pdf-diff.py

import fitz  # PyMuPDF
import difflib



def extract_text_from_pdf(file_path):
    doc = fitz.open(file_path)
    pages_text = []
    for page_num in range(len(doc)):
        page = doc.load_page(page_num)
        text = page.get_text()
        pages_text.append(text)
    return pages_text

def highlight_differences(pdf1_path, pdf2_path, output_path):
    doc1 = fitz.open(pdf1_path)
    pdf1_pages = extract_text_from_pdf(pdf1_path)
    pdf2_pages = extract_text_from_pdf(pdf2_path)

    # Keep track of highlighted instances to avoid duplicates
    highlighted_instances = set()

    for page_num, (text1, text2) in enumerate(zip(pdf1_pages, pdf2_pages)):
        if text1 != text2:
            # Use difflib to find differences line by line
            diff = list(difflib.ndiff(text1.splitlines(), text2.splitlines()))
            page = doc1.load_page(page_num)

            # Track lines that have been removed, added, or moved
            removed_lines = set()
            added_lines = set()
            changed_lines = []

            # Collect removed and added lines
            for line in diff:
                if line.startswith("- "):
                    removed_lines.add(line[2:].strip())
                elif line.startswith("+ "):
                    added_lines.add(line[2:].strip())

            # Detect moved lines and changed lines
            moved_lines = removed_lines.intersection(added_lines)
            removed_lines -= moved_lines  # Remove moved lines from the removed set
            added_lines -= moved_lines  # Remove moved lines from the added set

            # Identify changed lines
            sm = difflib.SequenceMatcher(None, text1.splitlines(), text2.splitlines())
            for tag, i1, i2, j1, j2 in sm.get_opcodes():
                if tag == 'replace':
                    for i in range(i1, i2):
                        changed_lines.append(text1.splitlines()[i].strip())

            for line in diff:
                print(line)  # Print each line of the diff output

                # Initialize variables to prevent uninitialized color usage
                color = None
                note_content = ""

                # Check for moved lines first
                if line.startswith("- ") or line.startswith("+ "):
                    text_to_highlight = line[2:].strip()
                    if text_to_highlight in moved_lines:
                        color = (0, 0, 1)  # Blue for moved lines
                        note_content = "Moved line: Found in a different position"
                # Check for changed lines next
                elif line.startswith("- ") and text_to_highlight in changed_lines:
                    color = (1, 1, 0)  # Yellow for changed lines
                    note_content = f"Changed line: Original was '{text_to_highlight}'"
                # Check for added lines after that
                elif line.startswith("+ "):
                    text_to_highlight = line[2:].strip()
                    if text_to_highlight in added_lines:
                        color = (0, 1, 0)  # Green for added lines
                # Finally, check for removed lines
                elif line.startswith("- "):
                    text_to_highlight = line[2:].strip()
                    if text_to_highlight in removed_lines:
                        color = (1, 0, 0)  # Red for removed lines
                        note_content = "Removed line: Not present in the second PDF"
                
                # Highlight only if a color has been set
                if color is not None:
                    # Search for the text in the page and highlight it
                    print(f"Searching for: '{text_to_highlight}' on page {page_num + 1}")
                    text_instances = page.search_for(text_to_highlight)

                    for inst in text_instances:
                        if inst not in highlighted_instances:  # Check if already highlighted
                            annot = page.add_highlight_annot(inst)
                            annot.set_colors(stroke=color)  # Set the highlight color
                            annot.update()
                            if note_content:  # Add note if present
                                annot.set_info(title="Note", content=note_content)
                            highlighted_instances.add(inst)  # Mark this instance as highlighted
                    else:
                        print(f"No match found for: '{text_to_highlight}' on page {page_num + 1}")

            print(f"Differences highlighted on page {page_num + 1}.")

    # Save the annotated PDF
    doc1.save(output_path)
    print(f"Annotated PDF saved as '{output_path}'.")
    
def log_differences_to_file(pdf1_path, pdf2_path, output_txt_path):
    pdf1_pages = extract_text_from_pdf(pdf1_path)
    pdf2_pages = extract_text_from_pdf(pdf2_path)

    with open(output_txt_path, "w", encoding="utf-8") as log_file:
        for page_num, (text1, text2) in enumerate(zip(pdf1_pages, pdf2_pages)):
            if text1 != text2:
                # Use difflib to find differences line by line
                diff = list(difflib.ndiff(text1.splitlines(), text2.splitlines()))

                log_file.write(f"Page {page_num + 1}:\n")
                print(f"Page {page_num + 1}:\n")  # Print to terminal
                line_number = 1

                i = 0
                while i < len(diff):
                    line = diff[i]
                    text_to_highlight = line[2:].strip()

                    # Check for changed lines with context indicators
                    if line.startswith("- ") and (i + 1 < len(diff) and diff[i + 1].startswith("? ")):
                        original_line = text_to_highlight
                        new_line = diff[i + 2][2:].strip()
                        
                        # Check if there's a change indicator
                        if (diff[i + 1].startswith("? ")):
                            log_entry = f"  Line {line_number}: '{new_line}' replaced '{original_line}'\n"
                            log_file.write(log_entry)
                            print(log_entry, end="")  # Print to terminal
                            i += 4  # Skip the next two lines (the new line and the change indicator)
                            line_number += 1
                            continue

                    # Check for removed lines
                    elif line.startswith("- "):
                        log_entry = f"  Line {line_number}: Removed line -> '{text_to_highlight}'\n"
                        log_file.write(log_entry)
                        print(log_entry, end="")  # Print to terminal

                    # Check for added lines
                    elif line.startswith("+ "):
                        log_entry = f"  Line {line_number}: Added line -> '{text_to_highlight}'\n"
                        log_file.write(log_entry)
                        print(log_entry, end="")  # Print to terminal

                    line_number += 1
                    i += 1  # Move to the next line in the diff

                log_file.write("\n")  # Separate different pages for clarity
                print()  # Print a newline for better readability in terminal

        log_file.write("Comparison complete.\n")
        print("Comparison complete.\n")  # Print completion message to terminal

# Specify paths
pdf1_path = "1.pdf"
pdf2_path = "4.pdf"
output_path = "output_annotated.pdf"
output_txt_path = "differences_log.txt"

# highlight_differences(pdf1_path, pdf2_path, output_path)
log_differences_to_file(pdf1_path, pdf2_path, output_txt_path)