压缩图片到指定的大小 MB 默认2MB； 原文件过大可能达不到指定的效果； 可以拿中间产物再跑一次脚本试试；

reduce_pdf.py

# pip install "pypdf[image]"
# https://pypdf.readthedocs.io/en/stable/user/file-size.html?highlight=size
# 压缩图片到指定的大小  MB  默认;

# python reduce_pdf.py source.pdf 2           #2MB
import io
import os
import sys  # NOQA

from pypdf import PdfReader, PdfWriter

source_pdf = sys.argv[1]
source_size = os.path.getsize(source_pdf)

require_size = 2 * 1024 * 1024  # MB
try:
    require_size = int(sys.argv[2]) * 1024 * 1024
except:
    pass
print(f"file:{source_pdf},require size={require_size / 1024 / 1024:.2f}; file size={source_size / 1024 / 1024:.2f}")

if require_size > source_size:
    exit(1)

_dir, _filename = os.path.split(source_pdf)
dest_pdf = f"{_dir}/reduce-{_filename}"

quality = round((require_size / source_size) * 100)
# quality = 40
reader = PdfReader(source_pdf)

for i in range(10):
    # print(f"Run Count:{i} {quality=}")
    if quality < 1:
        print(f"压缩失败,{quality=}")
        break
    writer = PdfWriter(dest_pdf)
    for page in reader.pages:
        writer.add_page(page)

    for page in writer.pages:
        for img in page.images:
            img.replace(img.image, quality=quality)

    _io_1 = io.BytesIO()
    writer.write(_io_1)
    dest_size = _io_1.getbuffer().nbytes
    print(f"{i}: {quality=},current_size={dest_size / 1024 / 1024:.2f}")
    if dest_size <= require_size:
        with open(dest_pdf, "wb") as f:
            writer.write(f)
        print(f"out file: < {dest_pdf} > ;size:{(os.path.getsize(dest_pdf) / 1024 / 1024):.2f}MB,Run Count:{i},{quality=}")
        break
    quality -= 4
    del writer