implement GetText

[Anonymous]

There are so many people who want to get text in a convenient way, maybe this may be useful in the next version of tinyxml?

char *TiXmlElement::GetText()
{
    return (char *)(TiXmlText *)(FirstChild()->ToText())->Value();
}

[Lee Thomason]

I'm totally up for considering this. But I have this concerns:

How to handle:
<foo> This is <bold>bold</bold> text.</foo>

If you GetText() on "foo", what do you get back? "This is"? "This is text"? "This is <bold>bold</bold> text."? All have pretty serious drawbacks and can be very misleading.

lee

[Ellers]

Good point.
But I think it is a common case to have a node like:

  <message>Hello</message>

It seems a very useful/reasonable call to do to get the text of the first child.

A "strawman" position (ie something to start with; knock it down if you like):

  someNode->GetText() returns the text of the first child of someNode IF someNode is an element whose first child is a text node.

Error handling is a bit of a pain.

e.g.
  text = pElement->GetText( );
  if ( text == 0 )
  {
    TiXmlGetLastError()
  }

or
  bool result = pElement->GetText(&text)
or
  use exceptions.

I like the first option.

WDYT?
Ellers

[Anonymous]

yope, but it would be better to move  if ( text == 0 ) inside GetText - I did it so.

[Lee Thomason]

The more I thought about it, the more I liked the "returns the text of the first child of someNode IF someNode is an element whose first child is a text node" approach.

Hopefully, adding GetText won't confuse the ToText() and Text() problem. Ughh.

I haven't a decent little batch of changes - I'll try to get that in soon.

lee

[Lee Thomason]

Implemented & checked in. Thanks for the input!

lee