In Python 3.8, how do I load image data from a file for inclusion ina JSON object?

Question

I'm using Python 3.8. I want to get image byte data into a JSON object. So I tried this

with open(os.path.join(dir_path, "../image_data", "myimg.jpg"), mode='rb') as img_file:
    image_data = img_file.read().decode("utf-16")
    my_json_data = {
        "image_data": image_data
        ...
    }

but the image_data = line is giving this error:

UnicodeDecodeError: 'utf-16-le' codec can't decode bytes in position 0-1: illegal UTF-16 surrogate

What's the proper way to load data for inclusion into a JSON object?

Probably because the image you're reading isn't encoded as utf-16? — Pranav Hosangadi, Aug 17 '20 at 20:18
Image data is usually binary (as appropriately handled by ``mode='rb'``). Why do you assume it is utf-16? — MisterMiyagi, Aug 17 '20 at 20:19
How are you planning to encode it for the json file - since JSON is text only don;t you need to encode it using base64 or similar ? — Tony Suffolk 66, Aug 17 '20 at 20:23

score 0 · Answer 1 · answered Aug 17 '20 at 20:21

decode and encode work on character data. You use them to covert between character encoding formats, such as utf-8 and ASCII. It doesn't make sense to take pure binary data -- such as an image -- and try to convert it to characters. Not every binary value can be converted to character; most character formats leave a few values as reserved or unused.

What you need is a simple raw byte format. Read the file as a sequence of bytes; this preserves the binary form, making it easy for your eventual JSON consumer to utilize the information.

In Python 3.8, how do I load image data from a file for inclusion ina JSON object?

1 Answers1