SGML Charset explorer..

[Andrew Thompson]

I was recently loading an HTML editor
so I could find the charcode of that
particularly obscure character using the
editor's 'insert special character' dialog.

It occured to me there had to be a
better way. There are probably dozens,
but here is my solution..
http://www.physci.org/codes/charset.jsp

This page is my 'charset explorer', it displays
character codes in a table 456 at a time.

It also has links to a page giving larger
representations of each character. Vis.
http://www.physci.org/codes/char.jsp?char=65
http://www.physci.org/codes/char.jsp?char=84
http://www.physci.org/codes/char.jsp?char=1944

I hope it brings a..
http://www.physci.org/codes/char.jsp?char=9786
...to your mug.

--
Andrew Thompson
* http://www.PhySci.org/ Open-source software suite
* http://www.PhySci.org/codes/ Web & IT Help
* http://www.1point1C.org/ Science & Technology

[Steve Pugh]

"Andrew Thompson" <Se********@www .invalid> wrote:

I was recently loading an HTML editor
so I could find the charcode of that
particularly obscure character using the
editor's 'insert special character' dialog.

It occured to me there had to be a
better way. There are probably dozens,
http://www.eki.ee/letter/ is my reference of choice.
but here is my solution..
http://www.physci.org/codes/charset.jsp

http://www.physci.org/codes/charset....8859-1&frame=1
Characters 127 to 159 in ISO-8859-1 (and all other ISO-8859 encodings)
are control characters. You seem to have some Windows-1252 characters
in there instead.

http://www.physci.org/codes/charset....8859-1&frame=2
There are only 256 characters in ISO-8859-1, so where did these come
from?

http://www.physci.org/codes/charset....8859-5&frame=1
Doesn't actually display any cyrillic characters. Mainly because
you've coded them as &#XXX; and numeric character references in HTML
always refer to unicode.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <st***@pugh.net > <http://steve.pugh.net/>

[Andreas Prilop]

On Wed, 18 Feb 2004, Steve Pugh wrote:

http://www.physci.org/codes/charset....8859-1&frame=1
Characters 127 to 159 in ISO-8859-1 (and all other ISO-8859 encodings)
are control characters. You seem to have some Windows-1252 characters
in there instead.
Blame your own browser!
http://www.physci.org/codes/charset....8859-1&frame=2
There are only 256 characters in ISO-8859-1, so where did these come
from?
The site is a bit confusing. Only "frame=..." is important for the
displayed characters. One and the same document is then sent with
different charset parameters. That should have no effect - but
actually browsers will take a different typeface for each charset
parameter.
http://www.physci.org/codes/charset....8859-5&frame=1
Doesn't actually display any cyrillic characters.

<http://www.physci.org/codes/charset.jsp?cs= iso-8859-5&frame=5>

[Steve Pugh]

Andreas Prilop <nh******@rrz n-user.uni-hannover.de> wrote:

On Wed, 18 Feb 2004, Steve Pugh wrote:

http://www.physci.org/codes/charset....8859-1&frame=1
Characters 127 to 159 in ISO-8859-1 (and all other ISO-8859 encodings)
are control characters. You seem to have some Windows-1252 characters
in there instead.

Blame your own browser!

Blame all my own browsers! Every browser I have incorrectly displays,
for example, ™ as a trademark sign. That's NN4, NN6, NN7, IE5,
IE5.5, IE6, Op5, Op6, Op7, Moz 1.6, Firefox 0.8 and even Lynx.

But the site is claiming that -
"SGML character 153. This is the character "?".
In HTML you would write it:
<p>This is the character "™".</p>"
<http://www.physci.org/codes/char.jsp?char=1 53>
which is just plain wrong.

http://www.physci.org/codes/charset....8859-1&frame=2
There are only 256 characters in ISO-8859-1, so where did these come
from?

The site is a bit confusing. Only "frame=..." is important for the
displayed characters. One and the same document is then sent with
different charset parameters. That should have no effect - but
actually browsers will take a different typeface for each charset
parameter.

It really is deeply misleading.

http://www.physci.org/codes/charset....8859-5&frame=1
Doesn't actually display any cyrillic characters.

<http://www.physci.org/codes/charset.jsp?cs= iso-8859-5&frame=5>

That is displaying unicode characters 0401-0500 (rather than the more
useful 0400-04FF).

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <st***@pugh.net > <http://steve.pugh.net/>

[Andrew Thompson]

Steve Pugh wrote:

Andreas Prilop <nh******@rrz n-user.uni-hannover.de> wrote:

On Wed, 18 Feb 2004, Steve Pugh wrote:

http://www.physci.org/codes/charset....8859-1&frame=1
Characters 127 to 159 in ISO-8859-1 (and all other ISO-8859
encodings) are control characters. You seem to have some
Windows-1252 characters in there instead.
Blame your own browser!

Blame all my own browsers! Every browser I have incorrectly displays,
for example, ™ as a trademark sign. That's NN4, NN6, NN7, IE5,
IE5.5, IE6, Op5, Op6, Op7, Moz 1.6, Firefox 0.8 and even Lynx.

But the site is claiming that -
"SGML character 153. This is the character "?".

Not here. My UA's (IE6/Moz 1.3 on XP) shows
http://localhost:8080/codes/char.jsp?char=153
as a 'tm' character.
In HTML you would write it:
<p>This is the character "™".</p>"
<http://www.physci.org/codes/char.jsp?char=1 53>
which is just plain wrong.

http://www.physci.org/codes/charset....8859-1&frame=2
There are only 256 characters in ISO-8859-1, so where did these come
from?

I got the impression that charset affected
the characters, that was not borne out by
my investigations, but I thought I would
leave it in there for the moment.

The site is a bit confusing. Only "frame=..." is important for the
displayed characters. One and the same document is then sent with
different charset parameters. That should have no effect - but
actually browsers will take a different typeface for each charset
parameter.

So I should remove all reference to charset?
It really is deeply misleading.
http://www.physci.org/codes/charset....8859-5&frame=1
Doesn't actually display any cyrillic characters.

<http://www.physci.org/codes/charset.jsp?cs= iso-8859-5&frame=5>

That is displaying unicode characters 0401-0500 (rather than the more
useful 0400-04FF).

....errr. That table is 16x16, ..or were you talking
hex there? I want to add the hex denomination
when I get a moment.

But on another note, I had a few questions
when I first posted, but the comments have
made me realise ..I have a lot of questions.

I'll cogitate the comments for a while before
I formulate my questions..

--
Andrew Thompson
* http://www.PhySci.org/ Open-source software suite
* http://www.PhySci.org/codes/ Web & IT Help
* http://www.1point1C.org/ Science & Technology

[Karl Smith]

"Andrew Thompson" <Se********@www .invalid> wrote:

http://www.physci.org/codes/charset.jsp

This page is my 'charset explorer', it displays
character codes in a table 256 at a time.

Yeah, anyway. How do I save the applet with the periodic table? That's
my reason for loathing Flash and Java: when they are something
worthwhile it's always such a hassle to save.

Pack it all into 100+ nested tables, I say. Then use:

td table {
display: none
}
td:hover table {
display: table;
position: absolute;
z-index: 2;
}

- or something. Would be cool for one of those "this only works in
Opera" demo pages.

[Steve Pugh]

"Andrew Thompson" <Se********@www .invalid> wrote:

Steve Pugh wrote:

Andreas Prilop <nh******@rrz n-user.uni-hannover.de> wrote:

On Wed, 18 Feb 2004, Steve Pugh wrote:

http://www.physci.org/codes/charset....8859-1&frame=1
Characters 127 to 159 in ISO-8859-1 (and all other ISO-8859
encodings) are control characters. You seem to have some
Windows-1252 characters in there instead.

Blame your own browser!

Blame all my own browsers! Every browser I have incorrectly displays,
for example, ™ as a trademark sign. That's NN4, NN6, NN7, IE5,
IE5.5, IE6, Op5, Op6, Op7, Moz 1.6, Firefox 0.8 and even Lynx.

But the site is claiming that -
"SGML character 153. This is the character "?".

Not here. My UA's (IE6/Moz 1.3 on XP) shows
http://localhost:8080/codes/char.jsp?char=153
as a 'tm' character.

It's just that TM characters aren't in acsii and hence couldn't be
transmitted in the news message. I should have edited the cut and
paste before sending.

That's what my browser shows as well, which is what's wrong.
™ is undefined in HTML and should not be used. It's a widespread
browser bug/feature to translate this into a Windows-1252 character.
The trademark character is actually ™

http://www.physci.org/codes/charset....8859-1&frame=2
There are only 256 characters in ISO-8859-1, so where did these come
from?
I got the impression that charset affected the characters,

It does. But you didn't use any characters in your pages. You used
numeric character references, which in HTML are always references to
unicode.

The site is a bit confusing. Only "frame=..." is important for the
displayed characters. One and the same document is then sent with
different charset parameters. That should have no effect - but
actually browsers will take a different typeface for each charset
parameter.
So I should remove all reference to charset?

Probably best. Users already have the ability to change the character
set used to display the page via their browsers. As your tables don't
use anything that's character set dependent there's really no point in
including them.

<http://www.physci.org/codes/charset.jsp?cs= iso-8859-5&frame=5>

That is displaying unicode characters 0401-0500 (rather than the more
useful 0400-04FF).

...errr. That table is 16x16, ..or were you talking
hex there?

Yes, wasn't the FF a clue? ;-)

The point I was making is that character sets are zero based, so by
counting 1 to 256 rather than 0 to 255, and so on for your higher
frames, you're doing it differently to every other reference.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <st***@pugh.net > <http://steve.pugh.net/>

[Andrew Thompson]

Karl Smith wrote:

Yeah, anyway. How do I save the applet with the periodic table?
??? Bit of a change of subject!

Anyway, I had never bothered to set-up
the periodic table _applet_ as an easy install,
but you can get the application here..
http://www.physci.org/install/download.jsp

It's around 1 Meg, and includes 5
other programs (the page lies and
says 6 - but the browser was so
dodgy I removed it)

Over the next year I plan to break the
software suite up into individual
programs (on the basis that not many
people need a software suite with both
a text editor and ..periodic table)
...That's
my reason for loathing Flash and Java: when they are something
worthwhile it's always such a hassle to save.
Java now offers Java Web-Start.
It offers a painless install for the user
(except for the 'you might die' security
warning that comes up when installing).

On the upside, program updates are
automatic, on the downside, even
Java developers have trouble finding
where the .jar files are actually installed.
Pack it all into 100+ nested tables, I say. Then use:

td table {
display: none
}
td:hover table {
display: table;
position: absolute;
z-index: 2;
}

- or something. Would be cool for one of those "this only works in
Opera" demo pages.

;-)

--
Andrew Thompson
* http://www.PhySci.org/ Open-source software suite
* http://www.PhySci.org/codes/ Web & IT Help
* http://www.1point1C.org/ Science & Technology

[Andrew Thompson]

Steve Pugh wrote:

"Andrew Thompson" wrote:

Steve Pugh wrote:

Andreas Prilop <nh******@rrz n-user.uni-hannover.de> wrote: .... That is displaying unicode characters 0401-0500 (rather than the
more useful 0400-04FF).
...errr. That table is 16x16, ..or were you talking
hex there?

Yes, wasn't the FF a clue? ;-)

Well shucks, I need *big* clues.
...about '4x2' should do.
The point I was making is that character sets are zero based, so by
counting 1 to 256 rather than 0 to 255, and so on for your higher
frames, you're doing it differently to every other reference.

I am starting a trend.
...OK, no, not really.
I'll adjust it next couple of days.

[Karl Smith]

"Andrew Thompson" <Se********@www .invalid> wrote:

Karl Smith wrote:

Yeah, anyway. How do I save the applet with the periodic table?

??? Bit of a change of subject!

Anyway, I had never bothered to set-up
the periodic table _applet_ as an easy install,
but you can get the application here..
http://www.physci.org/install/download.jsp

Not today. Java all crappy today. Can't get the page with the applet
to display, either.

So I've spent the last hour or so reviewing some available HTML
periodic tables and what a crappy bunch they are! The most appealing
(to look at) I've found so far is this:

http://www.dayah.com/periodic/

but it's a mess of JavaScript, font tags and rubbish underneath. And
I'm not certain the data is layed out correctly.

Shame 'bout the browser sniffing JavaScript that adds this admonition:
"Because of the complexity of this page, certain browsers may not
display it correctly. Your browser, Opera version 7.23, is
insufficient for viewing this page."
In IE6 it looks the same except for:
"Because of the complexity of this page, certain browsers may not
display it correctly. Your browser, Microsoft Internet Explorer
version 4.0, is sufficient for viewing this page."

Bugger that! I' gonna do my own pure CSS periodic table that ony works
in Opera.
Could you tell me what all those numbers mean and what the proper
layout for each element's data should be?