Python第二阶段学习总结

如果要从PDF文件中抽取中文，建议直接使用光学文字识别（OCR）

安装三方库：pip install easyocr

import PyPDF2
from PyPDF2.pdf import PageObject

reader = PyPDF2.PdfFileReader('resources/XGBoost.pdf')
page = reader.getPage(0)  # type: PageObject
print(page.extractText())

安装三方库：pip install fitz

import fitz
import os.path


def get_filename(file_path):
    """获取不带路径和后缀的文件名
    :param file_path: 文件路径
    :return: 文件名
    """
    
    # 拆分路径和文件名
    _, fullname = os.path.split(file_path)
    # 拆分文件名和后缀
    filename, _ = os.path.splitext(fullname)
    return filename


def pdf_image(pdf_file, img_path, zoom_x=4, zoom_y=4, rotation_angle=0):
    """将PDF文件转成PNG图片
    :param pdf_file: PDF文件路径
    :param img_path: 保存图片的路径
    :param zoom_x: 缩放比例（横向）
    :param zoom_y: 缩放比例（纵向）
    :param rotation_angle: 旋转角度
    """
    # 打开PDF文件
    pdf = fitz.open(pdf_file)
    # 逐页读取PDF
    for page_num in range(pdf.pageCount):
        page_obj = pdf[page_num]
        # 创建用于图像变换的矩阵
        trans = fitz.Matrix(zoom_x, zoom_y).preRotate(rotation_angle)
        # 将PDF页面处理成图像
        pm = page_obj.getPixmap(matrix=trans, alpha=False)
        temp = get_filename(pdf_file)
        pm.writePNG(f'{img_path}{temp}_{page_num + 1}.png')
    pdf.close()


def main():
    if not os.path.exists('resources/images/'):
        os.makedirs('resources/images/')
    pdf_image('resources/XGBoost.pdf', 'resources/images/')
    
if __name__ == '__main__':
    main()

import PyPDF2
from PyPDF2.pdf import PageObject

reader = PyPDF2.PdfFileReader('resources/XGBoost.pdf')
writer = PyPDF2.PdfFileWriter()
for page_num in range(reader.getNumPages()):
    # 读取原文件
    page_obj = reader.getPage(page_num)  # type:PageObject
    page_obj.rotateClockwise(90)
    # 写入新文件
    writer.addPage(page_obj)
    blank_page = writer.addBlankPage()  # type:PageObject
    blank_page.rotateClockwise(90)

with open('resources/XGBoost_modified.pdf', 'wb') as file:
    writer.write(file)

writer.encrypt('python')

import PyPDF2
from PyPDF2.pdf import PageObject

reader = PyPDF2.PdfFileReader('resources/watermark.pdf')
wm_page = reader.getPage(0)

reader = PyPDF2.PdfFileReader('resources/XGBoost.pdf')
writer = PyPDF2.PdfFileWriter()
for page_num in range(reader.getNumPages()):
    # 读取原文件
    page_obj = reader.getPage(page_num)  # type:PageObject
    page_obj.mergePage(wm_page)
    writer.addPage(page_obj)

with open('resources/XGBoost_watermarked.pdf', 'wb') as file:
    writer.write(file)

from reportlab.lib.pagesizes import A4
from reportlab.pdfbase import pdfmetrics
from reportlab.pdfbase.ttfonts import TTFont
from reportlab.pdfgen import canvas

pdf_canvas = canvas.Canvas('resources/demo.pdf', pagesize=A4)
width, height = A4

# 绘图
image = canvas.ImageReader('resources/guido.jpg')
pdf_canvas.drawImage(image, 20, height - 395, 250, 375)

# 显示当前页
pdf_canvas.showPage()

# 注册字体文件
pdfmetrics.registerFont(TTFont('Font1', 'resources/fonts/Action.ttf'))
pdfmetrics.registerFont(TTFont('Font2', 'resources/fonts/青呱石头体.ttf'))

# 写字
pdf_canvas.setFont('Font2', 40)
pdf_canvas.setFillColorRGB(0.9, 0.5, 0.3, 1)   # 前面三个参数是电脑三原色，最后一个参数0-1是透明度(阿尔法通道)
pdf_canvas.drawString(width // 2 - 120, height // 2, '你好，世界！')
pdf_canvas.setFont('Font1', 40)
pdf_canvas.setFillColorRGB(0, 1, 0, 0.5)
pdf_canvas.rotate(18)
pdf_canvas.drawString(250, 250, 'hello, world!')

# 保存
pdf_canvas.save()

import os

# 列出所有文件和文件夹
for file in os.listdir():
    # 判断是否是文件夹
    if os.path.isdir(file):     # isfile 是否是文件
        print(os.listdir(file))
    else:
        _, suffix = os.path.splitext(file)
        print(file, suffix)