如何解决“simplexml_load_file() 解析器错误：未定义实体 'nbsp'”？

Question

I use PHP to generate XML files. I have use some code below to avoid error.

$str = str_ireplace(array('<','>','&','\'','"'),array('&lt;','&gt;','&amp;','&apos;','&quot;'),$str);

but still cause fault.

simplexml_load_file() [function.simplexml-load-file] *[file name]* parser error : Entity 'nbsp' not defined in *[file name] [line]*

The error text here:

Dallas  Dallas () is the third-largest city in Texas and the ninth-largest in the United States.

In IE8, it seems to fault in () . So how many symbols should I notice?

Answer 1

  is a HTML entity, but doesn't exist in XML.

Either get rid of it (you're not saying where it comes from, so it's hard to give any more specific advice), or wrap your HTML data in CDATA blocks so the parser ignores them.

Answer 2

HTML specific entities - in this case   - are not valid xml entities, and that is what simplexml complains about; it reads the file as xml (not html) and finds entities which are not valid. You need to convert HTML entities back to their character representation first (you can use html_entity_decode() to do that)

$str = "some string containing html";
// this line will convert back html entities to regular characters
$str = html_entity_decode($str, ...);
// now convert special character to their xml entities
$str = str_ireplace(array('<','>','&','\'','"'),array('&lt;','&gt;','&amp;','&apos;','&quot;'),$str);

save_to_xml($str);

Note that if you use htmlentities() on your string before saving it in the xml, then that is the source of your problem (as you are converting html character to their respective html entities, which are not recognized by simplexml as xml entities).

// this won't work, the html entities it will uses are not valid xml entities
$str = htmlentities($str, ...)

save_to_xml($str);

If you have troubles understanding it, think of it as two different languages, like spanish (html) and english (xml), a valid word in spanish ( ) doesn't mean it is also valid in english, no matter the similarities between the two languages.

Answer 3

&nbsp ; is no-breaking space. You have to replace it. http://en.wikipedia.org/wiki/Non-breaking_space