我最初是从这里得到脚本的--〉Cropping pages of a .pdf file。
我仔细阅读了作者的话，终于意识到他说的是：
生成的文档有一个200 x200磅的裁切框，从媒体框内的25，25磅处开始。裁切框在裁切框内的25磅处。
意义

page.cropBox.upperRight = (200, 200)

必须控制最终利润，因此我将报表调整为

page.cropBox.upperLeft = (290, 792)

将裁剪镜像到另一侧并确保裁剪保持完整的垂直值

将每页切成两半（例如，如果源文件是以小册子形式创建的），然后将其重新组合以进行进一步处理（例如，文本提取）
导入所需的库

from PyPDF2 import PdfFileWriter,PdfFileReader,PdfFileMerger

拆分左侧部分

with open("docu.pdf", "rb") as in_f:
    input1 = PdfFileReader(in_f)
    output = PdfFileWriter()

    numPages = input1.getNumPages()

    for i in range(numPages):
        page = input1.getPage(i)
        page.cropBox.lowerLeft = (60, 50)
        page.cropBox.upperRight = (305, 700)
        output.addPage(page)

    with open("left.pdf", "wb") as out_f:
        output.write(out_f)

拆分右侧部分：

with open("docu.pdf", "rb") as in_f:
    input1 = PdfFileReader(in_f)
    output = PdfFileWriter()

    numPages = input1.getNumPages()

    for i in range(numPages):
        page = input1.getPage(i)
        page.cropBox.lowerLeft = (300, 50)
        page.cropBox.upperRight = (540, 700)
        output.addPage(page)

    with open("right.pdf", "wb") as out_f:
        output.write(out_f)

左右组合（两栏两页）

input1 = PdfFileReader(open("left.pdf","rb"))
input2 = PdfFileReader(open("right.pdf","rb"))
output = PdfFileWriter()
numPages = input1.getNumPages()

for i in range(numPages):
    l = input1.getPage(i)
    output.addPage(l)
    r = input2.getPage(i)
    output.addPage(r)

with open("out.pdf", "wb") as out_f:
    output.write(out_f)

我面临着同样的挑战，我写了这个（需要PyPDF2）：
https://gist.github.com/kintaro1981/eb6cfc6f40a7fb39744f5ae630d58fd5

# -*- coding: utf-8 -*-
"""
This script splits each pdf page in half vertically and merge them in a single PDF in order.
It asks how many pages do you want to skip to avoid splitting a book cover or a specific number of initial pages.
Usage:
python cutpdfpages.py <filename.pdf> <newfilename.pdf>
"""

import sys
import copy
from PyPDF2 import PdfWriter, PdfReader

file = str(sys.argv[1])
newfile = str(sys.argv[2])

with open(file, "rb") as pdf1:
    pdf = PdfReader(pdf1)
    output = PdfWriter()
    numpages = pdf.getNumPages()
    page2skip =  int(input('Insert how many pages do you want to skip: '))

    for i in range(page2skip, numpages):
        page = pdf.pages[i]
        pagebis = copy.copy(page)
        page.mediabox.upper_left = (page.mediabox.right / 2, page.mediabox.top,)
        output.add_page(page)
        pagebis.mediabox.upper_right = (pagebis.mediabox.right / 2, pagebis.mediabox.top,)
        output.add_page(pagebis)

    with open(newfile, "wb") as newpdf:
        output.write(newpdf)