On Wed, 02 Aug 2006 19:18:16 +0200, Subbu Allamaraju
<[EMAIL PROTECTED]> wrote:
True. Since the XHR spec is not really concerned about if and how
applications write/code to empty headers, it does not make sense to
require that this method return empty string. I think the semantics of a
single
getResponseHeader() could include the cases of both non-existent headers
as well as header with an empty value. But the current definition
prevents
this, prompting for methods isResponseHeaderPresent(), which IMO, would
not be required if the semantics are changed to return null.
They are changed.
--
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>