Base128 encoding, the intersection of latin1 and utf-8, which is basically ASCII, the most memory-efficient string encoding that can be written to disk as utf-8 without ballooning in size.
Hi, I have to encode large binary buffers and found out that using base128 give me the best results in terms of disk space (UTF-8).
However, I can't just send the whole buffer because Chrome crashes (out of memory), so I have to stream the buffer in chunks. Concatenating Buffer128.encode(chunk) does not work so I was thinking if you see possible a streamed version of the encoder/decoder.