Create utils.py

utils.py

@@ -0,0 +1,103 @@
+# /utils.py
+
+"""
+A collection of utility functions for data manipulation and formatting.
+
+This module provides helpers for tasks like converting chat history formats,
+processing images for multimodal models, cleaning model outputs, and
+applying code modifications.
+"""
+import base64
+import io
+import re
+from typing import Dict, List, Optional, Tuple
+
+import numpy as np
+from PIL import Image
+
+from config import SEARCH_START, DIVIDER, REPLACE_END, GRADIO_SUPPORTED_LANGUAGES
+
+# --- Type Definitions ---
+History = List[Tuple[Optional[str], Optional[str]]]
+Messages = List[Dict[str, any]]
+
+# --- History and Message Conversion ---
+
+def history_to_messages(history: History, system_prompt: str) -> Messages:
+    """Converts Gradio's history format to the list of messages format for an API call."""
+    messages = [{'role': 'system', 'content': system_prompt}]
+    for user_msg, assistant_msg in history:
+        # Handle potential multimodal user message (which comes as a list)
+        if isinstance(user_msg, list):
+            # Find the text part of the message for history
+            text_content = next((item.get("text", "") for item in user_msg if isinstance(item, dict) and item.get("type") == "text"), "")
+            messages.append({'role': 'user', 'content': text_content})
+        elif user_msg:
+             messages.append({'role': 'user', 'content': user_msg})
+        
+        if assistant_msg:
+            messages.append({'role': 'assistant', 'content': assistant_msg})
+    return messages
+
+def messages_to_history(messages: Messages) -> History:
+    """Converts a list of messages back to Gradio's history format."""
+    history = []
+    # Skip system message at index 0
+    for i in range(1, len(messages), 2):
+        user_msg = messages[i]['content']
+        assistant_msg = messages[i+1]['content'] if (i+1) < len(messages) else ""
+        history.append((user_msg, assistant_msg))
+    return history
+
+# --- Image Processing ---
+
+def process_image_for_model(image_data: np.ndarray) -> str:
+    """Converts a NumPy image array to a base64-encoded string."""
+    pil_img = Image.fromarray(image_data)
+    buffer = io.BytesIO()
+    pil_img.save(buffer, format="PNG")
+    img_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+    return f"data:image/png;base64,{img_str}"
+
+# --- Code Manipulation ---
+
+def remove_code_block(text: str) -> str:
+    """Extracts code from a markdown-style code block."""
+    pattern = r'```(?:[a-zA-Z]+)?\n(.*?)\n```'
+    match = re.search(pattern, text, re.DOTALL)
+    if match:
+        return match.group(1).strip()
+    return text.strip()
+
+def apply_search_replace(original_code: str, change_block: str) -> str:
+    """Applies a single search-and-replace block to the code."""
+    try:
+        parts = re.split(f"^{DIVIDER}$", change_block, flags=re.MULTILINE)
+        if len(parts) != 2: return original_code # Invalid block
+        
+        search_part, replace_part = parts
+        search_content = search_part.replace(SEARCH_START, "").strip()
+        replace_content = replace_part.replace(REPLACE_END, "").strip()
+
+        # To insert, search block is empty or just contains the line before insertion
+        if not search_content:
+             # Inserting at the beginning
+             return replace_content + "\n" + original_code
+
+        if search_content in original_code:
+            return original_code.replace(search_content, replace_content)
+        else:
+            # Handle insertion case where `search_content` is the line *before* insertion point
+            # and `replace_content` includes that line plus the new code.
+            # This is a common pattern LLMs use.
+            # We can simply return the original code, as more advanced logic is needed to reliably handle this.
+            print(f"Warning: Search block not found:\n---\n{search_content}\n---")
+            return original_code
+
+    except Exception as e:
+        print(f"Error applying changes: {e}")
+        return original_code
+
+def get_gradio_language(language: str) -> Optional[str]:
+    """Returns the language name if supported by Gradio, otherwise None."""
+    return language if language in GRADIO_SUPPORTED_LANGUAGES else None