Re: @ SCOS Message Format ?

On 24/02/2025 11:18, Richard Heathfield wrote:

On 24/02/2025 10:45, Richard Harnden wrote:

On 24/02/2025 10:28, Richard Heathfield wrote:

On 24/02/2025 10:17, Richard Harnden wrote:

On 24/02/2025 09:51, David Entwistle wrote:

On Sun, 23 Feb 2025 10:55:41 +0000, Richard Heathfield wrote:
>

I did. I found no articles under your name in the current feed for that
group.
I have found nothing of the kind in that group.
>

>
Oh dear. Stephan found them. Perhaps your reader. ;)
>
they should start at:
>
Subject: Character Test
Message-ID: <vpeakb$cgsc$1@dont-email.me>

>
I think you said uk.telecom.  That is uk.test :)

>
That would certainly explain it. Test groups tend to be huge. Let David post here if he wants us to read his examples.
>

>
That one isn't.

 Okay. Found them.
 

>
Anyway, in my thunderbird: x^2 superscripts, x ^ 2 does not.

 In mine, neither do.
 x^2 (x hat 2) is perfectly valid C code, so superscripting it wrongly changes its meaning.
 Given this input:
 32   ^  ^ *
...
126 ~ ^~ ^~*
 scos2 e 98 76 gives this output:

You don't need the e/d flag, just negate the numbers.
So: -5 -78 (the 98 has clocked around to 5).
It's more rot13ny that way.  It was Ben's idea.

 8q   l  U F
...
vfS X /@ zwT
 $md5sum charset.txt
d5c6d06587dbac07fed831293ff0580d  charset.txt
$ md5sum charset.scos2
87ea4967605a5ba4d69ff6cf0fb541f5  charset.scos2
 Anyone get anything different?
 

I get the same (without the scos headers).
One thing that is mildly annoying: if the encoded line of text happens to start with a "> ", then thunderbird thinks its quoted text.