Re: locale specific ordering in EN_US -- why is a<A<b<B<y<Y<z<Z?

[Linda Walsh]

Greg Wooledge wrote:

> On Mon, May 21, 2012 at 12:19:26PM -0700, Linda Walsh wrote:
> > Greg Wooledge wrote:
> > > For instance, on HP-UX 10.20, in the en_US.iso88591 locale:
> > >    A  a  ...  B  b
> > > Meanwhile, on Debian 6.0, in the en_US.iso88591 locale:
> > >    a A   ...  b B
>
> > So which is correct?
>
> Both.  Locale collating order is determined by the OS.  You cannot
> rely on it, unless you set the LC_COLLATE variable to "C" or "POSIX",
> in which case you get ASCII behavior (accented letters are not part
> of the character set at all).
>
> > Anyone wanting to reference an upper or lower case range
> > [a-z] or [A-Z], is gonna hurt from this.
>
> Correct.

----
        This is a prime example of Posix being stupid and bad for
computer science.

        They take a deterministic behavior and define it to be
non-deterministic and break 1000's of programs.

        They cannot justify this... as they are supposed to document
current practice -- which has never been to consider the interpretation
of a-z/A-Z as random!

        Thus they are violating their own rules!    How can anyone follow
such lame directions?  Who in their right mind would have voted to
make ranges "worthless"....i.e. -- established, standard practice has never
been for such ranges to be worthless -- yet that is exactly what they
voted for.

        How is posix following it's own rules?   If they don't follow
their own rules -- how can anyone be following these new specifications
which are obviously in conflict with established implementation?