Validation Issues

"downview" <downview (AT) gmail (DOT) com> wrote

Getting a little off topic here, but I'll give a couple answers. I'm
not supplying the URI because this site is a password-protected page
that I wrote and maintain for my wife. I'm an infantry soldier
currently on deployment in Iraq and it contains a lot of personal
stuff.

That's understandable, but perhaps you could make a copy of the page that
has the personal stuff removed. Check to see if that page still causes a
validation error and, if so, make that available. Otherwise, you'll have
narrowed down the location of the error, and you can continue removing and
restoring parts of the content until you find the source of the problem.

downview wrote:

I was, actually. Sorry about that. Minus the asterics and bullets,
it's the same though.

Nope

Source:

a: {
font-family: Trebuchet MS,sans-serif;

Pretty:

****{
**********o*font-family*:*Trebuchet*MS,*sans-serif;


You have a rogue ":" after the a, which is triggering the parse error. (The
validator is very bad about getting line numbers right).

--
David Dorward <http://blog.dorward.me.uk/> <http://dorward.me.uk/>
Home is where the ~/.bashrc is

"Harlan Messinger" <h.messinger (AT) comcast (DOT) net> wrote

"downview" <downview (AT) gmail (DOT) com> wrote in message
news:224dff78.0408240610.49d9fbd6 (AT) posting (DOT) google.com...
Getting a little off topic here, but I'll give a couple answers. I'm
not supplying the URI because this site is a password-protected page
that I wrote and maintain for my wife. I'm an infantry soldier
currently on deployment in Iraq and it contains a lot of personal
stuff.

That's understandable, but perhaps you could make a copy of the page that
has the personal stuff removed. Check to see if that page still causes a
validation error and, if so, make that available. Otherwise, you'll have
narrowed down the location of the error, and you can continue removing and
restoring parts of the content until you find the source of the problem.

Thank you very much, Harlan. I'm not sure if it was _exactly_ what you
ment, but you trigged a little common sense thinking and I'm happy to
report that my CSS markup now validates. If anyone else comes across
someone that has an extremely ambigious validation error, tell them do
what Harlan here suggested and what I did. Make a duplicate css page,
delete the first entry, save, upload, check for validation. Continue
this process one element after another, until it finally validates.
Then replace everything but that last element before it validated,
check it again; if it validates, you've found your problem. It took me
less than ten minutes to find the offending style and correct the
mistake using this method. To Harlan and everyone else that was kind
enough to post, a sincere thank you. It's refreshing to have found an
intelligent, (mostly) courteous, helpful newsgroup.

PS. The error was "a:" above the "a:link" Either this command is
superflous, or not supported (I had no idea) and by removing it, my
css validated and, more important, the site didn't change a pixel.
Good things.

On 24 Aug 2004 07:10:13 -0700, downview (AT) gmail (DOT) com (downview) wrote:

Jan Roland Eriksson <jrexon (AT) newsguy (DOT) com> wrote

On 23 Aug 2004 22:18:15 -0700, downview (AT) gmail (DOT) com (downview) wrote:

Sorry to be so ambiguous, but that's exactly the message.

And I quote:
To work as intended, your CSS style sheet needs a correct document
parse tree. This means you should use valid HTML.

ERRORS
* Line: 0
Parse Error - :
______________________________________

...I'm thinking perhaps there is a bug
A far fetched idea in your situation...

but here's the CSS markup to ponder as you requested:

* body {
o background-image : url(images/side_background.jpg);
[...]

Maybe this is just a case of a usenaut that has not read Q&A.6 in the
newly revised newsgroup FAQ?

Getting a little off topic here, but I'll give a couple answers. I'm
not supplying the URI because this site is a password-protected page
that I wrote and maintain for my wife....

I can respect that, but surely you do have your CSS file as an external
resource linked to your main document?

You can give us the URL for your CSS resource without compromising the
integrity of your web page(s).

...Also, this validation is for the CSS markup alone and that, in its
complete, natural form has been supplied.

Nope, you have not supplied us here with exactly the same as you did
send to the CSS "validator", if your CSS is external to your web pages.

I really should not guess, but things like a faulty server stated
character encoding, bogus "invisible" characters at the start of your
CSS and the likes may be liable to trip a checkup system. At least that
seems to be plausible given the error report for "line 0".

--
Rex

"downview" <downview (AT) gmail (DOT) com> wrote

PS. The error was "a:" above the "a:link" Either this command is
superflous, or not supported (I had no idea) and by removing it, my
css validated and, more important, the site didn't change a pixel.
Good things.

What's not supported is the colon followed by nothing. For just a plain "a",
use "a". Use the colon only when you have "a:link" or something like that.

On 24 Aug 2004 12:30:22 -0700, downview wrote:

Make a duplicate css page,
delete the first entry, save, upload, check for validation.

Just an (ever so slight) tweak on that suggestion..

Let us assume the number of CSS definitions is
*ridiculously* large, for my example I choose
1,000,000.

If you were to divide the sheet in half, check
each half alternately and pursue the half with
errors, you can actually narrow the error down
to the single CSS definition that is problematic
in just 20 checks*.

* It is a crude adaptation of 'binary search' logic.

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

On Tue, 24 Aug 2004 20:13:12 GMT, Andrew Thompson <SeeMySites (AT) www (DOT) invalid>
wrote:

Let us assume the number of CSS definitions is
*ridiculously* large, for my example I choose
1,000,000.

If you were to divide the sheet in half, check
each half alternately and pursue the half with
errors, you can actually narrow the error down
to the single CSS definition that is problematic
in just 20 checks*.

providing, of course, the problem is not one style rule but a combination,
some in one half, some in the other.

On Wed, 25 Aug 2004 13:13:44 -0400, Brian
<usenet3 (AT) julietremblay (DOT) com.invalid> wrote:

downview wrote:
I hereby call this thread finished.

Where did you get the idea that you can declare a thread finished?

Perhaps he invented the internet?

Maybe we could discuss whether the CSS validator is or isn't a validator
or a linter, just to prove him wrong...

On Tue, 24 Aug 2004 23:04:38 -0400, Neal wrote:

On Tue, 24 Aug 2004 20:13:12 GMT, Andrew Thompson wrote:

(1,000,000 CSS definitions)

.. you can actually narrow the error down
to the single CSS definition that is problematic
in just 20 checks*.

providing, of course, the problem is not one style rule but a combination,
some in one half, some in the other.

I can agree that perfectly valid CSS definitions
can combine in unpredictable and seemingly chaotic
ways, but I was referring specifically to invalid CSS
and/or CSS parse errors.

Even if there are 5 separate (invalid CSS) errors,
this 'halving' technique would allow you to identify
each of them quickly..

For the problems caused by *combinations* of
*valid* CSS rules, the above technique is most
likely to propound confusion, rather than
clarify anything.

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

Quoth the raven downview:

Late to the party, I'll submit that all font names with spaces must be
surrounded by quotes.

h2
{font-family: Arial Bold, sans-serif;

font-family: "Arial Bold", sans-serif;

a: {
font-family: Trebuchet MS,sans-serif;

Several occurrences of "Trebuchet MS".

#main a {
font-family: Times New Roman,sans-serif;

Ditto TNR. You get the idea now.

You should also be able to use <div>s to style areas and only use the
font properties for the <div> instead of repeated listing.

Numerous other little inconsistencies in there...

--
-bts
-This space intentionally left blank.