使用python批量提取docx文档中的图片

注意：此方法只能用于docx文档，doc的不行，doc的得另存为Html文件，进入到上述网页文件所在的文件夹中，你会发现有一个名称为“文件名.files”（如“图片.files”），进入其中，前面 Word 文档中的图片就一一被保存在里面了。

在解压的时候遇到点问题，1，不能解压，2，不能删除文件夹，3，不能移动文件，用try解决了

#revised by Stephen Shen 2020-3-8 19:51:46

#to extract pics from *.docx files

import zipfile

import os,shutil

import datetime

# import send2trash

# from tempfile import TemporaryDirectory

def renameFile(dstpath):

    fdirname,fbasename=os.path.split(dstpath)

    #文件名相同但大小不同

    fname,fext=os.path.splitext(fbasename)

    nowtime=datetime.datetime.now()                

    strtime=str(nowtime.year)+str(nowtime.month)+str(nowtime.day)+str(nowtime.hour)+str(nowtime.minute)

    newfbasename=fname+'-'+strtime+fext

    dstpath=os.path.join(fdirname,newfbasename)

    return dstpath

def extractPics(doc_inpath,outDir,zipDir):

    

    file_ext=['.docx']

    dirPath,fname,fext=splitPath(doc_inpath)

    if fext in file_ext:

        print(doc_inpath+' is extracting...')

        fnameNew=fname.strip()

        out_dir =os.path.join(outDir,fnameNew)

        if not os.path.exists(out_dir):#文件夹不存在

            if len(os.path.split(out_dir)[-1])>200:

                print('dirname is too long :'+out_dir)

            else:

                os.mkdir(out_dir)

        else:#文件夹已存在

            nowtime=datetime.datetime.now()                

            strtime=str(nowtime.year)+str(nowtime.month)+str(nowtime.day)+str(nowtime.hour)+str(nowtime.minute)

            out_dir=os.path.join(outDir,fnameNew+'-'+strtime)

        zip_path=os.path.join(zipDir,'docx.zip')

        shutil.copyfile(doc_inpath,zip_path)

        # zip_path=os.path.join(zip_dir,fnameNew+'.zip')

        # os.rename(copy_path,zip_path)

        try:

            with zipfile.ZipFile(zip_path,'r') as f:

                f.extractall()

        except:

            print(zip_path+' cannot be extracted')

        else:

            picsDir=os.path.join(zipDir,'word/media')

            if os.path.exists(picsDir):

                for pic in os.listdir(picsDir):

                    oldpic=os.path.join(picsDir,pic)

                    newpic=os.path.join(out_dir,pic)

                    try:

                        shutil.move(oldpic,newpic)

                    except:

                        print(doc_inpath+' is skipped')

        

        filelist=os.listdir(zipDir)                

        for f in filelist:

            filepath = os.path.join(zipDir, f )   

            if os.path.isfile(filepath):

                os.remove(filepath)

            elif os.path.isdir(filepath):

                shutil.rmtree(filepath,True)

def splitPath(doc_inpath):

    dirPath=os.path.dirname(doc_inpath)

    fname,fext=os.path.splitext(os.path.basename(doc_inpath))

    return dirPath,fname,fext

outDir=r'k:/fileExtracted/imagesFromDocs'

zipDir=r'k:/fileExtracted/zip'

# docDir=r'K:\imagsDocx'

docDir=r'K:/docs'

if not os.path.exists(outDir):

    os.makedirs(outDir)

if not os.path.exists(zipDir):

    os.makedirs(zipDir)

for folder,subFolderName,files in os.walk(docDir):

    for file in files:

        doc_inpath=os.path.join(folder,file)

        os.chdir(zipDir)

        extractPics(doc_inpath,outDir,zipDir)

        

print('done')