Re: HTML parse trees...

On Wed, 14 Jun 2000, Rasmus Lerdorf wrote: > > function _($tag, $attributes='', $content='') { > > $html = '<'.$tag; > > if (is_array($attributes)) { > > while ($attribute = each($attributes)) { > > if (is_int($attribute[1])) { > > $html .= ' '.$attribute[0].'='.$attribute[1]; > > } elseif (is_string($attribute[1])) { > > $html .= ' > > '.$attribute[0].'="'.$attribute[1].'"'; > > } elseif ($attribute[1]) { > > $html .= ' '.$attribute[0]; > > } > > } > > } else { > > $content = $attributes.$content; > > } > > if ($content) { > > return $html.'>'.$content.'</'.$tag.'>'; > > } else { > > return $html.'>'; > > } > > } > > Now I can't imagine that I'd be the first to do this, and I'm also seeing > > that this whole thing is going to end up being quite inefficient. All > > these string concatenations... > > > > Anyway, how come PHP doesn't have built in support for HTML (or XML) parse > > trees and functions to build them, store them and dump them efficiently? > PHP has 2 XML parsers. HTML is difficult to parse because it tends to not > conform to a clean dtd. I'm actually more interested in the UNparsing :) of HTML. I'd like to have primitives as above that will build the parse tree and then dump it as HTML. -- cg