Update ui/ui_core.py

ui/ui_core.py

@@ -3,35 +3,21 @@ import os
 import pandas as pd
 import pdfplumber
 import gradio as gr
-from tabulate import tabulate
-from typing import List, Optional
+from typing import List
 
-# ✅ Fix: Add src to Python path with correct parentheses
+# ✅ Fix: Add src to Python path
 sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "src")))
 
 from txagent.txagent import TxAgent
 
-def safe_extract_table_data(table: List[List[str]]) -> List[str]:
-    extracted_rows = []
-    if not table or not isinstance(table, list):
-        return extracted_rows
-    for row in table:
-        if not row or not isinstance(row, list):
-            continue
-        try:
-            clean_row = [str(cell) if cell is not None else "" for cell in row]
-            if any(clean_row):
-                extracted_rows.append("\t".join(clean_row))
-        except Exception as e:
-            print(f"Error processing table row: {e}")
-            continue
-    return extracted_rows
-
 def extract_all_text_from_csv_or_excel(file_path: str, progress=None, index=0, total=1) -> str:
     try:
         if not os.path.exists(file_path):
             return f"File not found: {file_path}"
 
+        if progress:
+            progress((index + 1) / total, desc=f"Reading spreadsheet: {os.path.basename(file_path)}")
+
         if file_path.endswith(".csv"):
             df = pd.read_csv(file_path, encoding="utf-8", errors="replace", low_memory=False)
         elif file_path.endswith((".xls", ".xlsx")):
@@ -39,62 +25,40 @@ def extract_all_text_from_csv_or_excel(file_path: str, progress=None, index=0, t
         else:
             return f"Unsupported spreadsheet format: {file_path}"
 
-        if progress:
-            progress((index + 1) / total, desc=f"Processed table: {os.path.basename(file_path)}")
-
-        group_column = None
-        for col in ["Booking Number", "Form Name"]:
-            if col in df.columns:
-                group_column = col
-                break
-
-        if group_column:
-            try:
-                groups = df.groupby(group_column)
-                result = []
-                for group_name, group_df in groups:
-                    if group_name is None:
-                        continue
-                    result.append(f"\n### Group: {group_name}\n")
-                    result.append(tabulate(group_df, headers="keys", tablefmt="github", showindex=False))
-                return "\n".join(result) if result else tabulate(df, headers="keys", tablefmt="github", showindex=False)
-            except Exception as e:
-                print(f"Error during grouping: {e}")
-                return tabulate(df, headers="keys", tablefmt="github", showindex=False)
-        else:
-            return tabulate(df, headers="keys", tablefmt="github", showindex=False)
+        lines = []
+        for _, row in df.iterrows():
+            line = " | ".join(str(cell) for cell in row if pd.notna(cell))
+            if line:
+                lines.append(line)
+        return f"\ud83d\udcc4 {os.path.basename(file_path)}\n\n" + "\n".join(lines)
 
     except Exception as e:
-        return f"Error parsing file {os.path.basename(file_path)}: {str(e)}"
+        return f"[Error reading {os.path.basename(file_path)}]: {str(e)}"
 
 def extract_all_text_from_pdf(file_path: str, progress=None, index=0, total=1) -> str:
-    extracted = []
     try:
         if not os.path.exists(file_path):
-            return f"PDF file not found: {file_path}"
+            return f"PDF not found: {file_path}"
 
+        extracted = []
         with pdfplumber.open(file_path) as pdf:
-            num_pages = len(pdf.pages) if hasattr(pdf, 'pages') else 0
-            for i, page in enumerate(pdf.pages if num_pages > 0 else []):
+            num_pages = len(pdf.pages)
+            for i, page in enumerate(pdf.pages):
                 try:
-                    tables = page.extract_tables() if hasattr(page, 'extract_tables') else []
-                    for table in tables if tables else []:
-                        extracted.extend(safe_extract_table_data(table))
-                    
-                    if progress and num_pages > 0:
-                        progress((index + (i / num_pages)) / total, 
-                                desc=f"Parsing PDF: {os.path.basename(file_path)} ({i+1}/{num_pages})")
-                except Exception as page_error:
-                    print(f"Error processing page {i+1}: {page_error}")
-                    continue
-
-        return "\n".join(extracted) if extracted else f"No extractable content found in {os.path.basename(file_path)}"
+                    text = page.extract_text() or ""
+                    extracted.append(text.strip())
+                    if progress:
+                        progress((index + (i / num_pages)) / total, desc=f"Reading PDF: {os.path.basename(file_path)} ({i+1}/{num_pages})")
+                except Exception as e:
+                    extracted.append(f"[Error reading page {i+1}]: {str(e)}")
+        return f"\ud83d\udcc4 {os.path.basename(file_path)}\n\n" + "\n\n".join(extracted)
+
     except Exception as e:
-        return f"Error parsing PDF {os.path.basename(file_path)}: {str(e)}"
+        return f"[Error reading PDF {os.path.basename(file_path)}]: {str(e)}"
 
 def create_ui(agent: TxAgent):
     with gr.Blocks(theme=gr.themes.Soft()) as demo:
-        gr.Markdown("<h1 style='text-align: center;'>📋 CPS: Clinical Patient Support System</h1>")
+        gr.Markdown("<h1 style='text-align: center;'>\ud83d\udccb CPS: Clinical Patient Support System</h1>")
         chatbot = gr.Chatbot(label="CPS Assistant", height=600, type="messages")
 
         file_upload = gr.File(
@@ -123,7 +87,7 @@ def create_ui(agent: TxAgent):
                     for index, file in enumerate(uploaded_files):
                         if not hasattr(file, 'name'):
                             continue
-                            
+
                         path = file.name
                         try:
                             if path.endswith((".csv", ".xls", ".xlsx")):
@@ -132,16 +96,14 @@ def create_ui(agent: TxAgent):
                                 extracted_text += extract_all_text_from_pdf(path, progress, index, total_files) + "\n"
                             else:
                                 extracted_text += f"(Uploaded file: {os.path.basename(path)})\n"
-                                if progress:
-                                    progress((index + 1) / total_files, desc=f"Skipping unsupported file: {os.path.basename(path)}")
                         except Exception as file_error:
-                            print(f"Error processing file {path}: {file_error}")
-                            extracted_text += f"\n[Error processing file: {os.path.basename(path)}]\n"
+                            extracted_text += f"[Error processing file: {os.path.basename(path)}] — {str(file_error)}\n"
                             continue
 
-                    message = f"{context}\n\n---\n{extracted_text.strip()}\n---\n\nBegin your reasoning."
+                    message = (
+                        f"{context}\n\n--- Uploaded File Content ---\n\n{extracted_text.strip()}\n\n--- End of File ---\n\nNow begin your reasoning:"
+                    )
 
-                final_response = None
                 generator = agent.run_gradio_chat(
                     message=message,
                     history=history,
@@ -153,7 +115,7 @@ def create_ui(agent: TxAgent):
                     uploaded_files=uploaded_files,
                     max_round=30
                 )
-                
+
                 for update in generator:
                     try:
                         if isinstance(update, list):
@@ -163,15 +125,13 @@ def create_ui(agent: TxAgent):
                                 and not (
                                     msg.role == "assistant"
                                     and hasattr(msg, 'content')
-                                    and msg.content.strip().startswith("🧰")
+                                    and msg.content.strip().startswith("\ud83e\udde0")
                                 )
                             ]
                             if cleaned:
-                                final_response = cleaned
                                 yield cleaned
-                        else:
-                            if isinstance(update, str) and not update.strip().startswith("🧰"):
-                                yield update.encode("utf-8", "replace").decode("utf-8")
+                        elif isinstance(update, str) and not update.strip().startswith("\ud83e\udde0"):
+                            yield update.encode("utf-8", "replace").decode("utf-8")
                     except Exception as update_error:
                         print(f"Error processing update: {update_error}")
                         continue