What does stty's iutf8 actually do (effect)?
On Fri, 17 Aug 2012 19:47:33 -0400, Dan B. wrote:
> The manual page for stty says that the input setting "iutf8" controls
> whether to "assume [that] input characters are UTF-8 encoded."
> What does setting actually do?
> (I understand UTF-8 and its mapping between byte sequences and
> characters. What I don't know is where character decoding and encoding
> are done in Linux input and output components (virtual consoles,
> xterm/etc., tty devices).)
> What does that iutf8 setting actually affect?
As I see it, this concerns the kernel and how it handles the data flaw
input/output coming from pipes redirection between programs or even from
a simple echo/cat command you execute when you're on a tty.
By turning that flag "on" (by default is "on") you are telling the kernel
it has to expect UTF-8 enconding and my guess is that this setting is
aimed to normalize the way console input/output is managed to avoid
problems coming from wrong character encoding transformations.
> Is the terminal (tty?) driver involved in decoding the input byte stream
> into characters for the process attached to the tty device?
> Or does the driver decode bytes into characters only for its own
> checking for special characters (e.g., for stty intf CHAR)?
My two cents for the former :-)
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact firstname.lastname@example.org