安装：
pip install pdfminer解析pdf文件用到的类：
PDFParser：从一个文件中获取数据
PDFDocument：保存获取的数据，和PDFParser是相互关联的
PDFPageInterpreter处理页面内容
PDFDevice将其翻译成你需要的格式
PDFResourceManager用于存储共享资源，如字体或图像。
PDFMiner的类之间的关系图：
LTCurve:表示一个通用的Bezier曲线
- 方法1
# -*- coding:utf-8 -*- import time,os.path,requests,re time1=time.time() from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter from pdfminer.converter import PDFPageAggregator from pdfminer.layout import LAParams,LTTextBoxHorizontal from pdfminer.pdfpage import PDFTextExtractionNotAllowed,PDFPage from pdfminer.pdfparser import PDFParser from pdfminer.pdfdocument import PDFDocument class CPdf2TxtManager(): def changePdfToText(self, filePath): # 以二进制读模式打开 file = open(path, 'rb') #用文件对象来创建一个pdf文档分析器 praser = PDFParser(file) # 创建一个PDF文档对象存储文档结构,提供密码初始化，没有就不用传该参数 doc = PDFDocument(praser, password='') ##检查文件是否允许文本提取 if not doc.is_extractable: raise PDFTextExtractionNotAllowed # 创建PDf 资源管理器 来管理共享资源，#caching = False不缓存 rsrcmgr = PDFResourceManager(caching = False) # 创建一个PDF设备对象 laparams = LAParams() # 创建一个PDF页面聚合对象 device = PDFPageAggregator(rsrcmgr, laparams=laparams) # 创建一个PDF解析器对象 interpreter = PDFPageInterpreter(rsrcmgr, device) # 获得文档的目录（纲要）,文档没有纲要会报错 #PDF文档没有目录时会报：raise PDFNoOutlines pdfminer.pdfdocument.PDFNoOutlines # print(doc.get_outlines()) # 获取page列表 print(PDFPage.get_pages(doc)) # 循环遍历列表，每次处理一个page的内容 for page in PDFPage.create_pages(doc): interpreter.process_page(page) # 接受该页面的LTPage对象 layout = device.get_result() # 这里layout是一个LTPage对象 里面存放着 这个page解析出的各种对象 # 一般包括LTTextBox, LTFigure, LTImage, LTTextBoxHorizontal 等等 for x in layout: if hasattr(x, "get_text"): fileNames = os.path.splitext(filePath) with open(fileNames[0] + '.txt','a+') as f: results = x.get_text() print(results) f.write(results + '\n') # 如果x是水平文本对象的话 # if (isinstance(x, LTTextBoxHorizontal)): # text = re.sub(replace, '', x.get_text()) # if len(text) != 0: # print(text) if __name__ == '__main__': path = './pdf文本处理/12.pdf' pdf2TxtManager = CPdf2TxtManager() pdf2TxtManager.changePdfToText(path) time2 = time.time() print('ok,解析pdf结束!') print('总共耗时：' + str(time2 - time1) + 's')- 方法2
# -*- coding: utf-8 -*- from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter from pdfminer.converter import TextConverter from pdfminer.layout import LAParams from pdfminer.pdfpage import PDFPage import requests,os,re try:#python3 from io import StringIO from urllib.request import urlopen except:#python2 from urllib import urlopen from cStringIO import StringIO def convert_pdf_to_txt(path,save_name): if debug: # 加载内存的方式 retstr = StringIO() fp = StringIO(path) else: #读取文件的方式 retstr = open(path, 'rb') fp = open(path, 'rb') # 创建一个PDF资源管理器对象来存储共享资源,caching = False不缓存 rsrcmgr = PDFResourceManager(caching=False) # 创建一个PDF设备对象 laparams = LAParams() device = TextConverter(rsrcmgr, retstr, codec='utf-8', laparams=laparams) # 创建一个PDF解析器对象 interpreter = PDFPageInterpreter(rsrcmgr, device) for page in PDFPage.get_pages(fp, pagenos=set(), maxpages=0, password='',caching=True, check_extractable=True): interpreter.process_page(page) fp.close()#关闭输入流 device.close()#关闭输出流 str = retstr.getvalue() retstr.close() try: with open("%s"%save_name,"w") as f: for i in str: f.write(i) print("%s Writing Succeed!"%save_name) except: print("Writing Failed!") if __name__ == '__main__': try: debug=True if debug: #这种方式暂时还有问题 url="http://pythonscraping.com/pages/warandpeace/chapter1.pdf" pdf_file = urlopen(url).read() # 也可以换成本地pdf文件，用open rb模式打开 # pdf_file = requests.get(url).content # 加载内存的方式 convert_pdf_to_txt(pdf_file, "123.txt") else: #读取文件的方式 convert_pdf_to_txt('11.pdf',"123.txt") except Exception as e: import traceback ex_msg = '{exception}'.format(exception=traceback.format_exc()) print(ex_msg)批量提取
# -*- coding:utf-8 -*- import os,re from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter from pdfminer.pdfpage import PDFPage from pdfminer.converter import TextConverter from pdfminer.layout import LAParams #将一个pdf转换成txt def pdfTotxt(filepath,outpath): try: fp = open(filepath, 'rb') outfp=open(outpath,'w') #创建一个PDF资源管理器对象来存储共享资源,caching = False不缓存 rsrcmgr = PDFResourceManager(caching = False) # 创建一个PDF设备对象 laparams = LAParams() device = TextConverter(rsrcmgr, outfp, codec='utf-8', laparams=laparams,imagewriter=None) #创建一个PDF解析器对象 interpreter = PDFPageInterpreter(rsrcmgr, device) for page in PDFPage.get_pages(fp, pagenos = set(),maxpages=0, password='',caching=False, check_extractable=True): page.rotate = page.rotate % 360 interpreter.process_page(page) #关闭输入流 fp.close() #关闭输出流 device.close() outfp.flush() outfp.close() except Exception as e: print("Exception:%s",e) #一个文件夹下的所有pdf文档转换成txt def fileTotxt(fileDir): files=os.listdir(fileDir) tarDir=fileDir+'txt' if not os.path.exists(tarDir): os.mkdir(tarDir) replace=re.compile(r'\.pdf',re.I) for file in files: filePath=fileDir+'\\'+file outPath=tarDir+'\\'+re.sub(replace,'',file)+'.txt' pdfTotxt(filePath,outPath) print("Saved "+outPath) if __name__ == '__main__': pdfTotxt(u'123.pdf', 'test.txt') fileTotxt('这里是目录的路径')文章来源: python提取pdf文本内容