CODEWITHSUNDEEP
pythoniobinarybase64
kramer65
kramer65

Reputation: 53873

How to decode base64 file into binary in Python?

I'm building a system which handles pdf file data (for which I use the PyPDF2 lib). I now obtain a base64 encoded PDF which I can decode and store correctly using the following:

import base64
# base64FileData  <= the base64 file data
fileData = base64.urlsafe_b64decode(base64FileData.encode('UTF-8'))
with open('thefilename.pdf', 'w') as theFile:
    theFile.write(fileData)

I now want to use this fileData as a binary file to split it up, but when I do type(fileData), the fileData turns out to be a <type 'str'>. How can I convert this fileData to be a binary (or at least not a string)?

All tips are welcome!

[EDIT]

if I do open(fileData, 'rb') I get an error, saying

TypeError: file() argument 1 must be encoded string without NULL bytes, not str

To remove the null bytes I tried, fileData.rstrip(' \t\r\n\0') and fileData.rstrip('\0') and fileData.partition(b'\0')[0], but nothing seems to work. Any ideas?

[EDIT2]

The thing is that I pass this string to the PyPDF2 PdfFileReader class, which on lines 909 to 912 does the following (in which stream is the fileData I provide):

if type(stream) in (string_type, str):
    fileobj = open(stream, 'rb')
    stream = BytesIO(b_(fileobj.read()))
    fileobj.close()

So because its a string, it assumes it is a filename, after which it tries to open the file. This then fails with a TypeError. So before feeding the fileData to the PdfFileReader I need to somehow convert it to something else than str so that it doesn't try to open it, but just considers fileData a file on itself. Any ideas?

Upvotes: 1

Views: 9687

Answers (2)

Muhammad Subair
Muhammad Subair

Reputation: 51

Example your input data is came from this:

with open(local_image_path, "rb") as imageFile:
    str_image_data = base64.b64encode(imageFile.read())

then to get the binary in variable you can try:

import io
import base64

binary_image_data = io.BytesIO(base64.decodebytes(str_image_data))

Upvotes: 2

Cake
Cake

Reputation: 81

Hence the open's binary mode you have to use 'wb' else it gets saved as "text" basically.

import base64
# base64FileData  <= the base64 file data
fileData = base64.urlsafe_b64decode(base64FileData.encode('UTF-8'))
with open('thefilename.pdf', 'wb') as theFile:
    theFile.write(fileData)

Upvotes: 3

Related Questions