詳解用Python把PDF轉(zhuǎn)為Word方法總結(jié)

# pip install pdf2docx #安裝依賴庫
from pdf2docx import Converter

pdf_file = r'C:\Users\Administrator\Desktop\新建文件夾\mednine.pdf'
docx_file = r'C:\Users\Administrator\Desktop\Python教程\02.docx'

# convert pdf to docx
cv = Converter(pdf_file)
cv.convert(docx_file, start=0, end=None)
cv.close()

from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.converter import TextConverter
from pdfminer.layout import LAParams
from pdfminer.pdfpage import PDFPage
from io import StringIO
import sys
import string
from docx import Document


def convert_pdf_2_text(path):
    rsrcmgr = PDFResourceManager()
    retstr = StringIO()
    
    device = TextConverter(rsrcmgr, retstr, codec='utf-8', laparams=LAParams())
    interpreter = PDFPageInterpreter(rsrcmgr, device)
    
    with open(path, 'rb') as fp:
        for page in PDFPage.get_pages(fp, set()):
            interpreter.process_page(page)
            #print(retstr.getvalue())  
        text = retstr.getvalue()
    device.close()
    retstr.close()
    return text


def pdf2txt():
    text=convert_pdf_2_text(path)
    with open('real.txt','a',encoding='utf-8') as f:
        for line in text.split('\n'):
            f.write(line+'\n')

        

def remove_control_characters(content):
    mpa = dict.fromkeys(range(32))
    return content.translate(mpa)
    
def save_text_to_word(content, file_path):     

    doc = Document()
    for line in content.split(''):
        print(line)
            
        paragraph = doc.add_paragraph()
        paragraph.add_run(remove_control_characters(line))
    doc.save(file_path)


if __name__ == '__main__':
    path = r'C:\Users\mayn\Desktop\程序臨時(shí)\培訓(xùn)教材.pdf'  # 你自己的pdf文件路徑及文件名 不適合掃描版 只適合標(biāo)準(zhǔn)PDF文件
    text = convert_pdf_2_text(path)
    save_text_to_word(text, 'output.doc')  #PDF轉(zhuǎn)為word方法
    #pdf2txt()  #PDF轉(zhuǎn)為txt方法

import pdfplumber
import pandas as pd
import time
from time import  ctime
import psutil as ps 
#import threading
import gc
pdf = pdfplumber.open(r"C:\Users\Administrator\Desktop\新建文件夾\mednine.pdf")
N=len(pdf.pages)
print('總共有',N,'頁')

def pdf2exl(i): # 讀取了第i頁，第i頁是有表格的，
    print('********************************************************************************************************************************************************')
    print('正在輸出第',str(i+1),'頁表格')
    print('********************************************************************************************************************************************************')   
    p0 = pdf.pages[i]
    try:
        table = p0.extract_table()
        print(table)
    
        df = pd.DataFrame(table[1:], columns=table[0])
    #print(df)
        df.to_excel(r"C:\Users\Administrator\Desktop\新建文件夾\Model"+str(i+1)+".xlsx")
    
        #df.info(memory_usage='deep')
        
        
    except Exception as e:
        print('第'+str(i+1)+'頁無表格，或者檢查是否存在表格')       
       
        pass
    #print('目前內(nèi)存占用率是百分之',str(ps.virtual_memory().percent),'    第',str(i+1),'頁輸出完畢')
    print('**********************************************************************************************************************************************************')
    print('\n\n\n') 
    time.sleep(5)


def dojob1():  #此函數(shù)  直接循環(huán)提取PDF里面各個(gè)頁面的表格 
    print('*********************')
    for i in range(0,N):
        pdf2exl(i)    

import camelot
import wand

# 從PDF文件中提取表格

def output(i):  
    #print(tables)
    #for i in range(5):
    tables = camelot.read_pdf(r'C:\Users\Administrator\Desktop\新建文件夾\mednine.pdf', pages=str(i), flavor='stream')
    print(tables[i])
    
# 表格數(shù)據(jù)
    print(tables[i].data)
    
    tables[i].to_csv(r'C:\Users\Administrator\Desktop\新建文件夾\002'+str(i)+r'.csv')


def plotpdf():
        # 這個(gè)是畫pdf 結(jié)構(gòu)的函數(shù) 現(xiàn)在不能用 不要打開
    
    #print(tables[0])
    tables = camelot.read_pdf(r'C:\Users\mayn\Desktop\vcode工作區(qū)\11\路基.pdf', pages='200', flavor='stream')
    camelot.plot(tables[0], kind='text')
    print(tables[0])
    plt.show()
    # 繪制PDF文檔的坐標(biāo)，定位表格所在的位置  
    #plt = camelot.plot(tables[0],kind='text')
    #plt.show()
    #table_df = tables[0].df

#plotpdf() 
#i=3
#output(i)
for i in range(0,2):
    try:    
        output(i)
    except Exception as e:
        print('第'+str(i)+'頁沒找到表格啊啊啊')
        pass 
    continue