Documentation bug with R2 multipart upload md5 example?

The documentation for calculating the final etag of a multipart upload seems like it might have a bug.

For example, consider a multipart upload with two parts. If they have the ETags bce6bf66aeb76c7040fdd5f4eccb78e6 and 8165449fc15bbf43d3b674595cbcc406 respectively, the ETag of the completed multipart upload will be f77dc0eecdebcd774a2a22cb393ad2ff-2 .

However, this is what I see when I try to calculate the expected final etag myself.

% md5 -s bce6bf66aeb76c7040fdd5f4eccb78e68165449fc15bbf43d3b674595cbcc406  
MD5 ("bce6bf66aeb76c7040fdd5f4eccb78e68165449fc15bbf43d3b674595cbcc406") = 158668b0d382a2d588dd141dafb5ec6f

I’ve concatenated the two etags and used that as input to md5 -s, but the result doesn’t match what’s in the docs.

Am I doing this wrong?

See: ETags

Also, nit, the docs contain a mis-spelling. concanated should be concatenated. :wink: