Python 办公 | PDF 文件处理指南

pip install PyPDF2 FPDF reportlab

import PyPDF2
# Open PDF filewith open('example.pdf', 'rb') as file:  # Create a PDF reader object  reader = PyPDF2.PdfReader(file)    # Get the number of pages in a PDF file.  num_pages = len(reader.pages)    # Extract text content page by page and print.  for page_num in range(num_pages):    page = reader.pages[page_num]    text = page.extract_text()    print(text)

from PyPDF2 import PdfMerger, PdfReader, PdfWriter
# Create a PDF mergermerger = PdfMerger()
# Merge multiple PDF filesmerger.append('example.pdf')merger.append('file2.pdf')
# Save the merged file.merger.write('merged.pdf')merger.close()
# Split a PDF filewith open('merged.pdf', 'rb') as file:    reader = PdfReader(file)    num_pages = len(reader.pages)      # Split every 10 pages into one file.    for start in range(0, num_pages, 10):        end = min(start + 9, num_pages - 1)        writer = PdfWriter()            # Add specified range of pages to a new file.        for page_num in range(start, end + 1):            writer.add_page(reader.pages[page_num])            # Save the split files.        with open(f'part_{start+1}-{end+1}.pdf', 'wb') as output_file:            writer.write(output_file)
print("Ta-da! The magic of merging and splitting is complete.！")

from PyPDF2 import PdfReader, PdfWriter
from reportlab.pdfgen import canvas
# Automatically fill in form fieldsdef fill_form(input_file, output_file, data):    c = canvas.Canvas(output_file)    c.setFont("Helvetica", 12)        # Read input file, process page by page.    reader = PdfReader(input_file)    for page_num, page in enumerate(reader.pages, start=1):        # Obtain the size of the page and create a canvas of corresponding size.        page_width = float(page.mediabox.width)        page_height = float(page.mediabox.width)        c.setPageSize((page_width, page_height))                # Draw page content        c.showPage()                # Check if the page has form fields.        if '/Annots' in page:            # Traverse all form fields            for annot in page['/Annots']:                # Check the field type as text area.                if '/T' in annot and '/V' in annot and annot['/Type'] == '/Annot':                    field_name = annot['/T'][1:-1]  # Get field name                    # Replace the field value with the incoming data.                    if field_name in data:                        field_value = data[field_name]                        c.drawString(annot['/Rect'][0], annot['/Rect'][1], field_value)        # Save the PDF with filled data.    c.save()
# Read the filled data.def read_form_data(input_file):    data = {}    reader = PdfReader(input_file)        # Traverse all pages.    for page in reader.pages:        # Check if there are form fields.        if '/Annots' in page:            # Traverse all form fields            for annot in page['/Annots']:                # Check the field type as text area.                if '/T' in annot and '/V' in annot and annot['/Type'] == '/Annot':                    field_name = annot['/T'][1:-1]  # Get field name                    field_value = annot['/V'][1:-1] if isinstance(annot['/V'], str) else ''                    data[field_name] = field_value        return data
# Create a new PDF formdef create_form(output_file, data):    c = canvas.Canvas(output_file)    c.setFont("Helvetica", 12)        # Add data line by line to the form.    y = 800    for field, value in data.items():        c.drawString(50, y, f"{field}: {value}")        y -= 20        # Save form    c.save()
# Fill in the form fields and save.fill_form('form_template.pdf', 'filled_form.pdf', {'name': 'joe', 'age': '18'})
# Read the filled data and print.form_data = read_form_data('filled_form.pdf')print(form_data)
# Create a new PDF formcreate_form('my_form.pdf', {'name': 'joe', 'age': '18'})
print("Magic completed! Now you can easily handle PDF forms!")

from pdf2image import convert_from_path
def pdf_to_image(input_file, output_file):    images = convert_from_path(input_file)    for i, image in enumerate(images):        image.save(f'{output_file}_{i}.jpg', 'JPEG')
pdf_to_image('input.pdf', 'output_image')

brew install poppler

sudo apt-get install poppler-utils

from PyPDF2 import PdfReader
def pdf_to_html(input_file, output_file):    with open(input_file, 'rb') as file:        reader = PdfReader(file)        text = ""                # Extract text content page by page        for page in reader.pages:            text += page.extract_text()                # Save as HTML file        with open(output_file, 'w') as html_file:            html_file.write(f"{text}")
pdf_to_html('input.pdf', 'output.html')

pip install pdfminer.six

from pdfminer.high_level import extract_text_to_fp
def pdf_to_text(input_file, output_file):    with open(output_file, 'w') as text_file:        with open(input_file, 'rb') as file:            extract_text_to_fp(file, text_file)
pdf_to_text('input.pdf', 'output.txt')

pip install python-docx PyPDF2

from docx import Documentfrom PyPDF2 import PdfReader
def pdf_to_word(input_file, output_file):    with open(input_file, 'rb') as file:        reader = PdfReader(file)        text = ""                # Extract text content page by page        for page in reader.pages:            text += page.extract_text()                # Create Word document        doc = Document()        doc.add_paragraph(text)                # Save as Word document        doc.save(output_file)
pdf_to_word('input.pdf', 'output.docx')

from PyPDF2 import PdfReader, PdfWriterfrom reportlab.pdfgen import canvasimport io
# Add watermark to PDFdef add_watermark(input_file, output_file, watermark_text):    reader = PdfReader(input_file)    writer = PdfWriter()
    watermark_buffer = io.BytesIO()
    # Create a PDF with watermark.    c = canvas.Canvas(watermark_buffer)    c.setFont("Helvetica", 48) # Pay attention to the choice of font.    c.rotate(45)    c.translate(-500, -500)    c.setFillAlpha(0.3)    c.drawString(400, 400, watermark_text)    c.save()
    watermark_buffer.seek(0)    watermark_pdf = PdfReader(watermark_buffer)
    # Traverse each page.    for i, page in enumerate(reader.pages, start=1):        watermark_page = watermark_pdf.pages[0]
        # Add watermark to the page        page.merge_page(watermark_page)        writer.add_page(page)
    # Save the file with watermark.    with open(output_file, 'wb') as file:        writer.write(file)
# Add digital signature to PDFdef add_signature(input_file, output_file, signature_image):    reader = PdfReader(input_file)    writer = PdfWriter()
    # Traverse each page.    for i, page in enumerate(reader.pages, start=1):        # Add a signature image to the bottom right corner of the page.        page.merge_page(signature_image)        writer.add_page(page)
    # Save the document with a signature.    with open(output_file, 'wb') as file:        writer.write(file)
# Use the functions add_watermark() and add_signature() to add watermark and signature.watermark_text = "Confidential document, do not disclose."signature_image = PdfReader("signature.pdf").pages[0]
add_watermark('part_21-30.pdf', 'document_with_watermark.pdf', watermark_text)add_signature('document_with_watermark.pdf', 'document_with_watermark_and_signature.pdf', signature_image)
print("Watermark and signature magic completed! You can now make your PDF files more secure and professional!")

pip install pytesseract

import matplotlib.pyplot as pltfrom reportlab.lib.pagesizes import A4from reportlab.platypus import SimpleDocTemplate, Table, Imagefrom reportlab.lib.styles import getSampleStyleSheetfrom reportlab.platypus import Paragraph, Spacer
# Create report contentdef create_report(output_file, data):    # Create PDF document object    doc = SimpleDocTemplate(output_file, pagesize=A4)
    # Load stylesheet    styles = getSampleStyleSheet()
    # Create report content elements    elements = []
    # Add title    title = Paragraph("Sales report", styles["Title"])    elements.append(title)    elements.append(Spacer(1, 20))
    # Add table    table_data = data    table = Table(table_data)    elements.append(table)    elements.append(Spacer(1, 20))
    # Generate charts and save as PNG images    plt.plot(data[1][1:], marker='o')    plt.xlabel("date")    plt.ylabel("Sales revenue")    plt.title("Sales Trend Chart")    plt.savefig("sales_plot.png")    plt.close()
    # Add charts to the report content.    image = Image("sales_plot.png", width=400, height=300)    elements.append(image)
    # Generate report    doc.build(elements)
# Create report datareport_data = [    ["date", "Sales revenue"],    ["1/1", 100],    ["1/2", 200],    ["1/3", 150],    ["1/4", 300],]
# Generate reportcreate_report('sales_report.pdf', report_data)
print("Table generation completed! Now you can view the generated report file.")

import pdf2imageimport pytesseract
# Convert PDF to imagedef pdf_to_image(input_file):    images = pdf2image.convert_from_path(input_file)    return images
# Use OCR to convert images into text.def image_to_text(image):    text = pytesseract.image_to_string(image)    return text
# Save the text to a filedef save_text_to_file(text, output_file):    with open(output_file, 'w', encoding='utf-8') as file:        file.write(text)
# Extract text from PDF.def extract_text_from_pdf(input_file, output_file):    # Convert PDF to image    images = pdf_to_image(input_file)        extracted_text = ""        # Extract text from each image.    for image in images:        text = image_to_text(image)        extracted_text += text + "\n"        # Save the extracted text to a file.    save_text_to_file(extracted_text, output_file)
# Extract text from scanned PDF.extract_text_from_pdf('scanned_document.pdf', 'extracted_text.txt')
print("OCR (Optical Character Recognition) magic completed! Now you can convert scanned PDF documents into editable text!")