Create app.py

app.py

@@ -0,0 +1,95 @@
+import gradio as gr
+import pandas as pd
+from pathlib import Path
+
+# --- CONFIGURATION ---
+
+# Path inside the Space repo
+FILE_PATH = Path("EPS_FILES_20K_NOV2025.csv")
+
+print("⏳ Loading data... (This may take 10–20 seconds)")
+
+try:
+    df = pd.read_csv(FILE_PATH, on_bad_lines="skip")
+    df.columns = [c.lower() for c in df.columns]
+
+    text_col = "text" if "text" in df.columns else df.columns[-1]
+    name_col = "filename" if "filename" in df.columns else df.columns[0]
+
+    print(f"✅ Success! Loaded {len(df)} documents.")
+
+except Exception as e:
+    print(f"❌ Error loading file: {e}")
+    df = pd.DataFrame({"error": ["File not found. Check path in app.py and that the CSV is in the repo."]})
+    text_col = "error"
+    name_col = "error"
+
+
+# --- APP LOGIC ---
+
+def search_documents(query):
+    """Filters the dataframe based on the search query."""
+    if not query:
+        return df.head(50)
+
+    mask = df.astype(str).apply(
+        lambda x: x.str.contains(query, case=False, na=False)
+    ).any(axis=1)
+
+    results = df[mask]
+    return results.head(1000)
+
+
+def display_document(evt: gr.SelectData, current_data):
+    """Runs when you click a row in the table."""
+    try:
+        row_index = evt.index[0]
+        full_text = current_data.iloc[row_index][text_col]
+        doc_name = current_data.iloc[row_index][name_col]
+        return f"📄 File: {doc_name}\n\n{full_text}"
+    except Exception as e:
+        return f"Error retrieving document text: {e}"
+
+
+# --- USER INTERFACE ---
+
+with gr.Blocks(title="Epstein Docs Browser", theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# 📂 Epstein Estate Document Browser")
+    gr.Markdown(
+        "Search 20,000+ documents. **Click on any row in the table to read the full file below.**"
+    )
+
+    with gr.Row():
+        search_box = gr.Textbox(
+            label="Search (Keywords, Names, Flight Logs)",
+            placeholder="Type here...",
+            scale=3,
+        )
+        search_btn = gr.Button("Search", variant="primary", scale=1)
+
+    results_table = gr.Dataframe(
+        headers=list(df.columns),
+        datatype="str",
+        label="Search Results (Click a row to view)",
+        interactive=False,
+        wrap=True,
+    )
+
+    doc_viewer = gr.TextArea(
+        label="Document Content",
+        lines=20,
+        placeholder="Select a document above to read it here...",
+    )
+
+    search_btn.click(fn=search_documents, inputs=search_box, outputs=results_table)
+    search_box.submit(fn=search_documents, inputs=search_box, outputs=results_table)
+
+    results_table.select(
+        fn=display_document,
+        inputs=results_table,
+        outputs=doc_viewer,
+    )
+
+# On Spaces, just launch; no share=True
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)