I rebased and documented the pull request #85.
Please check the patch and documentation,
and if there's something wrong or i can do further, please let me know.
This looks great. Just a quick thought, do you think it'd be better to decode the buffer after all the chunks have been received (then you could get rid of that comment about potential bugs on multi-byte characters spanning chunks).
It seems like it'd be fairly easy to do.
Either way, If I don't hear back from you I'll merge this :-)
Thank you for the reply.
I'll hack into it.
I commited a patch to decode the buffer after all the chunks've been received.
Plz check the patch, and if there's something wrong, let me know.
Rebased and documented by @noraesae
Process decoding after all the chunks have been received.
Prevent bugs that can occur when decode the large chunk with