python - ValueError: invalid literal for int() with base 10: ' ' when it worked before

Question

I'm having some issues with my program, basically what I'm trying to do is Stenography, insert an image into another image and then extract the secret image.

My program is able to insert just fine, but extracting it I get this error.

It was working fine the other day, I was able to read and write perfectly. I am also using the same BMP images as the other day as well. Anyone know the issue? I didn't change any of the code from it's working state, I even uploaded to github to make sure that version worked.

There are 3 files of code that is on my github.

https://github.com/am3ience/Steganography/blob/master/dcstego.py

https://github.com/am3ience/Steganography/blob/master/dcimage.py

https://github.com/am3ience/Steganography/blob/master/dcutils.py

The problem seems to be happening in dcutils.py in my read function. Any help would be amazing, i'm stumped.

#examine the lsb of each pixel, grouping into bytes
#check for nulls to signify if we are dealing with data or header info
#bytes determined to be data result in the hidden file
#---------------------------------------------------------------
def read(mainimage, output, password):
    lsbByte_Array = []
    dataString = ""
    secretFileName = ""
    lsbString = ""
    count = 0#iterator
    headerReceived=0#flags
    sizeReceived=0
    imageObject = dcimage.openFile(mainimage)
    pixels = imageObject.load()
    imageWidth, imageHeight = imageObject.size

    #cycle through each pixel
    for x in range(imageWidth):
        for y in range(imageHeight):
            r, g, b = pixels[x, y]
            #trim so we are dealing with only the least significant bit
            redPixel = str(bin(r)[2:].zfill(8))[7]
            greenPixel = str(bin(g)[2:].zfill(8))[7]
            bluePixel = str(bin(b)[2:].zfill(8))[7]
            secretBits = [redPixel, greenPixel, bluePixel]

            #for each of rgb
            for i in range(0,3):
                #check if our flags are set
                if (headerReceived == 0 or sizeReceived == 0):
                    lsbString += secretBits[i]

                    #verify each byte
                    if len(lsbString) == 8:
                        lsbByte_Array.append(lsbString)

                        #check if we have received a NULL byte
                        if lsbString == "00000000":
                            if headerReceived == 0:

                                #convert the the bit array into an ascii String
                                #set flag when header and size was received
                                fileName = ''.join(binascii.unhexlify('%x' % int(b,2)) for b in lsbByte_Array[0:len(lsbByte_Array) - 1])
                                print "File name: " + str(fileName)
                                headerReceived = 1
                            elif sizeReceived == 0:
                                 fileSize = ''.join(binascii.unhexlify('%x' % int(b,2)) for b in lsbByte_Array[0:len(lsbByte_Array) - 1])
                                print "File size: " + fileSize
                                sizeReceived=1

                            #reset the values
                            lsbByte_Array = []
                        lsbString = ""

                #once headers received, resulting data is hidden data
                elif (headerReceived == 1 and sizeReceived == 1):
                    if int(count) < int(fileSize):
                        #keep appending secret bits to the dataString until depleted
                        dataString += secretBits[i]
                        count += 1
                    else:
                        #send to have hidden file created
return dcimage.saveImage(output, dataString)

Answer 1

fileSize = ''.join(binascii.unhexlify('%x' % int(b,2)) for b in 
                   lsbByte_Array[0:len(lsbByte_Array) - 1])

if lsbByte_Array is an empty array then fileSize will be an empty string, which is obviously why int(fileSize) fails.

You should probably use a default value, for example

fileSize = ''.join(binascii.unhexlify('%x' % int(b,2)) for b in 
                   lsbByte_Array[0:len(lsbByte_Array) - 1]) or 0

As a side note, stuff like lsbByte_Array[0:len(lsbByte_Array) - 1]) can simply be written as lsbByte_Array[0:-1].