Update app.py

app.py

@@ -0,0 +1,75 @@
+import gradio as gr
+import pandas as pd
+import pickle
+from fuzzywuzzy import process
+
+# Load your data
+def load_lawyers(filename):
+    return pd.read_csv(filename)
+
+def load_history(filename):
+    with open(filename, "rb") as f:
+        return pickle.load(f)
+
+lawyers_data = load_lawyers('data/lawyers.csv')
+history_data = load_history('data/history.pkl')
+
+# Search function
+def search_solicitor(name):
+    if not name or not isinstance(name, str):
+        return "Invalid input"
+    
+    names_list = lawyers_data['name'].tolist()
+    matches = process.extract(name, names_list, limit=None)
+    threshold = 70
+    matched_names = [(names_list.index(match[0])+1, match[0]) for match in matches if match[1] >= threshold]
+    
+    if matched_names:
+        return pd.DataFrame(matched_names, columns=['idx', 'name'])
+    else:
+        return "No solicitors found."
+
+# Display history function
+def display_history(selected_index):
+    if selected_index is None or not (0 < selected_index <= len(lawyers_data)):
+        return "Invalid selection."
+    
+    solicitor = lawyers_data.iloc[selected_index - 1]
+    name = solicitor['name']
+    working_history_df = history_data[name][0]  # Get the DataFrame for the solicitor
+    
+    if working_history_df.columns.dtype == 'int64':
+        first_row_values = working_history_df.iloc[0]
+        working_history_df.columns = first_row_values
+        working_history_df = working_history_df.drop(index=0)
+    
+    # Clean the DataFrame: drop unwanted columns and replace NaN values
+    working_history_df = working_history_df.drop(columns=[col for col in working_history_df.columns if 'Unnamed' in str(col)], errors='ignore')
+    working_history_df = working_history_df.fillna('')  # Replace NaN with empty strings
+    nan_columns = [col for col in working_history_df.columns if pd.isna(col)]  # Identify columns with NaN as their name
+    working_history_df = working_history_df.drop(columns=nan_columns, errors='ignore')
+    
+    return working_history_df
+
+# Gradio interface
+with gr.Blocks() as iface:
+    gr.Markdown("# Solicitor Search and History Display")
+    
+    with gr.Row():
+        name_input = gr.Textbox(label="Enter Solicitor's Name")
+        search_button = gr.Button("Search Solicitor")
+    
+    search_results = gr.Dataframe(label="Search Results")
+    
+    with gr.Row():
+        index_input = gr.Number(label="Select Solicitor by Number", precision=0)
+        history_button = gr.Button("Display Working History")
+    
+    history_results = gr.Dataframe(label="Working History")
+
+    # Define button actions
+    search_button.click(search_solicitor, inputs=name_input, outputs=search_results)
+    history_button.click(display_history, inputs=index_input, outputs=history_results)
+
+if __name__ == "__main__":
+    iface.launch()