Extract pictures from Word and Excel with Python

Question

I was searching for a way to strip out pictures from these file types and this is the solution I came up with. It iterates through a given directory structure, copies any files with the proper extension, and renames the copy to filename.zip. Then it navigates through the zip structure and extracts all picture type files with the proper extension, and renames them to the original file name, with a number for uniqueness. Finally, it deletes the extracted directory trees it created.

Extracting pictures from text documents is part of my job, so this will actually save my company thousands of hours in the long run.

All of the code is below, and what I'm really asking is: Is there a better way? Is there something more efficient? Can it be scaled to include other formats? Could the text be extracted into a txt - for loading times on word vs notepad?

This solution works on my Linux machine, and I can extract the pictures, but I've yet to test on a Windows system.

#!/usr/bin/python3

import shutil
import os
import zipfile

def zipDoc(aFile,dirPath):
    dotNDX = aFile.index(".") # position of the .
    shortFN = aFile[:dotNDX] # name of the file before .
    zipName = dirPath + shortFN + ".zip" # name and path of the file only .zip
    shutil.copy2(dirPath + aFile, zipName) # copies all data from original into .zip format
    useZIP = zipfile.ZipFile(zipName) # the usable zip file
    return useZIP # returns the zipped file 

def hasPicExtension(aFile): # if a file ends in a typical picture file extension, returns true
    picEndings = [".jpeg",".jpg",".png",".bmp",".JPEG"".JPG",".BMP",".PNG"] # list of photo extensions
    if aFile.endswith(tuple(picEndings)): # turn the list into a tuple, because .endswith accepts that
        return True     
    else: # if it doesn't end in a picture extension
        return False

def delDOCXEvidence(somePath): # removes the .docx file structures generated
    ##################################################################
    # Working Linux code:
    os.rmdir(somePath + "/word/media") # removes directory
    os.rmdir(somePath + "/word") # removes more directory
    ##################################################################

    ##################################################################
    # Untested windows code:
    # os.rmdir(somePath + "\\\\word\\\\media") # removes directory
    # os.rmdir(somePath + "\\\\word") #removes more directory
    ##################################################################

def delXLSXEvidence(somePath): # removes the .xlsx file structures generated
    ##################################################################
    # Working Linux code:
    os.rmdir(somePath + "/xl/media") # removes directory
    os.rmdir(somePath + "/xl") # removes more directory
    ##################################################################

    ##################################################################
    # Untested windows code:
    # os.rmdir(somePath + "\\\\xl\\\\media") # removes directory
    # os.rmdir(somePath + "\\\\xl") #removes more directory
    ##################################################################

def extractPicsFromDir(dirPath=""):
# when given a directory path, will extract all images from all .docx and .xlsx file types
    if os.path.isdir(dirPath): # if the given path is a directory
        for dirFile in os.listdir(dirPath): # loops through all files in the directory
            dirFileName = os.fsdecode(dirFile) # strips out the file name
            if dirFileName.endswith(".docx"):
                useZIP = zipDoc(dirFile,dirPath) # turns it into a zip
                picNum = 1 # number of pictures in file
                for zippedFile in useZIP.namelist(): # loops through all files in the directory
                    if hasPicExtension(zippedFile): # if it ends with photo
                        useZIP.extract(zippedFile, path=dirPath) # extracts the picture to the path + word/media/
                        shutil.move(dirPath + str(zippedFile),dirPath + dirFileName[:dirFileName.index(".")] + " - " + str(picNum)) # moves the picture out
                        picNum += 1
                delDOCXEvidence(dirPath) # removes the extracted file structure
                os.remove(useZIP.filename) # removes zip file
                # no evidence
            if dirFileName.endswith(".xlsx"):
                useZIP = zipDoc(dirFile,dirPath) # turns it into a zip
                picNum = 1 # number of pictures in file
                for zippedFile in useZIP.namelist(): # loops through all files in the directory
                    if hasPicExtension(zippedFile): # if it ends with photo
                        useZIP.extract(zippedFile, path=dirPath) # extracts the picture to the path + word/media/
                        shutil.move(dirPath + str(zippedFile),dirPath + dirFileName[:dirFileName.index(".")] + " - " + str(picNum)) # moves the picture out
                        picNum += 1
                delXLSXEvidence(dirPath) # removes the extracted file structure
                os.remove(useZIP.filename) # removes zip file
                # no evidence

    else:
        print("Not a directory path!")
        exit(1)


uDir = input("Enter your directory: ")
extractPicsFromDir(uDir)

score 2 · Answer 1 · answered Mar 13 '19 at 07:29

Excel files are in the form of zip file.It is easy to extract images from excel or docx file:

import zipfile
from PIL import Image, ImageFilter
import io

blur = ImageFilter.GaussianBlur(40)

def redact_images(filename,FilePath):
    outfile = filename.replace(".xlsx", "_redacted.xlsx")
    with zipfile.ZipFile(filename) as inzip:
        with zipfile.ZipFile(outfile, "w") as outzip:
            i = 0
            for info in inzip.infolist():
                name = info.filename
                content = inzip.read(info)
                if name.endswith((".png", ".jpeg", ".gif")):
                        fmt = name.split(".")[-1]
                        Name = name.split("/")[-1]
                        img = Image.open(io.BytesIO(content))
                        img.save(FilePath + str(Name))
                        outb = io.BytesIO()
                        img.save(outb, fmt)
                        content = outb.getvalue()
                        info.file_size = len(content)
                        info.CRC = zipfile.crc32(content)
                        i += 1
                outzip.writestr(info, content)

filename : Location of input excel file

FilePath : Location to save extracted images

Extract pictures from Word and Excel with Python

1 Answers1

Linked