XML

Anon

Im writing an online HTML editor.

I'm stuck with a situation where I wan't to ensure that the user has closed any HTML tables with the appropriate ending tags.

I thought the best way to to this would be to use PHP's xml parser and write a function to return the number of start and end table tags.

Problem is the xml_parser function is very strict. it throws up an error when it encounters anything that doen't resemble xml.

Is there any way to bypass this strict error checking?

Anon

Paek, I've thought of the same thing but have not started work on one. Can you give an example of where the XML parser doesn't like valid HTML? (Don't forget to turn off the case sensitivity.)

There definitely are some non-container HTML tags that you would want to ignore, such as <BR>.

Anon

I could have a string like this which the user has entered into a form.

$string="
<table bgcolor='black'>
<tr><td>this is a table</td></tr>
</table> This is a statement outside of the html tags ";

PHP will then throw up this error message
XML error: junk after document element at line 4

I want to be able to force PHP to just ignore the junk elements.

Do you know if this is possible?

yelvington

I don't know exactly what rules expat enforces, but would it help to do this:

$string = "<p>$string</p>";

before handing it to the parser?

Anon

That worked well.I can temporarily add <p></p> tags to the string.

Now all I've got to do is to write the script so that it ignores single html tags such as <hr> <br> ect. and so that it ensures that attributes (eg. bgcolor='black') are enclosed in single or double quotation marks .

thanks

Anon

i too am making an online/offline html generator. i am having problem with special characters such as '&' and <br>. normally, with a language such as jsp i can use entities, such as & or >, but php seems to explode when these are parsed through.

any ideas?

Anon

here is the answer to my own problem:

see this link.
http://www.phpbuilder.com/annotate/message.php3?id=1006651

basically, the & entity is treated as a new set of character data, but because of the 'end_element' handler switch, this causes the character data to overwrite itself.

the answer is to concate the data each time.

character_data($parsere,$data){

$newData .= $data

}