Update app.py

app.py

@@ -1,59 +1,65 @@
-import gradio as gr
-from modules import analysis  # Correct import
-
-# Create the Gradio Blocks UI
-with gr.Blocks(title="Log Intelligence Assistant") as demo:
-    gr.Markdown("## 🔍 Interactive Network Log Assistant")
-
-    with gr.Tab("Inputs 🔽"):
-        with gr.Row():
-            uploaded_files = gr.File(
-                file_types=[".pdf", ".log", ".txt"],
-                file_count="multiple",
-                label="📁 Upload Log Files"
-            )
-            text_input = gr.Textbox(
-                label="Paste Logs Here",
-                lines=15,
-                placeholder="2023-10-05 14:22:11 [Firewall] BLOCKED: 192.168.1.5..."
-            )
-
-        with gr.Row():
-            query_input = gr.Textbox(
-                label="Ask About Logs",
-                placeholder="Find failed login attempts from 192.168.1.5"
-            )
-            quick_query = gr.Dropdown(
-                choices=["", "Detect anomalies", "Show blocked IPs", "Generate summary"],
-                label="Quick Actions"
-            )
-
-        with gr.Accordion("⚙️ Configuration Controls", open=False):
-            temperature = gr.Slider(minimum=0.1, maximum=1.0, value=0.7, label="Creativity (Temperature)")
-            start_time = gr.DateTime(label="Start Time")
-            end_time = gr.DateTime(label="End Time")
-
-    with gr.Tab("Outputs 🔼"):
-        analysis_result = gr.Textbox(label="Analysis Results", lines=10, interactive=False)
-        bar_plot = gr.Plot(label="Event Frequency by Hour")
-        pie_chart = gr.Plot(label="Event Type Distribution")
-        alert_feed = gr.HighlightedText(label="Security Alerts", color_map={
-            "CRITICAL": "red",
-            "WARNING": "orange",
-            "INFO": "blue"
-        })
-
-    with gr.Tab("Export 📤"):
-        export_csv = gr.Button("Save as CSV")
-        export_pdf = gr.Button("Export PDF Report")
-
-    run_button = gr.Button("🔎 Analyze Logs")
-
-    # ✅ Fixed the typo here: analysis not analys
-    run_button.click(
-        fn=analysis.run_analysis,
-        inputs=[uploaded_files, text_input, query_input, quick_query, temperature, start_time, end_time],
-        outputs=[analysis_result, bar_plot, pie_chart, alert_feed]
-    )
-
-demo.launch()
+from langchain.chains import RetrievalQA
+from langchain_community.llms import HuggingFacePipeline
+from transformers import pipeline
+from modules import parser, vectorizer
+from datetime import datetime
+import re
+
+def filter_logs_by_time(logs_text, start_time, end_time):
+    """
+    Filters log lines based on timestamp range.
+    """
+    if not start_time or not end_time:
+        return logs_text  # Skip filtering if not both are set
+
+    start = datetime.fromisoformat(str(start_time))
+    end = datetime.fromisoformat(str(end_time))
+
+    filtered_lines = []
+    for line in logs_text.splitlines():
+        match = re.match(r"(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2})", line)
+        if match:
+            timestamp = datetime.strptime(match.group(1), "%Y-%m-%d %H:%M:%S")
+            if start <= timestamp <= end:
+                filtered_lines.append(line)
+    return "\n".join(filtered_lines)
+
+def run_analysis(uploaded_files, text_input, query, quick_action, temperature, start_time, end_time):
+    logs_text = ""
+
+    # Combine uploaded + pasted logs
+    if uploaded_files:
+        logs_text += parser.parse_uploaded_files(uploaded_files)
+    if text_input:
+        logs_text += "\n" + text_input
+
+    if not logs_text.strip():
+        return "❌ No logs provided.", None, None, None
+
+    # Filter logs based on time range (if provided)
+    logs_text = filter_logs_by_time(logs_text, start_time, end_time)
+
+    # Use either typed query or quick action
+    query_text = query.strip() if query else ""
+    if not query_text and quick_action:
+        query_text = quick_action
+    if not query_text:
+        return "❌ No query or quick action selected.", None, None, None
+
+    # Process logs
+    docs = vectorizer.prepare_documents(logs_text)
+    vectordb = vectorizer.create_vectorstore(docs)
+
+    pipe = pipeline("text-generation", model="gpt2", max_length=512, temperature=temperature)
+    llm = HuggingFacePipeline(pipeline=pipe)
+
+    qa = RetrievalQA.from_chain_type(llm=llm, retriever=vectordb.as_retriever())
+    result = qa.run(query_text)
+
+    # Dummy charts and alerts for testing
+    bar_data = {"Hour": ["14:00", "15:00"], "Count": [8, 4]}
+    pie_data = {"Event Type": ["Blocked", "Scan"], "Count": [8, 4]}
+    alerts = [("CRITICAL", "8 blocked SSH attempts from 192.168.1.5"),
+              ("WARNING", "4 port scanning alerts from 10.0.0.8")]
+
+    return result, bar_data, pie_data, alerts