CODEWITHSUNDEEP
htmlxmlparsingtagsdoctype
Albus Dumbledore
Albus Dumbledore

Reputation: 12616

(x)HTML: Parsing bizarre tags

I am building my own humble (x)html parser. All is ok, but some doctype tags break it. Let me show you:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd" [
<!ENTITY D "&#x2014;">
<!ENTITY o "&#x2018;">
<!ENTITY c "&#x2019;">
<!ENTITY O "&#x201C;">
<!ENTITY C "&#x201D;">
]>

As far as I know, no other kind of tag is allowed to be nested in this way (I mean inside the tag name), i.e. incl. xml instructions and commenting tags.

My question is, what can you say about this issue. It looks so much against common sense to me. And of course, it's perfectly valid as far as XML is concerned.

Thanks!

Upvotes: 1

Views: 86

Answers (2)

mattbasta
mattbasta

Reputation: 13709

Wow, that's something you don't see every day. Code like that traces its roots all the way back to SGML. SGML had all sorts of fun bits that we use today: <![CDATA[, <?xml version="1.0"?>, and even PHP's open/close "tags": <?php ... ?>.

All in all, it's nothing to worry about. It's just a series of instructions for the parser, though you might be better off accomplishing the same thing via another means. Keep calm and carry on.

Upvotes: 1

Quentin
Quentin

Reputation: 943193

A Doctype Declaration isn't a tag, so it doesn't follow the syntax rules for tags. It is well specified though. (The trick to building a parser is to start with the specification and not from example documents)

Upvotes: 3

Related Questions