Python Code Help - pip install PyMuPDF python-docx pillow

[Splishsplash92]

I have this code below. The purpose of the code is to: extract the paragraphs that include an asterisk and its associated photos, from a PDF document, into a Word Document.

The code works and it exports the paragraphs that include an asterisk into a Word Doc, but it is not grabbing only the photos associated with the paragraph, it sometimes exports photos that are for another paragraph. How can I modify the code to make sure it ONLY exports the images directly below the paragraph?

import fitz  # PyMuPDF
from docx import Document
from docx.shared import Inches
import io
from PIL import Image

# Load the PDF document
pdf_document = fitz.open("Sample Home.pdf")

# Create a Word document
word_document = Document()

# Iterate through each page of the PDF
for page_num in range(pdf_document.page_count):
    page = pdf_document.load_page(page_num)
    blocks = page.get_text("blocks")

    for block in blocks:
        block_text = block[4]

        # Check if the paragraph includes an asterisk
        if '*' in block_text:
            # Add the paragraph to the Word document
            word_document.add_paragraph(block_text)

            # Extract images associated with this paragraph
            image_list = page.get_images(full=True)
            for image_index, img in enumerate(image_list):
                xref = img[0]
                base_image = pdf_document.extract_image(xref)
                image_bytes = base_image["image"]

                # Load image using PIL
                image = Image.open(io.BytesIO(image_bytes))
                image_filename = f"image_{page_num}_{image_index}.png"
                image.save(image_filename)

                # Add image to the Word document
                word_document.add_picture(image_filename, width=Inches(5))

# Save the Word document
word_document.save("Extracted_Paragraphs_and_Images.docx")

Gribouillis write May-31-2024, 09:28 PM:
Please post all code, output and errors (it it's entirety) between their respective tags. Refer to BBCode help topic on how to post. Use the "Preview Post" button to make sure the code is presented as you expect before hitting the "Post Reply/Thread" button.

[Pedroski55]

Got a sample PDF to experiment on?

[Splishsplash92]

I do, I tried uploading the PDF to my post but the file size is too large.

[Pedroski55]

Just a little bit of the PDF will do, say 4 or 5 pages, as long as they contain the type of data you are looking for.

I tried your code on the PDF manual for my new induction cooker. It worked ok!

pdf_document = fitz.open("/home/pedro/pdfs/pdfs/user_manual_ce208.pdf")
page = pdf_document.load_page(0)
# this returns tuples of the block coordinates and the text
blocks = page.get_text("blocks")
for block in blocks:
    if 'Instruction manual' in str(block):
        print(type(block), block)

The above prints:

Output:
<class 'tuple'> (320.31451416015625, 132.03631591796875, 445.85235595703125, 147.72772216796875, 'Instruction manual\n', 3, 0)
<class 'tuple'> (118.06179809570312, 468.69854736328125, 195.3158721923828, 478.35479736328125, 'Instruction manual\n', 5, 0)

Not sure what the last two integers in each tuple represent!

Just looked them up, the tuple is:

(x0, y0, x1, y1, "lines in block", block_no, block_type)