Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Same Image, but different base64

I used the following function to create a base64 encoded string of my Gravatar image (https://www.gravatar.com/avatar/cd5415f97afbe0177ba35ae31fbfd0db):

final BASE64Encoder encoder = new BASE64Encoder(); 

String encoded = encoder.encode(inputStreamToByteArray(is));
encoded = encoded.replaceAll("\r?\n", "");

return encoded;

I ran the method a couple of days ago and got the following base64 encoded string:

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

Just today I executed the method again, using the same image. But instead of retrieving the same encoded string, I received:

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

As you may notice, the first couple of characters are the same, but if you look at the end of the strings, you notice directly that they are completely different. Why is that? The image is still my Gravatar image (e.g., using http://codebeautify.org/base64-to-image-converter shows the same image for both strings).

Thanks for any hint, explanation! Is there anything wrong with my implementation? Is some of the implementation time or location dependent? How can I get the same base64 string for the same image?

PS: inputStreamToByteArray is implemented as follows:

public static byte[] inputStreamToByteArray(final InputStream is) {
    final ByteArrayOutputStream buffer = new ByteArrayOutputStream();
    final byte[] data = new byte[16384];

    try {
        int nRead;
        while ((nRead = is.read(data, 0, data.length)) != -1) {
            buffer.write(data, 0, nRead);
        }
        buffer.flush();
    } catch (final IOException e) {
        return null;
    } finally {
        try {
            buffer.close();
        } catch (final IOException e) {
            // ignore
        }
    }

    return buffer.toByteArray();
}
like image 670
Philipp Avatar asked Oct 04 '16 20:10

Philipp


People also ask

Is Base64 always same?

Artjom B. Base64 is not encryption. But yes, different input strings will always encode to different Base64-encoded strings, and the same input string will always encode to the same Base64-encoded string. It's not a hash though, so small changes in the input will only result in small changes in the output.

What does == mean in Base64?

This leaves us with one byte (8 bits); we pad the remaining byte with two bytes with value zero and encode the last sequence. Two equal signs ("==") are added to the encoded string. The size has extra two bytes (remainder of 16 bits when divided by 24): Same as above, but we pad just one byte.

Does Base64 encoding reduce image quality?

Encoding to/from Base64 is completely lossless.

Why does Base64 have ==?

It's padding. From http://en.wikipedia.org/wiki/Base64: In theory, the padding character is not needed for decoding, since the number of missing bytes can be calculated from the number of Base64 digits. In some implementations, the padding character is mandatory, while for others it is not used.


1 Answers

Converting both images back to .jpg, and using http://regex.info/exif.cgi, the following header comments appear:

One:

CREATOR: gd-jpeg v1.0 (using IJG JPEG v80), quality = 90

Other:

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), quality = 90

(can't remember which order I converted the images in, but there is a version change either way)

As you can see, they have upgraded the JPEG compressor (or spec?) to a newer version, resulting in a different JPEG encoding, or at least difference in the comments.

So to answer your question, there does not seem to be an issue with your base64 converter, but just that the two images actually are different

like image 106
Joseph Young Avatar answered Sep 29 '22 10:09

Joseph Young