Yoan Blanc Reporter
	Description • 20 years ago

The Midas implementation isn't making goog html/xhtml (well-formed but not valid).

DTD said :

<!ELEMENT ul (li)+>
<!ELEMENT ol (li)+>

Step to reproduce :

1) pick a rich text editor (http://www.kevinroth.com/rte/demo.htm)
2) build some nested lists like :

* A
   * B
* C

3) look at the source code (here the result)

<ul>
 <li>A</li>
 <ul>
  <li>B</li>
 </ul>
 <li>C</li>
</ul>

	Yoan Blanc Reporter
	Comment 1 • 20 years ago

Looking at the spec the resould should be (in xhtml) :

<ul>
 <li>A
  <ul>
   <li>B</li>
  </ul>
 </li>
 <li>C</li>
</ul>

or (html valid) :

<ul>
 <li>A
 <ul>
  <li>B
 </ul>
 <li>C
</ul>

a solution (with regex or other) :

1) Deleting every </li>
2) Replace :
 * </ul>        ->  </li></ul>
 * </ul><li>    ->  </ul></li></ul>
 * (?!<ul>)<li> ->  </li><li> (a <li> not preceded by a <ul>)

	Yoan Blanc Reporter
	Comment 2 • 20 years ago

(In reply to comment #1)
>  * </ul><li>    ->  </ul></li></ul>

</ul><li>   ->  </ul></li><li>

Sorry

	Kathleen :Brade
	Comment 3 • 20 years ago

Do either of you happen to know if this is a regression?

The output generated by Composer doesn't pass the validator @w3.org but I swear
it did long ago.  Any ideas or am I thinking of something else?

I haven't tested this on a recent trunk build (I see it in Mozilla 1.7.3 on Mac).

	Daniel Glazman (:glazou) (NOT MONITORING BMO ANY MORE)
	Comment 4 • 20 years ago

This is not a regression. Duping to bug 54479

*** This bug has been marked as a duplicate of 54479 ***

	Bieler Batiste
	Comment 5 • 20 years ago

this PHP code correct the misteaks 
 
$html = preg_replace("#</li>#","",$html); 
while( preg_match("#</ul>[^>]*<ul>[^>]*<ul>#",$html) ) 
{ 
 $html = preg_replace("#</ul>[^>]*<ul>[^>]*<ul>#","<ul>",$html); 
} 
$html = preg_replace("#</ul>#","</li></ul>",$html);