wc validator says: non SGML character number ...

[Jarek]

Hi all!

I'd like to validate my web application and I see that validator.w3.org
says that some of valid characters from ISO-8859-2
are non SGML. Is this a bug in validator, or some charactes are in some
way forbidden ?
What is strange for me, that I'm using a lot of national characters with
codes above A0h, and only few of them are problematic.

Jarek

 

[Jukka K. Korpela]

I'd like to validate my web application and I see that
validator.w3.org
says that some of valid characters from ISO-8859-2
are non SGML.

Maybe the reason is that they are. For example, NUL is a non-SGML
character.

Is this a bug in validator,
Hardly.

or some charactes are in some way forbidden ?
Yes.

What is strange for me, that I'm using a lot of national characters
with codes above A0h, and only few of them are problematic.

They are probably not ISO-8859-2 characters at all; you have probably
confused ISO-8859-2 with a proprietary encoding by Microsoft.

To get more detailed help, you should post a more detailed question.
That spells "post the URL, Luke!"

 

[Jarek]

They are probably not ISO-8859-2 characters at all; you have probably
confused ISO-8859-2 with a proprietary encoding by Microsoft.

I'm not using widoze at all, but it looks that I've found where is the
problem: application sends valid ISO-8859-2 characters, but browser
transode them depending on locally available font. Ie. if web server
sends character C3h, View Source in browser it may show 84h if the font
is Arial, C3h if font is Helvetica and probably something else if font
will be UTF-8. Really surprising.

To get more detailed help, you should post a more detailed question.
That spells "post the URL, Luke!"

I'm developing internal applications, available only internally, so URL
will not help. I took html from View Source and put it to validator as text.

Best regards
Jarek.