ll5270
2022-09-02 18:57:45 +08:00
这个我有经验
import io
import os
import PyPDF2
import httpx
import pikepdf
from app.general.exception import ValidationException
def get_pdf_by_url(url):
try:
rsp = httpx.get(url)
except Exception:
raise ValidationException(msg="下载地址访问失败")
raw_pdf_io = io.BytesIO(rsp.content)
pdf = pikepdf.Pdf.open(raw_pdf_io)
pike_pdf_io = io.BytesIO()
pdf.save(pike_pdf_io)
return PyPDF2.PdfFileReader(pike_pdf_io)
def add_watermark(water_file, page_pdf):
"""
将水印 pdf 与 pdf 的一页进行合并
"""
pdfReader = PyPDF2.PdfFileReader(water_file)
page_pdf.mergePage(pdfReader.getPage(0))
return page_pdf
def execute_watermark_pdf(path, url):
pdf_reader = get_pdf_by_url(url)
pdf_writer = PyPDF2.PdfFileWriter() # 用于写 pdf
# page_pdf = add_watermark("/Users/xx/Downloads/aa.pdf", pdf_reader.getPage(0))
for page in range(pdf_reader.numPages):
if page == 0:
page_pdf = add_watermark("/Users/xx/Downloads/aa.pdf", pdf_reader.getPage(page))
elif page == 5:
page_pdf = add_watermark("/Users/xx/Downloads/aa6.pdf", pdf_reader.getPage(page))
else:
page_pdf = pdf_reader.getPage(page)
pdf_writer.addPage(page_pdf)
new_pdf = io.BytesIO()
pdf_writer.write(new_pdf)
new_pdf.seek(0)
with open(f"pdf/{path}.pdf", "wb") as f:
f.write(new_pdf.getvalue())
from PyPDF2 import PdfFileReader, PdfFileWriter, PdfFileMerger
from os import remove, listdir, mkdir
from os.path import join, isdir, split, splitext, basename
from pdf2image import convert_from_path
from reportlab.lib.pagesizes import A4, landscape, portrait
from reportlab.pdfgen import canvas
# 把 pdf 文件拆成 jpg 图片，每页一张
def pdf2jpgs(path, file):
# 路径和文件名
# 创建同名文件夹
mkdir("temp/" + path)
# 转换图片，图片像素 dpi 为 480
print("pdf/" + file)
images = convert_from_path("pdf/" + file, dpi = 100, fmt = 'JPEG',
thread_count = 4,)
# 保存图片
for index, image in enumerate(images):
image.save('temp/{}/{}.jpg'.format(path,index))
# 把 jpg 图片合并为 pdf 文件
def merge_jpg2pdf(jpgpath):
# 要合并的图片
jpg_files = [join(jpgpath, fn) for fn in listdir("temp/" + jpgpath)
if fn.endswith('.jpg')]
jpg_files.sort(key = lambda fn: int(splitext(basename(fn))[0]))
result_pdf = PdfFileMerger()
# 临时文件
temp_pdf = 'temp.pdf'
# 依次转 pdf ，再合并 pdf
for fn in jpg_files:
# 转 pdf ，portrait 纵向页面，landscape 横向页面
c = canvas.Canvas(temp_pdf, pagesize = portrait(A4))
c.drawImage("temp/" + fn, 0 , 0, *portrait(A4))
c.save()
# 合并
with open(temp_pdf, 'rb') as fp:
pdf_reader = PdfFileReader(fp)
result_pdf.append(pdf_reader)
# 保存结果
result_pdf.write("result/" + jpgpath + '.pdf')
result_pdf.close()
# remove("temp/" + jpgpath)
if __name__ == '__main__':
data = []
for i in data:
if not os.path.exists("result" + "/" + i["user_loan_id"] + ".pdf"):
execute_watermark_pdf(i["user_loan_id"], i["url"])
# 转图片
pdf2jpgs(i["user_loan_id"], i["user_loan_id"] + ".pdf")
# 图片合并成 pdf
merge_jpg2pdf(i["user_loan_id"])
# break