xhtml teething troubles

[Ted]

This page
http://homepage.ntlworld.com/r.a.mcc..._local_dtd.xml
doesn't work properly in Firefox or IE6. The faults are different. In
Firefox the TestText entity is not recognised. In IE6, the <br /> tag
doesn't cause a line break. Can anyone tell me what I'm doing wrong?

[Michael Winter]

On 31/05/2006 12:00, VK wrote:

Michael Winter wrote:
[snip]

Firefox can process external subsets, but only does so in certain
circumstances and they do not include DTDs found on the Web.

Yeh, right... Firefox can process external subsets but it cannot
process external subsets. :-)

I don't expect you to have taken the time to understood what was
written, but it would have been nice if you had.

[snip]

How many times do I have to repeat this: it is /not/ a bug!

You may repeat it as many times as you want:

Yes, I know. You're a stubborn fool that refuses to learn anything from
anyone.

As I've stated before, I don't reply to your posts any more to attempt
to further your knowledge. It's wasted effort. However, I have no
intention of letting you mislead others who ask this group for help or
informed opinion.
it doesn't change the production mechanics.
None of the grammar productions, nor the prose that accompanies them,
have any relevance here. Section 5.1 Validating and Non-Validating
Processors[1] states exactly what's expected from conforming processors.
Specifically,

Non-validating processors are REQUIRED to check only the
document entity, including the entire internal DTD subset,
for well-formedness.

noting that Firefox (and others) use non-validating processors. Furthermore,

While they are not required to check the document for validity,
they are REQUIRED to process all the declarations they read in
the internal DTD subset and in any parameter entity that they
read ...

Note the conspicuous absence of any reference to an external entity.
<q>The productions later in this specification for individual
nonterminals (elementdecl, AttlistDecl, and so on) describe the
declarations after all the parameter entities have been included.</q>

Plain and simple: either you include all declared entities and start
production, or you don't bother with the production at all.

It would seem that you don't know the difference between an entity
reference and a parameter entity reference. The latter has nothing to do
with the problem described by the OP.

[snipped gibberish]

Mike
[1] 5.1 Validating and Non-Validating Processors, XML 1.0
Specification. <http://www.w3.org/TR/REC-xml/#proc-types>

--
Michael Winter
Prefix subject with [News] before replying by e-mail.

[VK]

Michael Winter wrote:

However, I have no
intention of letting you mislead others who ask this group for help or
informed opinion.

That is exactly my intention too, unfortunately.

All differences between external DTD subsets and internal DTD subsets
are well spelled at <http://www.w3.org/TR/REC-xml/>, section 2.8

Section 2.9 also defines that in the OP case standalone in prolog
presumed "no" and /cannot/ be "yes": <q>entities (other than amp, lt,
gt, apos, quot), if references to those entities appear in the
document</q>

The idea that internal DTD is /a must to process/ while external DTD is
something /optional to process/ is plain stupid. It took a long brain
waching from one organisation to start to believe it.

A standard-compliant XML processor is not a capricious lady as one is
trying to present: "if I'm moody, I'll retrieve external subsets before
parsing; otherwize I'll ignore them and just break parsing on the first
non-declared entity".

The real reason of external DTD complications for someone is because of
bogus DTD's for (X)HTML. It is like a primitive tribe would find a TV
set and use it as a local god for years - with dancing around and
putting flowers on it. What would be a hard discover for them that it
is actually something practically useful and the actual purpose of this
thing is not to stay in the middle of the village and be covered with
flowers.

[Ted]

"VK" <sc**********@y ahoo.com> wrote in message
news:11******** *************@f 6g2000cwb.googl egroups.com...

Also it is not clear why did you call the post " xhtml teething
troubles" as the linked document has no relation neither to XHTML nor
to HTML. It's a well-formed (accounting extra entities in DTD) XML
document served as XML document.

Sorry if you think I'm posting to the wrong group. I did a |Google search to
find newsgroups which dealt with xhtml, and this seemed to be the best
group. xhtml is supposed to be served as xml
In both cases it is really out of the scope of (X)HTML authoring,
<comp.text.xm l> is more relevant.

Thaks for the tip.

[VK]

Ted wrote:

did a Google search to find newsgroups which dealt with xhtml
And you came to the right one. The problem is that your code is not
XHTML and has nothing to do with it - though I believe you that you
fairly thought otherwise.
xhtml is supposed to be served as xml

Noop. Never. XHTML is supposed to be served as application/xhtml+xml
XML/XSL is supposed to be served as text/xml.

XHTML code must be /well-formed in accordance with XML rules/, but it
is /not/ served as XML.

[Ted]

"VK" <sc**********@y ahoo.com> wrote in message
news:11******** **************@ c74g2000cwc.goo glegroups.com.. .

Ted wrote:
xhtml is supposed to be served as xml

Noop. Never. XHTML is supposed to be served as application/xhtml+xml
XML/XSL is supposed to be served as text/xml.

XHTML code must be /well-formed in accordance with XML rules/, but it
is /not/ served as XML.

According to : http://www.w3.org/MarkUp/2004/xhtml-faq#texthtml
"XHTML is an XML format; this means that strictly speaking it should be sent
with an XML-related media type (application/xhtml+xml, application/xml, or
text/xml). "

As long as I stick with the free webspace provided by NTL, any file with the
extension .xml will be served as application/xml

[Ted]

<di*****@codesm iths.com> wrote in message
news:11******** **************@ j55g2000cwa.goo glegroups.com.. .

Ted wrote:

http://homepage.ntlworld.com/r.a.mcc..._local_dtd.xml doesn't work properly in Firefox or IE6.

Pragmatically, I think you're stuffed. It appears to be implemented
correctly as a piece of XML work, but this just isn't how the web
works, barely how XML is used (in practice) and is _certainly_ not a
technique that's usable on the web for the forseeable future.

There's a raft of valid techniques out there that just aren't commonly
used in practice, so support for them varies from poor to none.

Great! Can anyone give me some links to web pages which teach the techniques
which are supported?

[VK]

Ted wrote:

According to : http://www.w3.org/MarkUp/2004/xhtml-faq#texthtml
"XHTML is an XML format; this means that strictly speaking it should be sent
with an XML-related media type (application/xhtml+xml, application/xml, or
text/xml). "

Does it really say that?! A second... I be damned, indeed! Sometimes I
feel like to call marines to smash W3C down. Unfortunately the absence
of W3C would be even more harmful than its presence, so needs to be
tolerated :-( :-)

No, text/xml has no relation with XHTML. XML parser has no idea about
additional layout/styling rules for (X)HTML element. You already
discovered it on <br> sample. Each Content-Type for a particular
content, this is how the Web works.

There is a lot of scum connected with XHTML, so before to proceed you
may want to check that your hosting provider supports XHTML as such.
For this you may use this primitive probe: save it as say probe.html
and upload it.

Open it: it shows up fine because it is served as Content-Type
text/html, so browser uses HTML parser on it, and as HTML it is a valid
markup: you are allowed to skip closing tag for list elements.

Now let's try to find out if server admin has an extension associated
with Content-Type application/xhtml+xml (XHTML). In the US the most
common extensions for this are .xhtml and .xht
So try to change the extension first to .xhtml then to .xht, upload and
view in Firefox. The sign that the page is served and parsed as XHTML
will be error message "closing tag missing". From this point out you
have a real XHTML page you can play with. It also means that
unfortunately this page is not available to Internet Explorer users
anymore (IE doesn't support XHTML and it will not do it in any near
future).

// probe page

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type"
content="text/html; charset=iso-8859-1" />
<title>XHTML Probe</title>
</head>

<body>
<ul>
<li>Item 1
<li>Item 2
<li>Item 3
</ul>
</body>
</html>

[Andy Dingley]

Ted wrote:

According to : http://www.w3.org/MarkUp/2004/xhtml-faq#texthtml
"XHTML is an XML format; this means that strictly speaking it should be sent
with an XML-related media type (application/xhtml+xml, application/xml, or
text/xml). "

Be aware that VK is a clueless idiot (recent threads passim), although
some believe instead that he's part troll (just not as funny as Bill
Bailey). Typically about 2/3rd of what he posts is wrong.

This ng is probably your best place to discuss techniques for pushing
XHTML. The local attitude is against it, but that's from a position of
some serious knowledge on the matter. Do some archive searching -
you'll find useful advice on many topics and soon recognise who the
usefully knowledgeable posters are (Jukka, Alan and Henri spring to
mind for this topic).

[VK]

VK wrote:

<meta http-equiv="Content-Type"
content="text/html; charset=iso-8859-1" />

Oops... sorry. I used W3C Amaya for quick XHTML page generation and it
inserts this nonsense automatically - it has to be removed (though
harmful as server-side generated Content-Type has higher priority over
http-equiv).

[Michael Winter]

On 31/05/2006 15:01, VK wrote:

Michael Winter wrote:

However, I have no intention of letting you mislead others who ask
this group for help or informed opinion.
That is exactly my intention too, unfortunately.

You don't qualify as informed, particularly as you've already admitted
to not understanding the XML Specification properly.

At the same time I miss a lot of professional knowledge in XML
and XSLT, definitely weak in reading W3C docs and my English
may fell down - especially after midnight.
-- 11************* ********@j73g20 00...legro ups.com,
comp.infosystem s.www.authoring.html

[snip]
Section 2.9 also defines that in the OP case standalone in prolog
presumed "no" and /cannot/ be "yes": [...]
The standalone document declaration is irrelevant here.

[snip]
The idea that internal DTD is /a must to process/ while external DTD
is something /optional to process/ is plain stupid. [...]
Your opinion on the matter is irrelevant. The required behaviour is what
it is. Validating processors must read and process the complete DTD, and
any external parsed entities. Non-validating processors are only
required to process the internal subset (if present) up to the first
unread parameter entity (except when processing a standalone document).
A standard-compliant XML processor is not a capricious lady as one is
trying to present: [...]
I've done no such thing. The behaviour of the XML processor used by
Firefox is well-defined in this regard.
The real reason of external DTD complications for someone is because
of bogus DTD's for (X)HTML.

And what bogus DTDs would these be? The official ones, knowing you.

[snipped drivel]

--
Michael Winter
Prefix subject with [News] before replying by e-mail.