自动化给pdf作业添加标注

背景

已知扣分和扣分原因
给每个学生的pdf报告上特定位置添加批注

实现功能

• 查找pdf上特定文字所在页面
• 在该页面添加批注

代码

有借鉴，后补引用

#-*- coding = utf-8 -*-
import pandas as pd
from PyPDF2 import PdfFileWriter, PdfFileReader
import io
from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import letter,A4 #纸大小
from reportlab.pdfbase import pdfmetrics #可以写中文
from reportlab.pdfbase.ttfonts import TTFont
import pandas as pd
import pdfplumber
import os
import re

#%% read files
path = r'D:\AAAAA Files\Season2021-2022\下学期'
classes_ = os.listdir(path)

class_ = classes_[0]

s_files = os.listdir(os.path.join(path, class_))
s_files = pd.DataFrame(s_files,columns=['origin_name'])
s_files['stdid'] = s_files['origin_name'].str.slice(0,7)
s_files['isfile'] = s_files['origin_name'].\
    apply(lambda s: os.path.isfile(os.path.join(path, class_, s)))
s_files['ispdf'] = s_files['origin_name'].apply(lambda s: s[-4:]=='.pdf')
s_files = s_files[(s_files['isfile']==False)|(s_files['ispdf']==True)]

#%%
name_table = \
    pd.read_excel(r'D:\AAAAA Files\Season2021-2022\下学期\学生名单-期末作业打分表.xlsx')
name_table = name_table[['学生期末作业成绩', 'Unnamed: 4', 'Unnamed: 8']]
name_table.dropna(inplace=True)
name_table = name_table.rename(columns={'学生期末作业成绩':'学号',
                                        'Unnamed: 4':'分数',
                                        'Unnamed: 8':'评价'})
name_table['isfinished'] = False


#%%
pdfmetrics.registerFont(TTFont("SimSun",
                               r"D:\AAAAA Files\Season2021-2022\注\Simsun.ttf"))

#%%

def markpdf(filepath, isfile, stdid):
    
    pdf_path = 0
    if isfile == False:
        filepathes = os.listdir(os.path.join(path,class_,filepath))
        for file_ in filepathes:
            if (file_[-4:]=='.pdf')&('代码' not in file_):
                pdf_path = os.path.join(path, class_, filepath, file_)
                break
            else:
                continue
        if pdf_path == 0:
            return -1
    elif (filepath[-4:]=='.pdf')&('代码' not in filepath):
        pdf_path = os.path.join(path, class_, filepath)
    else:
        print(filepath+'未完成')
        return -1

#% 寻找文字
    pdf = pdfplumber.open(pdf_path)
    pagei = -1
    pagenum = len(pdf.pages)
    for i in range(pagenum):
        pagetext = pdf.pages[i].extract_text()
        if re.search('.*?(文字一|文字二|文字三).*?',pagetext) is not None:
            pagei = i
            break
        else:
            continue
    if pagei == -1:
        print(filepath+'未完成')
        return -1
    pdf.close()


    packet = io.BytesIO()
    # 使用Reportlab创建一个新的PDF
    can = canvas.Canvas(packet, pagesize=A4)
    can.setFont("SimSun", 10)
    
    can.setFillColorRGB(255, 0, 0)
    temp = name_table[name_table['学号']==stdid]
    if temp.shape[0]==0:
        print(filepath+'未完成')
        return -1
    
    can.drawString(80, 50, "任务二：%s \n扣分情况：-%s"%(temp['评价'].values[0],
                                                25-temp['分数'].values[0]))
    can.save()
    
    #buffer从偏移0开始
    packet.seek(0)
    new_pdf = PdfFileReader(packet)
    #读取已有的PDF
    existing_pdf = PdfFileReader(open(pdf_path, "rb"))
    output = PdfFileWriter()
    #
    page = existing_pdf.getPage(pagei)
    page.mergePage(new_pdf.getPage(0))

    for i in range(pagenum):
        if i ==pagei:
            output.addPage(page)
        else:
            output.addPage(existing_pdf.getPage(i))
    
    # 最后，向目标的pdf写出
    outputStream = open('%s-批改完成.pdf'%pdf_path[:-4], "wb")
    output.write(outputStream)
    outputStream.close()
    packet.close()
    # print(filepath+'完成')
    # os.remove(pdf_path)
    # readed = readed + [stdid]
    return stdid

#%%
readed = []
for fi in s_files.iterrows():
    i, f = fi
    stdid = markpdf(f['origin_name'],f['isfile'],int(f['stdid']))
    if stdid>0:
        readed = readed+[stdid]

readed = pd.DataFrame(readed, columns=['学号'])
readed['完成'] = True
name_table = pd.merge(name_table, readed, on='学号',how='left')

结果