comparing results of validators

**Steve Pugh** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

On Tue, 14 Dec 2004 10:29:16 -0600, "Richard" <Anonymous@127. 001>
wrote:
[color=blue]
>http://geocities.com/r_bullis/gallery2.html
>
>Validator 3 at w3c totally annhialates the page.
>It doesn't like comment tags, nor does it like "alt" not being used with img
>tags.
>While the jigsaw version now says the page is compliant.[/color]

They're validating different things.
validator.w3.or g is an HTML validator.
jigsaw.w3.org/css-validator/ is a CSS 'validator'.
[color=blue]
>Validator 3 doesn't like the added stuff yahoo puts on for their embedded
>ads, while Jigsaw totally ignores them.[/color]

Because they're HTML errors not CSS errors.
[color=blue]
>So what is the proper use of comment tags within a division?
> is not approved of.[/color]

A comment starts and ends with two dashes. So you have an extra dash
after the end of the comment but before the close of the declaration.
That's an error.
[color=blue]
>Excuse me. It's a comment. It should be ignored.[/color]

It is ignoring the comment (everything between -- and --). It's the
stuff after the comment that's an error.
[color=blue]
>It ignored the // script comments.[/color]

Of course, the content of <script> blocks is CDATA and hence subject
to different parsing rules.

Steve

**Arne** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

Once upon a time *Richard* wrote:
[color=blue]
> http://geocities.com/r_bullis/gallery2.html
>
> Validator 3 at w3c totally annhialates the page.[/color]

What is "Validator 3"? Do you mean validator.w3.or g?
[color=blue]
> It doesn't like comment tags, nor does it like "alt" not being used with img
> tags.
> While the jigsaw version now says the page is compliant.[/color]

The "jigsaw version" ( jigsaw.w3.org ) is a CSS validator only. It
does'nt validate anything else. The "validator.w3.o rg" validates HTML
[color=blue]
> Validator 3 doesn't like the added stuff yahoo puts on for their embedded
> ads, while Jigsaw totally ignores them.[/color]

The validator don't like it because it's on top of the file, where the
doctype spec. should be. Jigsaw don't see any CSS in the "Yahoo stuff",
since it's javascript only(?)
[color=blue]
> Validator 3 apparently does not like the use of inline <style> while jigsaw
> does.[/color]

Oh yes, it does. It's something else that cause the error.
[color=blue]
> So what is the proper use of comment tags within a division?
>  is not approved of.
> Excuse me. It's a comment. It should be ignored.
> It ignored the // script comments.[/color]

The validator want just two hyphens in the comments, like

Don't ask me why, it's just like that it is :-)

--
/Arne

http://web.telia.com/~u97802964/

The old believe everything, the middle-aged suspect everything and the
young know everything. (Oscar Wilde)

**Darin McGrew** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

*Richard* wrote:[color=blue][color=green]
>>  is not approved of.
>> Excuse me. It's a comment. It should be ignored.[/color][/color]

Arne <user@domain.in valid> wrote:[color=blue]
> The validator want just two hyphens in the comments, like
> 
> Don't ask me why, it's just like that it is :-)[/color]

As the FAQ <http://www.htmlhelp.co m/faq/html/basics.html#htm l-comments>
explains:

A comment declaration starts with "<!", followed by zero or more
comments, followed by ">". A comment starts and ends with "--", and
does not contain any occurrence of "--" between the beginning and
ending pairs.

So, the first example should be interpretted as:

<! Start the declaration
-- Start a comment
-- End a comment
image 101 Text not within a comment (not allowed)
-- Start a comment
-> Text within a comment

And neither the second comment nor the declaration itself are closed.

But browsers don't really support SGML comments. The practical rule for
HTML comments (from the FAQ) is:

An HTML comment begins with "", and does not
contain "--" or ">" anywhere in the comment. Do not put comments
inside tags (i.e., between "<" and ">") in HTML markup.
--
Darin McGrew, mcgrew@stanford alumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp. com, http://www.HTMLHelp.com/

"FAILURE IS NOT AN OPTION. It comes bundled with the software."

**Spartanicus** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

Darin McGrew <mcgrew@stanfor dalumni.org> wrote:
[color=blue]
>*Richard* wrote:[/color]

Could you change your quote line to something more standard? Removing
the asterikses or including the email address allows for more effective
filtering of follow ups.

--
Spartanicus

**Richard** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

The main reason I use  is because the editor I use won't
accept anything less.
<! comment > is not acceptable.
<!- comment > is not acceptable.

I have used the same style before with validated pages.
So something else may have triggered the errors.

**Michael Winter** · Jul 21 '05, 12:36 AM

Re: comparing results of validators

On Tue, 14 Dec 2004 13:27:44 -0600, Richard <Anonymous@127. 001> wrote:
[color=blue]
> The main reason I use  is because the editor I use
> won't accept anything less.[/color]

If your editor won't accept[1]



then it's broken and time to find a new one.

[snip]

Mike

[1] I find it difficult to believe that an editor would somehow prevent
you from entering valid SGML comment delimiters.

--
Michael Winter
Replace ".invalid" with ".uk" to reply by e-mail.

**Steve Pugh** · Jul 21 '05, 12:37 AM

Re: comparing results of validators

On Tue, 14 Dec 2004 13:27:44 -0600, "Richard" <Anonymous@127. 001>
wrote:
[color=blue]
>The main reason I use  is because the editor I use won't
>accept anything less.[/color]

IF your editor doesn't except  then your editor is
very, very broken and you should drop it.
[color=blue]
><! comment > is not acceptable.
><!- comment > is not acceptable.[/color]

Those aren't comments.
[color=blue]
>I have used the same style before with validated pages.
>So something else may have triggered the errors.[/color]

You probably used  before and simply didn't realise
that this time you'd used --- instead of --.

Steve

**Harlan Messinger** · Jul 21 '05, 12:37 AM

Re: comparing results of validators

"Richard" <Anonymous@127. 001> wrote in message
news:cpnern0lin @news3.newsguy. com...[color=blue]
> The main reason I use  is because the editor I use won't
> accept anything less.
> <! comment > is not acceptable.
> <!- comment > is not acceptable.[/color]

What you've just said is that you're using one wrong format because two
other wrong formats don't work. What have you got against using the right
format, the one with *two* hyphens on each end?
[color=blue]
>
> I have used the same style before with validated pages. So something else[/color]
may have triggered the errors.

I don't think so. When I run the following code through the validator:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">
<title>Comments </title>
<! comment >
<!- comment ->


<p>HELLO!</p>

I get the following error messages:

1. Line 3, column 0: character data is not allowed here

<! comment >

You have used character data somewhere it is not permitted to appear.
Mistakes that can cause this error include putting text directly in the body
of the document without wrapping it in a container element (such as a
<p>aragraph</p>) or forgetting to quote an attribute value (where characters
such as "%" and "/" are common, but cannot appear without surrounding
quotes).

2. Line 6, column 16: invalid comment declaration: found name character
outside comment but inside comment declaration



3. Line 6, column 0: comment declaration started here



In error message two, column 16 is the third hyphen in the second trio of
hyphens, and that particular hyphen is colored red. So that is indeed being
flagged as an error.

What I don't understand is why the version with *one* hyphen on each side is
accepted.

**Jan Roland Eriksson** · Jul 21 '05, 12:37 AM

Re: comparing results of validators

On Tue, 14 Dec 2004 10:29:16 -0600, "Richard" <Anonymous@127. 001> wrote:
[color=blue]
>http://geocities.com/r_bullis/gallery2.html[/color]

[...]
[color=blue]
>So what is the proper use of comment tags within a division?
> is not approved of.
>Excuse me. It's a comment.[/color]

Nope, it's not.

It's a Comment Start, directly followed by a Comment End, followed by
some misplaced text, followed by yet another Comment Start that has the
characters '->' in it and then never ends as a comment until the parser
finds yet another sequence of two consecutive 'dashes'.

At which place your page is already hopelessly lost for the validator
already.

You have totally misunderstood the correct syntax of an SGML(HTML)
comment.

It is built from the following bits and pieces.

<! = MDO = Markup Declaration Open
-- = COM = Comment Start or End[color=blue]
> = MDC = Markup Declaration Close[/color]

so the following is a valid comment in SGML(HTML)



Here is another one that is correct...

<!-- this is another comment --
-- with two comment lines in it --[color=blue]
>[/color]

....now move on to count the balance of 'double dashes' in your page
source.

Summary:

Two consecutive 'dashes' starts a comment, the next occurrence of two
consecutive 'dashes' ends that same comment.

The very first two 'dashes' _MUST_ follow directly after an initial MDO.

After the last two closing 'dashes' there can be arbitrary white space
(but absolutely no NAME characters) before the closing MDC.

You have some length of road to travel yet, but its nice to see that you
have an interest to learn :-)

Fix your obvious comment syntax errors first and then come back so we
can help you out with the rest of your problems.

--
Rex

**Jan Roland Eriksson** · Jul 21 '05, 12:37 AM

Re: comparing results of validators

On Tue, 14 Dec 2004 16:17:20 -0500, "Harlan Messinger"
<h.messinger@co mcast.net> wrote:

[...]
[color=blue]
>What I don't understand is why the version with *one* hyphen on each side is
>accepted.[/color]

It's not accepted, just an effect of the parser needing to find a place
in the data stream where it can "recover" from total "mumbo-jumbo".

--
Rex

**Nick Kew** · Jul 21 '05, 12:37 AM

Re: comparing results of validators

In article <3293j9F3jnju5U 1@individual.ne t>,
"Harlan Messinger" <h.messinger@co mcast.net> writes:
[color=blue]
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
> "http://www.w3.org/TR/html4/strict.dtd">
> <title>Comments </title>
> <! comment >
> <!- comment ->
> 
> 
> <p>HELLO!</p>
>
> I get the following error messages:[/color]

[chopped]
[color=blue]
> 1. Line 3, column 0: character data is not allowed here
> 2. Line 6, column 16: invalid comment declaration: found name character
> outside comment but inside comment declaration
> 3. Line 6, column 0: comment declaration started here
>
> What I don't understand is why the version with *one* hyphen on each side is
> accepted.[/color]

It isn't. Lines 3 and 4 are character data, which has already been
reported as an error in the first message. Put it inside a <div>
(so character data *is* allowed) and get another surprise.

"Errors" 2 and 3 are a single error - which you have correctly
identified - and reporting it as two errors is a longstanding bug.
Validate it at Page Valet for a saner report.

--
Nick Kew