represent any Unicode character by means of a markup string coded in us-ascii

[lkrubner]

>Alan J. Flavell Oct 7 2004, 1:44 pm show options

On Thu, 7 Oct 2004, Shmuel (Seymour J.) Metz wrote:
at 08:24 PM, "Alan J. Flavell" <[e-mail]> said:

>I think you mean "multiple character encoding schemes".

Yes, although a different character set would imply a different
encoding scheme.

Absolutely not. That's the whole point!

In (X)HTML you can (if you so choose) represent any Unicode character
by means of a markup string coded in us-ascii, even. The use of other
encoding schemes is merely a convenience when the desired character
repertoire fits a particular pattern, but whichever encoding scheme
you choose, you still - in principle - have access to any other
Unicode character you need, by means of &-notation.

I could change any Unicode character to its html notation, if only I
had a way to find out the Unicode value of the characters in the string
I'm given. But given a random set of string inputs, possibly copy and
pasted from WordPerfect or Microsoft Word or BBedit on a Mac, I don't
know how to find the Unicode value of those characters.

[Alan J. Flavell]

On Sat, 27 May 2005, lk******@geocities.com wrote:

I could change any Unicode character to its html notation, if only I
had a way to find out the Unicode value of the characters in the
string I'm given.
What's the context here? In order to know what "characters" you have
been given, you need to know what encoding they are represented in. If
they're not an encoding of Unicode itself, then you can normally refer
to the appropriate cross-mapping table at the Unicode site to
determine the corresponding hexadecimal Unicode value. That's the
value that you'd need (converted to decimal if you so choose) in the
&#...; representation in HTML.
But given a random set of string inputs, possibly copy and pasted
from WordPerfect or Microsoft Word or BBedit on a Mac, I don't know
how to find the Unicode value of those characters.

If you're talking about forms submission, then the usual arrangement
is that the characters are submitted using the same character encoding
as the page which contains the form which they're submitted from.
For working with modern browsers, I'd normally recommend that you use
utf-8 for that. (No good with NN4.*).

http://ppewww.ph.gla.ac.uk/~flavell/...form-i18n.html

(But if you've been sent utf-8 and you're willing to store files in
utf-8 then you don't really *have* to use &#...; representation
anyway. It's your choice, really.)

You're then reliant on what the client platform actually does when
copy/pasting from another application window into the form.

That can have some unexpected glitches, since Word (especially older
versions) has a nasty habit of changing to a non-standard font e.g
Symbol and inserting a Latin letter (e.g W) to get a symbol (e.g Omega
or Ohm sign). This doesn't really work in HTML - MS of course will
fool its users by repeating the error in MSIE, but a properly
conforming www-compatible browser will display the W that the markup
asked for - not the symbol that was intended.