python pdf 转换成txt，csv，doc 及doc转换为pdf初级

pdf，搜索相关路径下pdf文档，把pdf文档移动到指定的路径下，再对该路径下的pdf文档进行转换

缺点：无法转换带有图片的pdf文档，转换后的文档格式问题

安装相应的库

1）pip install pdfminer3k 是pdfminer的Python 3端口

2）安装docx库

pip install python_docx

使用了os.walk对特定路径下的pdf文档进行查找，并对该文档进行移动

import  os
import shutil
import importlib
import sys
import re
def load_file():walk = os.walk(r'C:\Users\ALFIEL\Desktop\20190527')i=1for root, dirs,files in walk:print((root,dirs,files))for name in files:if len(re.findall('(\w)\.pdf',name))>=1:#添加条件对pdf文件进行筛选shutil.move(os.path.join(root,name), "C:\\Users\\SALFIEL\\Desktop\\pdfdocement\\"+str(i)+".pdf")i+=1load_file()

转换完整的代码如下：

#pdf转换器
from pdfminer.pdfparser import PDFParser, PDFDocument
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter,process_pdf
from pdfminer.layout import LAParams
from pdfminer.converter import PDFPageAggregator
from pdfminer.pdfinterp import PDFTextExtractionNotAllowed
from docx import Document
from pdfminer.layout import *
document = Document()
import warnings
warnings.filterwarnings("ignore")
from pdfminer.converter import TextConverter
from pdfminer.layout import LAParams
from io import StringIO
from urllib.request import urlopen
import pandas as pd
import  os
def readPDF(pdfFile):#读取pdf文件#创建PDf 资源管理器 来管理共享资源rsrcmgr = PDFResourceManager()retstr = StringIO()#数据保存到内存中laparams = LAParams()device = TextConverter(rsrcmgr, retstr, laparams=laparams)process_pdf(rsrcmgr, device, pdfFile)device.close()content = retstr.getvalue()retstr.close()return content
def save_to_file(file_name, contents):#可以存储为相应的格式的文档，单纯以文字的pdf转换fh = open(file_name, 'w',encoding='utf-8')#若没有encoding='utf-8'则会爆出gbk编码错误。写入txt，csv需要转换格式utf-8fh.write(contents)fh.close()# save_to_file('mobiles.txt', 'your contents str')def main():#当某个目录下面寻找pdf文档文件并把它移动到某个路径下面#进行txt或则csv的转换walk=os.walk(r"C:\Users\ALFIEL\Desktop\pdfdocement")for root, dirs,files in walk:for name in files:if len(re.findall('(\w)\.pdf',name))>=1:#添加条件对pdf文件进行筛选pdfFile = open("{}".format(os.path.join(root,name)),'rb')outputString = readPDF(pdfFile)save_to_file('c.csv',outputString)def save_to_doxc(file_name,path):#以二进制读取打开文件fn = open(file_name,'rb')#用文件对象创建pdf文档分析器parser = PDFParser(fn)#创建一个pdf文档doc = PDFDocument()#连接分析器与文档对象parser.set_document(doc)doc.set_parser(parser)# 提供初始化密码# 如果没有密码 就创建一个空的字符串doc.initialize()# 检测文档是否提供txt转换，不提供就忽略if not doc.is_extractable:raise PDFTextExtractionNotAllowedelse:resource = PDFResourceManager()laparams = LAParams()device = PDFPageAggregator(resource,laparams=laparams)# 创建一个PDF解释器对象interpreter = PDFPageInterpreter(resource,device)num_page, num_image, num_curve, num_figure, num_TextBoxHorizontal = 0, 0, 0, 0, 0for i in doc.get_pages():#循环遍历列表，每次处理一个page的内容num_page += 1 interpreter.process_page(i)layout = device.get_result()print(layout)for x in layout:if isinstance(x,LTImage):num_image += 1if isinstance(x,LTCurve): num_curve += 1if isinstance(x,LTFigure):  # figure对象num_figure += 1if isinstance(x, LTTextBoxHorizontal):#判断变量时水平文本框对象num_TextBoxHorizontal += 1# 水平文本框对象增一# 保存文本内容if hasattr(x,"get_text"):#判断x对象是否有get_text方法with open(r'{}'.format(path), 'a',encoding='utf-8') as f:    #生成doc文件的文件名及路径results = x.get_text()print(results)f.write(results)f.write('\n')# 获取文本内容print('对象数量：\n','页面数：%s\n'%num_page,'图片数：%s\n'%num_image,'曲线数：%s\n'%num_curve,'水平文本框：%s\n'%num_TextBoxHorizontal)print ('处理完成')def main2():walk=os.walk(r"C:\Users\ALFIEL\Desktop\pdfdocement")for root, dirs,files in walk:for name in files:if len(re.findall('(\w)\.pdf',name))>=1:#添加条件对pdf文件进行筛选pdfFile = os.open(r"{}".format(os.path.join(root,name)),os.O_RDWR)save_to_doxc(pdfFile,path=os.path.join(root,name).replace('.pdf','.doc'))if __name__ == '__main__':main2()

注意：当写入txt，csv文件时，需要吧编码模式encoding为utf-8

doc转pdf或其他

还有通过调用win32接口进行doc的转换

from win32com.client import Dispatch, constantsdef doc2pdf(input1, output):w = Dispatch('Word.Application')try:doc = w.Documents.Open(input1, ReadOnly=1)doc.SaveAs(output, 17)return Trueexcept Exception as e:print(e)return Falsefinally:doc.Close()
def main():input1 = r'C:\Users\ALFIEL\Desktop\pdfdocement\1.doc'output = r'C:\Users\ALFIEL\Desktop\pdfdocement\6.pdf'rc = doc2html(input1, output)if rc:print('转换成功')else:print('转换失败')if __name__ == '__main__':main()

其中SaveAs(output,wdformat)

wdfomat参数如下：

wdFormatDocument                    = 0
wdFormatDocument97                  = 0
wdFormatDocumentDefault             = 16
wdFormatDOSText                     = 4
wdFormatDOSTextLineBreaks           = 5
wdFormatEncodedText                 = 7
wdFormatFilteredHTML                = 10
wdFormatFlatXML                     = 19
wdFormatFlatXMLMacroEnabled         = 20
wdFormatFlatXMLTemplate             = 21
wdFormatFlatXMLTemplateMacroEnabled = 22
wdFormatHTML                        = 8
wdFormatPDF                         = 17
wdFormatRTF                         = 6
wdFormatTemplate                    = 1
wdFormatTemplate97                  = 1
wdFormatText                        = 2
wdFormatTextLineBreaks              = 3
wdFormatUnicodeText                 = 7
wdFormatWebArchive                  = 9
wdFormatXML                         = 11
wdFormatXMLDocument                 = 12
wdFormatXMLDocumentMacroEnabled     = 13
wdFormatXMLTemplate                 = 14
wdFormatXMLTemplateMacroEnabled     = 15
wdFormatXPS                         = 18照着字面意思应该能对应到相应的文件格式，如果你是office 2003可能支持不了这么多格式。word文件转html有两种格式可选wdFormatHTML、wdFormatFilteredHTML（对应数字8、10），区别是如果是wdFormatHTML格式的话，word文件里面的公式等ole对象将会存储成wmf格式，而选用wdFormatFilteredHTML的话公式图片将存储为gif格式，而且目测可以看出用wdFormatFilteredHTML生成的HTML明显比wdFormatHTML要干净许多。

参考自https://blog.csdn.net/binger819623/article/details/6770932