Text 16860, 50 rader
Skriven 2013-11-23 18:46:46 av Maurice Kinal (1:261/38.9)
  Kommentar till text 16856 av mark lewis (1:3634/12.71)
Ärende: utf-8 fidonet messaging
===============================
Hey mark!

 ml> your_msg -> quoted in a reply -> i replied to the reply

Ah!  Yes you are correct.  Sorry about that chief.

 ml> what do you mean "will catch"?

utf-8 identifies itself as such and if there is anything other then ascii or
utf-8 characters within that same file or message (in this case) then it really
isn't utf-8, is it?

In Fidonet-speak, this means that ALL the so-called binary must be stripped out
of the header(s) before the actual message can be properly identified if indeed
the CHRS kludge is set to UTF-8.  The ascii control characters can be left in
since they are ascii which is totally acceptable in utf-8.  Anything else will
cause corruption within that message (eg 0xfe and 0xff in hex-speak).  It is
possible to send mixed encodings within a packet of messages BUT anything
identifying itself as utf-8 MUST be ascii and/or utf-8 characters or it really
isn't a utf-8 message.  This fact makes utf-8 messaging vastly superior to any
other codepage as far as verification is concerned, especially when it concerns
language or graphical characters given that they all have an unique code that
isn't shared by anything else which is not the case in non-unicode 8 bit
codepages including CP437 which some people were suckered into calling extended
or higher ascii for whatever reason.  I suspect evil and demented purposes such
as MSDOS.  :::snicker:::

 ml> read as the two byte character for the lower case o

Ah!  I wasn't sure that was your plan as there was nothing in your message to
indicate it.  Now that you tell me then looking at your original message the
first character as a multibyte character yeilds 'ö' for the 0xc3 + 0xb6 valid
utf-8 code but then that screws up the second character (0x94) which when
converted from cp437 produces the exact same character as the first but then
the act of conversion to utf-8 alters the first character into two seperate
characters as I already posted you about.  This is a prime example of why you
can't have it both ways and if you had only sent the first multibye sequence
then it would have gotten it right despite you not seeing it in your editor as
such.

Personally I say keep utf-8 messaging pure.  Doing otherwise will reduce it to
the same low level of DOS-think and thus cause far too much grief not unlike
what Microsoft has been doing all along.  ;-)

Life is good,
Maurice

... A Møøse once bit my sister ...
--- GNU bash, version 4.2.45(2)-release (x86_64-unknown-linux-gnu)
 * Origin: Pointy Stick Society - Ladysmith BC, Canada (1:261/38.9)