Daniel Veillard wrote:
On Fri, Mar 17, 2006 at 03:46:46PM -0800, Rush Manbert wrote:I am trying to write a function that removes all attributes from an element using the tree interface. It seems that xml:lang, xml:id, and xml:space are special cases, since they have their own get and set functions.Do I need to do anything special to remove these attributes from the node?No, they are normal attributes, only their namespace is special.My current code for "remove all attributes" is this: inline void removeAllNodeAttributes (xmlNodePtr pNode) { xmlAttrPtr pAttr; while (NULL != (pAttr = pNode->properties)) { xmlRemoveProp (pAttr); } }that should work. That won't remove namespace declarations, but that's normal they are a different kind of nodes, and removing them is way harder.
Thanks Daniel, maybe this thread will work. My apologies, I wrote xml:id when I means xml:base in the original post.I can see that xml:base is applied to an element in the normal way. The W3C Recommendation has this example:
<doc xml:base="http://example.org/today/";
xmlns:xlink="http://www.w3.org/1999/xlink";>
It makes sense that remove all attributes should remove that one. Or at
least have the option of removing it.
However, xml:space only seems to appear in the DTD, within an <!ATTLIST>. Does that just end up adding a normal attribute named xml:space on every affected element, but it does it "behind the scenes"?
If I wanted to change my code to preserve xml:space and xml:base, would I just do a string compare of "xml" to xmlAttribute.prefix and, either "base" or "space" to xmlAttribute.name?
Thanks, Rush