s3 read is pretty savage - it's designed for reading
text objects, but assumes they don't have a line terminator at the
end? I would argue that an object with no terminating newline isn't a
text file, it's binary data which happens to be in
ascii/utf8/whatever. This is certainly mitigated somewhat by the
existence of s3 cat, but it'd be nice to have a utf8-handling
human-oriented read which doesn't assume its input isn't a valid
text file.
Are there hysterical raisins such that everything will break if we
remove the extra newline, or any other objections to changing s3 read to just decode utf-8 and print it?
s3 read
is pretty savage - it's designed for reading text objects, but assumes they don't have a line terminator at the end? I would argue that an object with no terminating newline isn't a text file, it's binary data which happens to be in ascii/utf8/whatever. This is certainly mitigated somewhat by the existence ofs3 cat
, but it'd be nice to have a utf8-handling human-orientedread
which doesn't assume its input isn't a valid text file.Are there hysterical raisins such that everything will break if we remove the extra newline, or any other objections to changing
s3 read
to just decode utf-8 and print it?