python 如何将ppt和word转化为txt文档

　　最近做了关于计算文档中关键词的程序，使用Tf-idf方法去提取，其中需要使用python读取MS word文档和 MS powerpoint中的内容，现将部分讲解和代码贴出来，请指正。

　　首先，介绍一下win32com，这是个和window链接的模块，实话说，功能是很强大的，在网上看到很多功能，可以用来打开word，ppt，Excel，Access，模拟浏览器等行为，下载地址：http://starship.python.net/~skippy/win32/Downloads.html，现在，介绍其功能，是从其他网上的资料粘贴过来的。

1.word功能：http://my.oschina.net/duxuefeng/blog/64137（这个写的比较清楚）　

import win32com
from win32com.client import Dispatch, constants

w = win32com.client.Dispatch('Word.Application')
# 或者使用下面的方法，使用启动独立的进程：
# w = win32com.client.DispatchEx('Word.Application')

# 后台运行，不显示，不警告
w.Visible = 0
w.DisplayAlerts = 0

# 打开新的文件
doc = w.Documents.Open( FileName = filenamein )
# worddoc = w.Documents.Add() # 创建新的文档

# 插入文字
myRange = doc.Range(0,0)
myRange.InsertBefore('Hello from Python!')

# 使用样式
wordSel = myRange.Select()
wordSel.Style = constants.wdStyleHeading1

# 正文文字替换
w.Selection.Find.ClearFormatting()
w.Selection.Find.Replacement.ClearFormatting()
w.Selection.Find.Execute(OldStr, False, False, False, False, False, True, 1, True, NewStr, 2)

# 页眉文字替换
w.ActiveDocument.Sections[0].Headers[0].Range.Find.ClearFormatting()
w.ActiveDocument.Sections[0].Headers[0].Range.Find.Replacement.ClearFormatting()
w.ActiveDocument.Sections[0].Headers[0].Range.Find.Execute(OldStr, False, False, False, False, False, True, 1, False, NewStr, 2)

# 表格操作
doc.Tables[0].Rows[0].Cells[0].Range.Text ='123123'
worddoc.Tables[0].Rows.Add() # 增加一行

# 转换为html
wc = win32com.client.constants
w.ActiveDocument.WebOptions.RelyOnCSS = 1
w.ActiveDocument.WebOptions.OptimizeForBrowser = 1
w.ActiveDocument.WebOptions.BrowserLevel = 0 # constants.wdBrowserLevelV4
w.ActiveDocument.WebOptions.OrganizeInFolder = 0
w.ActiveDocument.WebOptions.UseLongFileNames = 1
w.ActiveDocument.WebOptions.RelyOnVML = 0
w.ActiveDocument.WebOptions.AllowPNG = 1
w.ActiveDocument.SaveAs( FileName = filenameout, FileFormat = wc.wdFormatHTML )

# 打印
doc.PrintOut()

# 关闭
# doc.Close()
w.Documents.Close(wc.wdDoNotSaveChanges)
w.Quit()

2. Excel功能：同上的网址

3. Access功能：

import win32com.client
oAccess = win32com.client.Dispatch('Access.Application')
DbFile = r'C:UsershansDocumentsNewDb.mdb'
dbLangGeneral = ';LANGID=0x0409;CP=1252;COUNTRY=0'
# dbVersion40 64
dbVersion = 64
oAccess.DBEngine.CreateDatabase(DbFile, dbLangGeneral, dbVersion)
oAccess.Quit()
del oAccess

4. 模拟浏览器行为：http://www.cnblogs.com/chenzehe/archive/2010/09/01/1814397.html

Code highlighting produced by Actipro CodeHighlighter (freeware)http://www.CodeHighlighter.com/--># -*- coding:UTF-8 -*- 
#!/user/bin/env python
'''
Created on 2010-9-1
@author: chenzehe
'''
import win32com.client
from time import sleep

loginurl='http://passport.cnblogs.com/login.aspx'
loginouturl='http://passport.cnblogs.com/logout.aspx'
username='XXX'
password='XXX'

ie = win32com.client.Dispatch("InternetExplorer.Application")    
ie.Visible = 0


ie.Navigate(loginurl)    
state = ie.ReadyState    
print "打开登陆页面"    
while 1:    
    state = ie.ReadyState    
    if state ==4:    
        break    
    sleep(1)
print "页面载入完毕，输入用户名密码"    
state = None

ie.Document.getElementById("tbUserName").value=username    
ie.Document.getElementById("tbPassword").value=password    
ie.Document.getElementById("btnLogin").click()   


while 1:    
    state = ie.ReadyState    
    print state    
    if state ==4 and str(ie.LocationURL) == "http://home.cnblogs.com/":    
        break
    sleep(1)
print "登陆成功" 
print '你的昵称是：'
print ie.Document.getElementById('lnk_current_user').title

#博客园只能登录一次，注销
print '注销！'
ie.Navigate(loginouturl)

5.播放mp3文件：http://www.sharejs.com/codes/python/5733

from win32com.client import Dispatch
mp = Dispatch("WMPlayer.OCX")
# use an mp3 file you have ...
#tune = mp.newMedia("C:/Program Files/Common Files/HP/Memories Disc/2.0/audio/Swing.mp3")
# or copy one to the working folder ...
#tune = mp.newMedia("Bier1.mp3")
# you can also play wma files, this cool sound came with XP ...
tune = mp.newMedia("C:/WINDOWS/system32/oobe/images/title.wma")
mp.currentPlaylist.appendItem(tune)
mp.controls.play()
# to stop playing use
raw_input("Press Enter to stop playing")
mp.controls.stop()

　　真心感觉这个东西很强大呀！言归正传，该上将word和ppt转化为txt的代码了！如下：

#coding:utf-8
import win32com
import win32con
import win32gui
import codecs
from win32com.client import Dispatch
import pythoncom

class MSOffice2txt():
    def __init__(self, fileType=['doc','ppt']):
        self.docCom = None
        self.pptCom = None
        pythoncom.CoInitialize()
        if type(fileType) is not list:
            return 'Error, please check the fileType, it must be list[]'
        for ft in fileType:
            if ft == 'doc':
                self.docCom = self.docApplicationOpen()
            elif ft == 'ppt':
                self.pptCom = self.pptApplicationOpen()

    def close(self):
        self.docApplicationClose(self.docCom)
        self.pptApplicationClose(self.pptCom)

    def docApplicationOpen(self):
        docCom = win32com.client.Dispatch('Word.Application')
        docCom.Visible = 1
        docCom.DisplayAlerts = 0
        docHwnd = win32gui.FindWindow(None, 'Microsoft Word')
        win32gui.ShowWindow(docHwnd, win32con.SW_HIDE)
        return docCom

    def docApplicationClose(self,docCom):
        if docCom is not None:
            docCom.Quit()

    def doc2Txt(self, docCom, docFile, txtFile):
        doc = docCom.Documents.Open(FileName=docFile,ReadOnly=1)
        doc.SaveAs(txtFile, 2)
        doc.Close()

    

    def pptApplicationOpen(self):
        pptCom = win32com.client.Dispatch('PowerPoint.Application')
        pptCom.Visible = 1
        pptCom.DisplayAlerts = 0
        pptHwnd = win32gui.FindWindow(None, 'Microsoft PowerPoint')
        win32gui.ShowWindow(pptHwnd, win32con.SW_HIDE)
        return pptCom

    def pptApplicationClose(self, pptCom):
        if pptCom is not None:
            pptCom.Quit()

    def ppt2txt(self, pptCom, pptFile, txtFile):
        ppt = pptCom.Presentations.Open(pptFile,ReadOnly=1, Untitled=0, WithWindow=0)
        f = codecs.open(txtFile,"w",'gb18030')
        slide_count = ppt.Slides.Count
        for i in xrange(1,slide_count + 1):
            shape_count = ppt.Slides(i).Shapes.Count
            for j in xrange(1,shape_count + 1):
                if ppt.Slides(i).Shapes(j).HasTextFrame:
                    s = ppt.Slides(i).Shapes(j).TextFrame.TextRange.Text
                    f.write(s)
        f.close()
        ppt.Close()

    def translate(self, filename, txtFilename):
        if filename.endswith('doc') or filename.endswith('docx'):
            if self.docCom is None:
                self.docCom = self.docApplicationOpen()
            self.doc2Txt(self.docCom, filename, txtFilename)
            return True
        elif filename.endswith('ppt') or filename.endswith('pptx'):
            if self.pptCom is None:
                self.pptCom = self.pptApplicationOpen()
            self.ppt2txt(self.pptCom, filename, txtFilename)
            return True
        else:
            return False

if __name__=='__main__':
    msoffice = MSOffice2txt()
    filename = u'F:\study.docx'
    if msoffice.translate(filename, 'temp.txt'):
        print 'Successed!'
    else:
        print 'Failed!'
    msoffice.close()