W3C home > Mailing lists > Public > xmlschema-dev@w3.org > May 2003

Re: whitespace reality check: when is whitespace significant in element-only content

From: Jeni Tennison <jeni@jenitennison.com>
Date: Tue, 27 May 2003 17:12:37 +0100
Message-ID: <104431152584.20030527171237@jenitennison.com>
To: "Paul Kiel" <paul@hr-xml.org>
CC: xmlschema-dev@w3.org

Hi Paul,

> The element "Parent" is defined as an element-only content
> consisting of the child element "Child". The "Child" element is
> optional.
> Example instance 1:
> <Parent>
>      <Child/>
> </Parent>
> Example instance 2:
> <Parent>
>  </Parent>
> I know that Example instance # 1 is valid - and all parsers agree.
> The whitespace in between the end of the openning <Parent> and the
> beginning of the <Child> element (a line break and a few spaces) is
> ignorable.
> On Example instance # 2, there is one parser that says the white
> space is significant, and therefore not valid. Most of the parsers
> say it is valid. Is the whitespace in Example 2 significant? Does it
> prevent validation? (even though it is the same whitespace that is
> in Example 1).

Example instance # 2 is valid. In this case you have elementOnly
content; the schema spec says:

 "If the {content type} is element-only [sic], then the element
  information item has no character information item [children] other
  than those whose [character code] is defined as a white space in [XML
  1.0 (Second Edition)]."

It is only when the content type is actually empty (i.e. you're not
allowed any content at all in the element) that whitespace is
significant. For example, if <Parent> were declared with:

<xs:element name="Parent">
  <xs:complexType />

then it wouldn't be legal for the <Parent> element to contain

So you've discovered a bug in one of the validators and should report



Jeni Tennison
Received on Tuesday, 27 May 2003 12:12:47 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:15:11 UTC