Re: UTF-8 breaks 'ascii'

Re: UTF-8 breaks 'ascii'

[Charles Wilson]

Kenneth wrote:
> Maybe it shouldn't attempt to display code points 128..255
> since they aren't really ascii.

You're probably right.  Originally, ascii didn't display those code
points; I added it back when a big concern was getting the "correct"
fonts for rxvt-windows and cmd.exe so that line graphics (pstree -G,
etc) would display properly.  It was a quick way to tell that "yes,
those line art characters are present at the appropriate char values".

This is less of a concern now -- because (a) rxvt is, if not deprecated,
at least discouraged, and (b) other terminals like rxvt-unicode, xterm,
and mintty have different and better mechanisms to support such line
art.  And cygwin-in-cmd.exe "just works" too, thanks to improvements in
cygwin's terminal handling code.

So...yeah, I'll remove those in the next release.

--
Chuck

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Re: UTF-8 breaks 'ascii'

[Andy Koppe]

Charles Wilson wrote:
> Kenneth wrote:
>> Maybe it shouldn't attempt to display code points 128..255
>> since they aren't really ascii.
>
> You're probably right.  Originally, ascii didn't display those code
> points; I added it back when a big concern was getting the "correct"
> fonts for rxvt-windows and cmd.exe so that line graphics (pstree -G,
> etc) would display properly.  It was a quick way to tell that "yes,
> those line art characters are present at the appropriate char values".

I still find it useful for much the same reason, when testing the
plethora of supported charsets. How about disabling the extended code
points by default and having an option for enabling them?

Andy

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Re: UTF-8 breaks 'ascii'

[Charles Wilson]

On 4/14/2010 11:45 AM, Charles Wilson wrote:
> Kenneth wrote:
>> Maybe it shouldn't attempt to display code points 128..255
>> since they aren't really ascii.
> 
> You're probably right.  Originally, ascii didn't display those code
> points; I added it back when a big concern was getting the "correct"
> fonts for rxvt-windows and cmd.exe so that line graphics (pstree -G,
> etc) would display properly.  It was a quick way to tell that "yes,
> those line art characters are present at the appropriate char values".

Andy Koppe wrote:
> I still find it useful for much the same reason, when testing the
> plethora of supported charsets. How about disabling the extended code
> points by default and having an option for enabling them?

Done, for 1.4.6: I added a new option '--extended'.

--
Chuck



--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

UTF-8 breaks 'ascii'

[Nellis, Kenneth]

Cygwin defaulting to UTF-8 breaks the 'ascii' program.
It seems 'ascii' presumes a character encoding of LATIN1.
To get intended output:

ascii | iconv -f LATIN1 -t UTF-8

I would have considered this an upstream problem except for
the reference to cygutils:

$ ascii --help
ascii is part of cygutils version 1.4.2
  Prints nicely formatted table of the ascii character set
...

Maybe it shouldn't attempt to display code points 128..255
since they aren't really ascii.

--Ken Nellis