python's YENC.DECODE -> weird output

[John Savage]

I save posts from a midi music newsgroup, some are encoded with
yenc encoding. This gave me an opportunity to try out the decoders
in Python. The UU decoder works okay, but my YENC effort gives
results unexpected:

import yenc, sys

fd1=open(sys.ar gv[1],'r')
#yenc.encode(sy s.argv[1],"outfile.yenc" ,bytes=0)
yenc.decode(sys .argv[1],"outfile.mid", bytes=0,crc_in= '')

I confirmed that yenc.decode exactly reverses yenc.encode, BUT the
encoding itself seems to differ from the USENET standard. That is,
when I decode USENET files the result isn't a valid music file.
(I did try both with and w/o the headers.)

Maybe it uses a different character set? I can't quite put my
finger on what might be happening. What I can say is that the
yenc coding from the newsgroup article, when viewed with Linux
'more', displays roughly 10% of its characters as a question mark,
whereas when I give Python's yenc.encode a binary music file and
view its output using 'more', it displays about 90% of the output
as a question mark.

Any ideas?
--
John Savage (my news address is not valid for email)

[John Machin]

Re: python's YENC.DECODE -> weird output

On Jul 16, 11:33 pm, John Savage <rooksw...@subu rbian.com.auwro te:

I save posts from a midi music newsgroup, some are encoded with
yenc encoding. This gave me an opportunity to try out the decoders
in Python. The UU decoder works okay, but my YENC effort gives
results unexpected:
>
import yenc, sys
>
fd1=open(sys.ar gv[1],'r')
#yenc.encode(sy s.argv[1],"outfile.yenc" ,bytes=0)
yenc.decode(sys .argv[1],"outfile.mid", bytes=0,crc_in= '')
>
I confirmed that yenc.decode exactly reverses yenc.encode, BUT the
encoding itself seems to differ from the USENET standard. That is,
when I decode USENET files the result isn't a valid music file.
(I did try both with and w/o the headers.)
>
Maybe it uses a different character set? I can't quite put my
finger on what might be happening. What I can say is that the
yenc coding from the newsgroup article, when viewed with Linux
'more', displays roughly 10% of its characters as a question mark,
whereas when I give Python's yenc.encode a binary music file and
view its output using 'more', it displays about 90% of the output
as a question mark.
>
Any ideas?

1. It isn't "Python's yenc". Consider directing your question to
whoever (if anyone) is maintaining whichever third-party module you
are talking about.

2. Consider using a tool that will display what is going on in
characters and hex e.g. http://linux.about.com/library/cmd/blcmdl1_hexdump.htm

[Terry Reedy]

Re: python's YENC.DECODE -&gt; weird output



John Savage wrote:

I save posts from a midi music newsgroup, some are encoded with
yenc encoding. This gave me an opportunity to try out the decoders
in Python. The UU decoder works okay, but my YENC effort gives
results unexpected:
>
import yenc, sys
>
fd1=open(sys.ar gv[1],'r')
#yenc.encode(sy s.argv[1],"outfile.yenc" ,bytes=0)
yenc.decode(sys .argv[1],"outfile.mid", bytes=0,crc_in= '')
>
I confirmed that yenc.decode exactly reverses yenc.encode, BUT the
encoding itself seems to differ from the USENET standard.

I do not believe that there really is a yenc standard. I have noticed
that the yenc decoder in Mozilla Thunderbird never decodes posts from
certain posters (and hence certain encoders) whereas uu decoding only
sporadically fails.