45 +71,33 @@ if st.button("Validate Blockchain Integrity"):
|
|
| 65 |
is_valid = blockchain_logger.is_blockchain_valid()
|
| 66 |
st.write("Blockchain Integrity:", "Valid ✅" if is_valid else "Invalid ❌")
|
| 67 |
|
| 68 |
-
#
|
| 69 |
-
st.subheader("
|
| 70 |
-
|
| 71 |
-
# Initialize DocumentRetriever
|
| 72 |
-
retriever = DocumentRetriever()
|
| 73 |
-
|
| 74 |
-
@st.cache_resource
|
| 75 |
-
def load_retriever():
|
| 76 |
-
"""Load documents into the retriever."""
|
| 77 |
-
retriever.load_documents()
|
| 78 |
-
return retriever
|
| 79 |
-
# Load the retriever and documents
|
| 80 |
-
st.write("Loading documents...")
|
| 81 |
-
retriever = load_retriever()
|
| 82 |
-
st.write("Documents successfully loaded!")
|
| 83 |
-
|
| 84 |
-
# Streamlit UI
|
| 85 |
-
st.title("Document Search App")
|
| 86 |
-
st.subheader("Enter a query to search for related documents")
|
| 87 |
|
| 88 |
# Query Input
|
| 89 |
query = st.text_input("Enter your query (e.g., 'sports news', 'machine learning')")
|
| 90 |
|
| 91 |
if st.button("Search"):
|
| 92 |
if query:
|
| 93 |
-
#
|
| 94 |
-
|
| 95 |
-
if
|
| 96 |
-
st.
|
| 97 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
st.warning("No relevant documents found for your query.")
|
| 99 |
-
|
| 100 |
-
st.
|
| 101 |
-
for idx, doc in enumerate(results, start=1):
|
| 102 |
-
st.write(f"### Document {idx}")
|
| 103 |
-
st.write(doc[:500]) # Display first 500 characters of each document
|
| 104 |
else:
|
| 105 |
st.warning("Please enter a query to search.")
|
| 106 |
|
| 107 |
# Debugging Section
|
| 108 |
if st.checkbox("Show Debug Information"):
|
| 109 |
-
st.write(f"Total documents loaded: {len(retriever.documents)}")
|
|
|
|
| 4 |
from PIL import Image
|
| 5 |
from rag_sec.document_search_system import DocumentSearchSystem
|
| 6 |
from chainguard.blockchain_logger import BlockchainLogger
|
|
|
|
|
|
|
|
|
|
| 7 |
|
| 8 |
# Blockchain Logger
|
| 9 |
blockchain_logger = BlockchainLogger()
|
| 10 |
|
| 11 |
# Initialize DocumentSearchSystem
|
| 12 |
+
@st.cache_resource
|
| 13 |
+
def initialize_system():
|
| 14 |
+
"""Initialize the DocumentSearchSystem and load documents."""
|
| 15 |
+
home_dir = Path(os.getenv("HOME", "/"))
|
| 16 |
+
system = DocumentSearchSystem(
|
| 17 |
+
neo4j_uri="neo4j+s://0ca71b10.databases.neo4j.io",
|
| 18 |
+
neo4j_user="neo4j",
|
| 19 |
+
neo4j_password="HwGDOxyGS1-79nLeTiX5bx5ohoFSpvHCmTv8IRgt-lY"
|
| 20 |
+
)
|
| 21 |
+
system.retriever.load_documents()
|
| 22 |
+
return system
|
| 23 |
+
|
| 24 |
+
# Initialize and load system
|
| 25 |
+
st.write("Initializing the Document Search System...")
|
| 26 |
+
system = initialize_system()
|
| 27 |
+
st.success("System initialized and documents loaded!")
|
| 28 |
|
| 29 |
# Directory for storing uploaded files
|
| 30 |
UPLOAD_DIR = "uploaded_files"
|
| 31 |
os.makedirs(UPLOAD_DIR, exist_ok=True)
|
| 32 |
|
| 33 |
# Streamlit Layout
|
| 34 |
+
st.title("Memora: Secure File Upload and Search with Blockchain & Neo4j")
|
| 35 |
st.subheader("Securely upload, organize, and query your files")
|
| 36 |
|
| 37 |
+
# File Upload Section
|
| 38 |
+
uploaded_files = st.file_uploader(
|
| 39 |
+
"Upload your files", accept_multiple_files=True, type=['jpg', 'jpeg', 'png', 'mp4', 'avi']
|
| 40 |
+
)
|
| 41 |
|
| 42 |
if uploaded_files:
|
| 43 |
for uploaded_file in uploaded_files:
|
|
|
|
| 47 |
f.write(uploaded_file.getbuffer())
|
| 48 |
st.success(f"File saved locally: {file_path}")
|
| 49 |
|
| 50 |
+
# Display uploaded file details
|
| 51 |
if uploaded_file.type.startswith('image'):
|
| 52 |
image = Image.open(uploaded_file)
|
| 53 |
st.image(image, caption=uploaded_file.name, use_column_width=True)
|
|
|
|
| 56 |
album = st.text_input(f"Album for {uploaded_file.name}", "Default Album")
|
| 57 |
tags = st.text_input(f"Tags for {uploaded_file.name} (comma-separated)", "")
|
| 58 |
|
| 59 |
+
# Log Metadata and Transaction
|
| 60 |
if st.button(f"Log Metadata for {uploaded_file.name}"):
|
| 61 |
metadata = {"file_name": uploaded_file.name, "tags": tags.split(','), "album": album}
|
| 62 |
blockchain_details = blockchain_logger.log_data(metadata)
|
| 63 |
blockchain_hash = blockchain_details.get("block_hash", "N/A")
|
| 64 |
|
| 65 |
+
# Use Neo4jHandler from DocumentSearchSystem to log the transaction
|
| 66 |
+
system.neo4j_handler.log_relationships(uploaded_file.name, tags, blockchain_hash, [album])
|
| 67 |
st.write(f"Metadata logged successfully! Blockchain Details: {blockchain_details}")
|
| 68 |
|
| 69 |
# Blockchain Integrity Validation
|
|
|
|
| 71 |
is_valid = blockchain_logger.is_blockchain_valid()
|
| 72 |
st.write("Blockchain Integrity:", "Valid ✅" if is_valid else "Invalid ❌")
|
| 73 |
|
| 74 |
+
# Document Search Section
|
| 75 |
+
st.subheader("Search Documents")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
# Query Input
|
| 78 |
query = st.text_input("Enter your query (e.g., 'sports news', 'machine learning')")
|
| 79 |
|
| 80 |
if st.button("Search"):
|
| 81 |
if query:
|
| 82 |
+
# Process query through the DocumentSearchSystem
|
| 83 |
+
result = system.process_query(query)
|
| 84 |
+
if result["status"] == "success":
|
| 85 |
+
st.success(f"Query processed successfully!")
|
| 86 |
+
st.write("### Query Response:")
|
| 87 |
+
st.write(result["response"])
|
| 88 |
+
st.write("### Retrieved Documents:")
|
| 89 |
+
for idx, doc in enumerate(result["retrieved_documents"], start=1):
|
| 90 |
+
st.write(f"**Document {idx}:**")
|
| 91 |
+
st.write(doc[:500]) # Display the first 500 characters
|
| 92 |
+
st.write("### Blockchain Details:")
|
| 93 |
+
st.json(result["blockchain_details"])
|
| 94 |
+
elif result["status"] == "no_results":
|
| 95 |
st.warning("No relevant documents found for your query.")
|
| 96 |
+
elif result["status"] == "rejected":
|
| 97 |
+
st.error(result["message"])
|
|
|
|
|
|
|
|
|
|
| 98 |
else:
|
| 99 |
st.warning("Please enter a query to search.")
|
| 100 |
|
| 101 |
# Debugging Section
|
| 102 |
if st.checkbox("Show Debug Information"):
|
| 103 |
+
st.write(f"Total documents loaded: {len(system.retriever.documents)}")
|