pdf_processing.py

# pdf_processing.py
import fitz  # PyMuPDF
import pymupdf4llm
import os
import tempfile
import traceback
from typing import Tuple, Optional, List, Dict, Any

def convert_rect_to_dict(rect: fitz.Rect) -> Optional[Dict[str, float]]:
    """Converts a fitz.Rect object to a dictionary."""
    if not rect or not isinstance(rect, fitz.Rect):
        print(f"Warning: Invalid rect object received: {rect}")
        return None
    return {
        "x0": rect.x0,
        "y0": rect.y0,
        "x1": rect.x1,
        "y1": rect.y1,
        "width": rect.width,
        "height": rect.height
    }

def try_map_issues_to_page_rects(
    issues_to_map_for_context: List[Dict[str, Any]],
    pdf_rects: List[fitz.Rect],
    page_number_for_mapping: int # 1-based page number
) -> int:
    """Helper function for mapping LT issues to PDF rectangles."""
    mapped_count = 0
    num_issues_to_try = len(issues_to_map_for_context)
    num_available_rects = len(pdf_rects)
    limit = min(num_issues_to_try, num_available_rects)

    for i in range(limit):
        issue_to_update = issues_to_map_for_context[i]
        if issue_to_update['is_mapped_to_pdf']: # Check the correct flag name
            continue
        pdf_rect = pdf_rects[i]
        coord_dict = convert_rect_to_dict(pdf_rect)
        if coord_dict:
            issue_to_update['pdf_coordinates_list'] = [coord_dict] # Store as list of dicts
            issue_to_update['is_mapped_to_pdf'] = True
            issue_to_update['mapped_page_number'] = page_number_for_mapping
            mapped_count += 1
        else:
            print(f"      Warning: Could not convert rect for context '{issue_to_update['context_text'][:30]}...' on page {page_number_for_mapping}")
    return mapped_count

def extract_pdf_text(file_input: Any) -> str:
    """Extracts full text from a PDF file using PyMuPDF4LLM (as Markdown)."""
    temp_file_path_for_pymupdf4llm = None
    actual_path_to_process = None
    try:
        if isinstance(file_input, str):
            actual_path_to_process = file_input
        elif hasattr(file_input, 'read') and callable(file_input.read):
            temp_file_obj = tempfile.NamedTemporaryFile(suffix=".pdf", delete=False)
            temp_file_path_for_pymupdf4llm = temp_file_obj.name
            file_input.seek(0)
            temp_file_obj.write(file_input.read())
            temp_file_obj.close()
            actual_path_to_process = temp_file_path_for_pymupdf4llm
        else:
            raise ValueError("Input 'file_input' must be a file path (str) or a file-like object.")

        doc_for_page_count = fitz.open(actual_path_to_process)
        page_count = len(doc_for_page_count)
        doc_for_page_count.close()
        print(f"PDF has {page_count} pages. Extracting Markdown using pymupdf4llm.")
        
        markdown_text = pymupdf4llm.to_markdown(actual_path_to_process)
        
        print(f"Total extracted Markdown text length: {len(markdown_text)} characters.")
        return markdown_text
        
    except Exception as e:
        print(f"Error extracting text from PDF: {str(e)}")
        traceback.print_exc()
        return ""
    finally:
        if temp_file_path_for_pymupdf4llm and os.path.exists(temp_file_path_for_pymupdf4llm):
            os.remove(temp_file_path_for_pymupdf4llm)